How to integrate the European Energy Markets: a new vision

How to integrate the European Energy Markets:
a new vision
Abel M. Mateus1
Portuguese Competition Authority
The European Commission has today adopted the Final Report of its competition
inquiry into the gas and electricity markets… Its disappointing conclusion is that more
than a decade after having launched the drive for liberalisation, we are still far from
having a single, competitive and well-functioning European energy market.
Nelly Kroes, Press Conference, Brussels, 10th January, 2007.
Prepared for
Fordham University Conference
New York, September 2007
Version: August 27th, 2007
I would like to acknowledge the assistance of João Lopes. We also thank information and comments
from numerous persons in different agencies.
European Energy Markets
Abel M. Mateus
1- Introduction
Starting in the 1980s, reform of energy markets addressed four major areas: (i) privatization
of the parts of the system that could be subject to competition; (ii) introduction of
wholesale and retail market competition, allowing for new entry into generation and retail
supply; (iii) restructuring through vertical unbundling of generation, transmission,
distribution and retail supply activities, as well as horizontal splitting of generation and
retail supply; and (iv) regulation of the monopoly segments by establishing an independent
regulator, provision of third party network access and incentive regulation of transmission
and distribution networks. After almost two decades of reform, a number of EU
governments and the Commission as well as competition authorities are dissatisfied with
the results achieved so far. The paper deals with some of the major problems that need to
be addressed, and assesses the policies required to solve these problems. We cannot cover
all the issues, and questions like security of supply, emissions policies and supply mix are
not addressed here. We concentrate on four major issues: (i) ownership unbundling; (ii)
cross border trade and interconnection; (iii) merger policies; and (iv) competition policy in
a world of concentration of energy resources. We start by showing that the EU is still a
despairingly and segmented set of energy markets, and that the level of integration achieved
so far is quite low. Moreover, we also show two things. First, that the level of competition
in energy markets is still low in most of the countries, since they do not even pass the most
primary tests and it is also quite dispersed. And second, that competition matters for energy
prices. An important study for the Commission has shown that the latest price increases
have sometimes been exploited for increases in margins. We also point out an issue that
would merit a separate paper, e.g. that energy policy is sometimes tainted by serious flaws
in planning systems. There is a need to develop and apply more widely computable general
equilibrium models that replicate the market, and project evaluation would benefit from
more use of real options theory. Governments and the Commission establish targets and
European Energy Markets
Abel M. Mateus
supply mixes, defining targets for renewable energy, without an in-depth study of its impact
in consumer welfare and the competitiveness of the economy, as well as without a study of
the policy instruments required for supporting the decisions of economic agents.
Another tenet of the paper is that ownership unbundling with careful regulation and
enforcement of third party access is necessary for reform. We demonstrate the benefits for
investment and competition with empirical evidence for the UK and Portugal. We also
show how to solve coordination problems in practice. However, and contrary to a number
of positions, we also argue that unbundling is not sufficient. It is essential to create a
competitive market in generation, which may require divestiture of assets, VPP caps on
market power and other measures. Evidence in support of this position is provided from
the UK, Portugal, the Nordic Market and the USA.
The second pillar for creating an integrated market is to increase the level of
interconnection. First, we show that this is no substitute for creating national competitive
markets, but it is important for market integration. Second, our position is that there
should be better regulation of interconnectors, and that the amount of required capacity
needs to be addressed by economic modelling and not by a single rule that “fits all”. There
are quite a number of strategic issues involved. An efficient incentives network based
system could also provide a market solution for the problem. Third, a minimum
harmonization across national regulations has to be carried out and the issue of an EUwide regulator should be addressed.
Another important area of concern is the present merger policy at national and Community
level. First, we start by dismissing the theory of national champions in the area of energy
that seems to be gaining popularity in recent times. But we also dismiss the theory of
European-wide champions. We also show that the present rule of 2/3 needs to be
qualified, and that there is a need for harmonization of merger policy across the EU, largely
as a result of the externalities across markets created by energy flows. Furthermore, there is
European Energy Markets
Abel M. Mateus
a problem of level-playing field in cross border mergers: state enterprises or state
controlled undertakings can buy private enterprises in other countries, but these private
enterprises cannot buy the former. Besides, there is a problem of governance and
contestability in those cases. Finally, we also show that there is a need to improve merger
analysis for national and cross-border cases due to the specific nature of electricity markets,
to the definition of dominance, and to the shifting nature of those markets with
interconnection and technological change.
Finally, we show the importance of having a common external policy on energy trade by
the European Union with our major trade partners, and especially in regard to energy. We
defend that all bilateral deals between Member States and major producing countries, such
as Russia or Algeria, should not be conducted in the absence of that common framework.
The paper starts by identifying some major issues on integration of European Union
energy markets. Section 3 discusses the problems and provides evidences on unbundling.
Section 4 discusses the problems of cross-border trade. Section 5 discusses merger policies
at national and community levels. Section 6 addresses competition policy in an
environment of global concentration in natural resources, and section 7 concludes. There
are also two Annexes on the UK and Portuguese experience2 in market liberalization and
2- Issues in Integration of the European Union Energy Markets
The EU electricity and gas markets are a despairingly association of almost closed national
markets: substantial price differences persist among Member States and the problem has
been aggravated with large increases in energy prices since 2003 (Figure 1). In the last five
years there was hardly any progress in achieving a convergence process, demonstrating the
Annexes will be supplied upon request.
European Energy Markets
Abel M. Mateus
lack of integration of national markets. The European Commission Sector Inquiry3, carried
out in 2005-06 identifies three major structural problems. Firstly, many energy markets are
too highly concentrated, with dominant positions of incumbents prevailing in many
markets. This has led incumbents to exploit their market power and raise prices, and to
engage in various practices that make it harder for new entrants to compete.
Secondly, many energy markets are characterised by a high degree of vertical integration.
Bundling of network and supply activities allows incumbents to control the network and
the supply market. They use them as strategic assets to exclude competition through
discrimination. Moreover, in this case, there are too few incentives to invest in networks - a
major obstacle to new entry and a threat to security of supply. Many of Europe's electricity
interconnectors are chronically congested. The total amount of revenues collected by these
Transmission System Operators (TSOs) between 2001 and 2005 was about 1.3 billion
euros. However, only 250 million euros, e.g. less than 20% of these revenues, was invested
back into increased capacity. There are insufficient tradable supplies on energy markets, e.g.
they are illiquid. And long-term contracts contribute to locking-in the markets.
Thirdly, there is an absence of cross-border integration and of cross-border competition.
Incumbents largely keep to their traditional markets, and rarely enter other national
markets as large scale competitors. The report reveals that the energy prices for commercial
users vary significantly from Member State to Member State. These differences are not
eroded through import competition. One of the reasons is that incumbents stick to their
home markets. Different market designs between Member States contribute to the gloomy
picture rendering it difficult to move energy from one point in Europe to another.
Figure 1
European Commission, DG Competition report on Energy Sector Inquiry, January 2007
European Energy Markets
Abel M. Mateus
Source: European Commission, DG Competition Report on Energy Sector Inquiry (2007).
More recently, the London Economics (LE) price study4 has clearly revealed that the more
concentrated wholesale markets are the higher the price-cost mark-up. More importantly,
the LE study has shown the usefulness of a set of measures that characterize market power
in electricity markets. Since electricity cannot be stored and generation must be balanced
with demand at all time, the indispensability of power generation to meet demand gives a
precise notion of how, in short-term markets, market power can be exercised. The more
indispensable a specific generator is the higher its market power is. Thus, its market power
depends on market conditions and can change dramatically across one trading day.
Several institutions use a set of market indicators to characterize the level of competition,
e.g. Oxera produces annually for the DTI a benchmark report about the degree of
competition in the UK market, compared with other EU countries. Only countries that
have third-party access to transmission regulated, and a certain degree of unbundling and
supply-market opening, are considered. Then, batteries of indicators, aggregated in five sets
London Economics and Global Energy Decisions report to DG Competition, European Commission,
Structure and Performance of Six European Wholesale Electricity Markets in 2003, 2004 and 2005, February
European Energy Markets
Abel M. Mateus
as indicated in Table 1, are considered for the electricity and gas markets. The results show
three groups of countries in electricity: low level of competition (Ireland and Portugal),
medium level (Spain, Italy, Austria and Germany) and high level of competition (Nordic
countries), the UK having the highest score. The gas markets have systematically a lower
level of competition, from Portugal that benefits from derogation to the UK, the only
country scoring above 6.
Table 1
Downstream supply
Score-all market areas
Score network area
Overall elctricity score
Upstream market
Disaggregated scores for select EU electricity markets (preliminary 2005)
Downstream supply
Score-all market areas
Score network area
Overall elctricity score
Upstream market
Disaggregated scores or select EU gas markets (preliminary 2005)
Source: Oxera. Energy market competition in the EU and G7: preliminary 2005 rankings. October 2006.
Part of the price differences between the countries analysed in the LE study reflect the
supply mix that may depend on the geographic characteristics of the country and most of
the times on Government discretionary decisions (most follow a licensing scheme). The
LE study has also identified market power as another major factor in price formation. In
fact, entry in generation and retail supply has only to a limited extent eroded the
incumbent’s market power inherited from the past monopolies framework.
The study of LE was carried out for six MS: UK, France, Germany, Belgium, Spain and
Netherlands. It used data from hourly observations and studied simulations and related
European Energy Markets
Abel M. Mateus
price to outcomes of output and marginal cost, and market structure for almost every
generation unit in every hour for 2002-4.5 Besides estimating traditional measures of
specific electricity market structure related to concentration, it calculated the PSI (Pivotal
Supplier Index) that shows how “pivotal” is a firm to supply electricity at a given hour, and
the RSI (Residual Supply Index) that shows how the market depends, on a continuous
basis, on this company capacity to meet demand. In fact, in a tight market situation, even a
small company can exert significant market power, and in an excess supply situation even a
large company has to engage in a competitive game. The estimates show that France and
Belgium are highly concentrated – only one company dominates the market most of the
time; Germany, Spain and Netherlands have average levels – with two companies with
significant market position 20 to 77 percent of the time; and the UK is moderately
Next, it calculated measures of market outcomes, like the Lerner Index and the Price-Cost
Mark-up, estimating marginal costs based on the GED simulation model, used by the
industry to compute the optimal market despatch. The highest margins are for France and
the lowest for GB and Netherlands.
The final step was to regress market outcomes to market structures. Figure 2 shows the
plot of the data. The conclusion is that there is evidence of a high negative correlation
between measures of concentration and margins. Moreover, it concludes that most of the
price increases that occurred lately were not due to fuel costs but to exploitation of market
power and environmental costs.
Figure 2
Equivalent to 8 760 hours per year.
European Energy Markets
Abel M. Mateus
Source: London Economics, op. cit.
Europe faces major challenges: (i) EU emissions may increase by 5 percent by 2030, rather
than falling; (ii) EU energy import dependence will jump from 50 percent of today’s total
energy consumption to 65 percent in 2030; and (iii) energy markets remain far from
integrated markets.
According to several policy statements by the European Commission, the “energy policy
for a changing world” has a broad objective: “well functioning energy markets that ensure
secure energy supplies at competitive prices as key to achieving growth and consumer
welfare in the European Union”. There are three pillars: (1) building a true Internal Energy
Market, which gives real choices for EU energy users, whether citizens or businesses, and
that triggers the huge investments needed, in order to promote competitiveness,
sustainability and security. To reach this aim it is necessary to achieve effective unbundling,
to have an independent regulatory control, and to reach a certain minimum capacity in
interconnection;6 (2) accelerating the shift to a low carbon energy, by cutting greenhouse
gas emissions by 20 percent by 2020 through an expansion of renewables or expansion of
The Commission has set a 10 percent of total demand target.
European Energy Markets
Abel M. Mateus
nuclear power,7 and increasing energy research; and (3) increasing energy efficiency, by
saving 20 percent of total primary energy consumption by 2020. These objectives would be
complemented by an international energy policy: speaking with one voice, cooperating with
the rest of the world on reducing emissions8 and developing effective solidarity
mechanisms for solving any energy crisis.
Despite the importance of these objectives of energy policy, we did not find a single
estimate of the costs for the EU of reaching those targets.
This paper concentrates on the first pillar, with a focus on competition issues. In fact,
building an integrated energy market in the European Union requires: (i) confidence in the
security of high quality supply involving adequate production capacity, security of primary
fuel supply, and reliable and adequate networks; (ii) sustainable competitive outcomes,
referring to ability of the sector to finance and deliver efficient and reliable energy, within a
market mechanism that provides market signals for entry, that is efficient in location,
timing, scale and fuel choice, with environmentally friendly processes, and at minimum
costs to consumers; (iii) absence of market abuse and ability to mitigate market power; and
(iv) credible regulation for efficient free entry and investment.
According to Newbery9 and others, the preconditions for liberalising energy markets are: (i)
guarantee a non-discriminatory and free third party access to entrants; (ii) total unbundling;
(iii) adequate and secure supply involving an adequate and reliable network, enough
production capacity and secure sourcing of primary fuels; (iv) sufficient competition; and
(v) independent and appropriate regulation of monopoly segments.
At present nuclear represents 30 percent of EU electricity.
By strengthening the Energy Charter Treaty, review the post-Kyoto climate regime and extension of
emissions trading to global partners.
Liberalizing Electricity Markets, Cambridge University, 2002.
European Energy Markets
Abel M. Mateus
Market integration means for most economists that European costs are minimized, trade
takes place guided by comparative advantage, importing competition into more
concentrated markets. To build an integrated EU market10 there are two important preconditions. The first is to build enough capacity in interconnectors in order to create a
system of “connecting vases” such that flows of energy equilibrate the market and equalize
the prices across the EU. Competition pressure arising from imports is then sufficient to
discipline domestic incumbent’s behaviour. The second is to effectively reduce market
concentration, either trough divestment or trough other remedies that reduce the
possibility of incumbents to behave independently. However, less concentrated markets –
and unbundling – also diminish the incentives to explore congestion bottlenecks by
domestic players and therefore reduce the needed interconnection capacity in order to
reduce the market power of incumbents.
In the 2002 Barcelona European Council, EU Member-States agreed to increase minimum
interconnection levels to 10% of domestic electricity production capacity. However,
interconnection investment is lagging. The congestion rents are only to a limited extent
used to invest in interconnection capacity expansion. Actually, as noted by the European
Commission, only about 5% of total annual investment for electricity grids in the EU,
Norway, Switzerland and Turkey was directed to increase interconnection capacity.11
Today a significant number of Member States have still not achieved the Barcelona target.
There is a difference in building a Single Market that presupposes similar prices throughout the EU, or an
Integrated Market that refers to building a competitive market throughout the EU with enough
interconnection. In the last case price differentials may persist because of different costs of production or
transport. The paper concentrates on this last aim, which we think makes more economic sense.
European Commission, Communication from the Commission to the Council and the European
Parliament, Priority Interconnection Plan, (2007)
European Energy Markets
Abel M. Mateus
The recent blackouts that occurred in Europe, like the Italian blackout of September 2003
and the German initiated blackout12 of November 4, 2006, also showed the need for a
more intensified cooperation between TSOs and a minimum set of rules for managing the
Therefore, and following the Sector Inquiry conclusions, the European Commission, both
in merger control and in anti-trust intervention, foresees the application of remedies in
order to tackle the high concentration levels seen at the domestic level in most countries.
These remedies range from large scale Virtual Power Plants (VPPs)13 to structural measures
as divestitures.1415
Europe is not the only place were reform is taking place. There have been several energy
reform policies throughout the world. There is a consensus that some of the most
successful involve countries/regions of UK, Nordic countries, New Zealand, Victoria and
South Australia, New York, Texas, PJM, Chile, and Argentina.16 In almost all cases
ownership unbundling has played an important role.
Let us state at the outset that there is still a need to inject better economic methodologies
in project evaluation and in system planning. The field is still largely dominated by
engineers and physicists. All the scenarios that have been published by the European
Union, available in the site of the Commission, do not report the costs of those policies for
the different economic agents (households, enterprises, electricity generators, etc.) and the
instruments necessary to induce their behaviour in a market economy.17
Started within the E.On network and was caused by a boat reaping a power cable.
In practice VPPs are not very effective because the power of the incumbent to accumulate reserve power
and its need in the balancing system.
Another measure proposed by the Spanish Libro Blanco is to limit the market share of the large producers
by time of the day.
The Spanish White Paper proposed the following measures to limit market power of Endesa and Iberdrola:
plant divestiture, VPPs (following the Fench and Dutch experiences), and long-term contracts where both
price and quantity are set by the regulator. The report is published as Libro Blanco, dated 30 June 2005 at
For a description of each experience see Sioshansi and Pfaffenberger. 2006.
This problem does not seem to affect the DoE model.
European Energy Markets
Abel M. Mateus
In some EU countries, governments intervene and substitute to the market in choosing the
supply mix. And most of times they do not even have enough studies for justifying their
decision, introducing heavy distortions and raising consumer prices.
Naturally, governments should intervene in correcting externalities, like making the polluter
pay the social cost. But they should intervene by distorting as less as possible the market
mechanism. E.g. there is still no basic approach to the incentives for renewables in the EU.
Several governments give a guarantee to the tariff18 and priority of access for wind power.
However, in the US, the government gives a tax credit that is much less distorting.19
1- Vertical unbundling: solution or false start?
What is unbundling?
Electricity supply consists of four distinct activities: generation, high-tension transmission,
lower voltage local distribution, and supply – contracting for and billing final consumers. In
most countries generation accounts, on average,20 for about 65-70 percent of industry
costs; transmission for 10-12 percent; distribution for 20 percent; and supply for the
remaining 5-7 percent. Transmission and distribution (the “wires businesses”) are natural
monopolies that require regulation. The first requires a national grid to transport energy
from the production points to feed the local markets, but also requires balancing the
network, a characteristic of electricity systems. Regional distribution networks are fed by
the national grid and transport the energy to consumer homes or business sites. Generation
and supply are potentially competitive. What is a liberalized and competitive electricity
supply industry? It is a market(s) into which generators sell and from which final
consumers and traders can buy, a transmission and distribution system that balances
continuously the system and acts as a common carrier, with freedom of entry in generation
In Portugal the government guarantees an 8 percent rate of return on investment.
And also a loan guarantee in some cases.
Although for households the part of distribution is higher than for an industrial large consumer.
European Energy Markets
Abel M. Mateus
(subject to balancing and congestion in transmission) and freedom for final consumers to
choose from whom to contract for supply.
In countries that have gone through a process of liberalisation, it is recognized that
incumbents have usually a large set of contracts with the national grid (e.g. rights of priority
in transmission) that undermine competition and makes it difficult third party access. Thus,
there is a need for a Network Code in order to level the playing field and establish good
In the case of gas markets, unbundling means separating the high-pressure network from
the service of importing or taking gas from producer sources to supply customers.
Competition in source intakes means that there should be enough players importing gas
through an international pipeline or LNG plants. Storage facilities are also important parts
of the network, for balancing in the short- and medium-term the market. Since they can be
used as “essential facilities”, like points of intake of the gas for the country, its access
should be regulated and there should be enough competitive capacity to prevent market
foreclosure. Finally, market liberalization requires that there are enough suppliers to buy
gas from “producers” and sell it to customers. The high pressure network constitutes a
national monopoly and local distribution systems are regional monopolies. If there are few
“points of entry” for gas and storage they should also be regulated. Competition occurs
then between “producers” or in-takers and suppliers.
After analyzing the present EU situation, we will look at the costs and benefits of
unbundling and will support those with theoretical and empirical evidence. In the fourth
section, we will survey briefly the status of the current policy debate. In the fifth section,
we will discuss an alternative that is being proposed for the TSO, e.g. the Independent
System Operator (ISO) model current in the USA.21 In the sixth section we will address the
As we will see below, the ISO being discussed in Europe is not the same as the one prevalent in the US. In
the US is more like an extension arm of the regulator.
European Energy Markets
Abel M. Mateus
issue of public versus private ownership in the context of unbundling, with special
reference to the Nordic situation, whereas in the final section we will discuss also briefly
some legal issues.
There has been a discussion if unbundling should also be applied to the regional
distribution networks. In fact, the present EU directives also require legal and functional
unbundling of the distribution network. Most of the present thinking in the EU considers
that it is not justified to apply ownership unbundling to regional distribution networks,
following the experience of the UK.22
The present EU situation
The present second generation EU directives of electricity and gas23 impose minimum
obligations on energy network operators with regard to legal and functional unbundling
between transmission/distribution networks on the one hand and upstream (generation or
production)/downstream (supply) functions on the other. The companies are obliged to: (i)
create separate legal entities for network activities; (ii) separate executive management and
decision-making; (iii) maintain separate accounts; and (iv) create information barriers
between those activities (“chinese walls”). In addition, a regulated Third Party Access
regime was introduced, and the creation of national regulators became mandatory.
Network operators have to submit annual reports on the implementation of the
compliance programmes to the regulatory authority. Finally, Regulations were introduced
This is the opinion of the Council of Regulators, expressed in several documents of the ERGEG.
Directive 2003/54/EC of the European Parliament and of the Council of 26 June 2003 concerning
common rules for the internal market in electricity and repealing Directive 96/92/EC. (OJ L 176, p. 37-56),
hereafter Electricity Directive. Directive 2003/55/EC of the European Parliament and of the Council of 26
June 2003 concerning common rules for the internal market in gas and repealing Directive 98/30/EC. (OJ L
176, p. 57-78), hereafter Gas Directive.
European Energy Markets
Abel M. Mateus
which allowed for the adoption of legally binding guidelines with the aim of facilitating
cross border competition, and in particular interconnector activity.24
Among all the Member States less then ten have ownership separation of transmission25
from the incumbent for electricity and five for gas. In most of the cases, only a few years
have passed thereof.26 Table 7 shows the status of unbundling vis-à-vis the present
Source: European Commission27
Regulation (EC) No 1228/2003 of 26 June 2003 on conditions for access to the network for cross-border
exchanges in electricity (OJ L 176, 15.7.2003, p. 1-10) and Regulation (EC) No 1775/2005 of 28 September
2005 on conditions for access to the natural gas transmission network (OJ L 289, 3.11.2005, p. 1-13)
UK (England and Wales), Netherlands, Sweden, Finland, Denmark, Spain, Portugal, Slovenia and Czech
Republic for electricity; and Great Britain, Spain, Netherlands, Denmark and Romania for gas.
The situation of unbundling in the EU has been the object of a report for the Commission produced by
Gómez-Acebo Pombo Abogados and Charles Russel, Unbundling of Electricity and Gas Transmission and
Distribution System Operators, Final Report, December 1st, 2005 and updated in the Benchmarking reports.
European Commission, Commission Staff Working Document, Report on Progress in Creating the
Internal Gas and Electricity Market, 2005
European Energy Markets
Abel M. Mateus
Notwithstanding, the Energy Sector Inquiry demonstrated that legal and functional
unbundling is not sufficient28. The same opinion has been expressed by the European
Council of Regulators: “The ineffectiveness of the existing unbundling arrangements is a
significant reason for the slow pace of market integration and the slow growth in cross
border trade observed in EU electricity and gas markets. The “legal unbundling” regime …
hasn’t worked. The rules were too vague and the implementation by Member States so far
is patchy.” (ERGEG, Fact Sheet on Unbundling, May 2007).
As noted by the European Commission, vertically integrated incumbents have incentives to
discriminate new entrants. Investment incentives are also distorted. No incentives exist to
develop the network in the overall interest of the market, namely for facilitating new entry
or to rise import levels. It was also showed that there is considerable evidence that
investment decisions of vertically integrated companies are biased to the needs of supply
Also in the US, where unbundling had a legal basis, several court decisions against violation
of anti-trust laws demonstrated the detrimental effects for competition of vertical
integration. In the Otter Tail Power vs US, in 1973, the US Supreme Court ruled against
the behaviour of a vertically integrated utility, which refused, through its transmission
affiliate, to source cheaper electricity supplies. This case lead to a significant debate over
the competition benefits of vertical de-integration of electricity utilities. More recently,
FERC (Federal) Order 2000 was designed to resolve what it “perceived to be problems
created by the balkanized control of transmission networks and alleged discriminatory
See the summary conclusions in Lowe, Pucinskaite, Webster and Lindberg. Effective unbundling of energy
transmission networks: lessons from the Energy Sector Inquiry. Competition Policy Newsletter. Sping 2007,
p. 23-34.
European Commission, Communication from the Commission to the Council and The European
Parliament, Prospects for the internal gas and electricity market, January 2007
European Energy Markets
Abel M. Mateus
practices of generators and energy traders seeking to use the transmission networks of
vertically integrated firms”.30
Benefits and Costs of unbundling: theoretical and empirical evidence
What criteria should we use to judge the importance of unbundling? The criteria used to
sort out models should: (i) prevent discrimination in access; (ii) optimize use of
infrastructure; (iii) incentivize economic investment; and (iv) enable effective regulatory
oversight of monopolistic activities.
More importantly, the main question to be analysed is to what extent full ownership
unbundling leads to the loss of efficiencies, namely in what concerns the coordination of
short run operation of the electricity system, long term planning, the quality of service, and
grid losses. Since ownership unbundling has been established in a few countries for some
time, it is important to analyse those national experiences.
In this subsection, and after enunciating the major problems, we will address four major
issues raised by a cost-benefit analysis: (i) impact on competition and potential for
discrimination; (ii) transaction costs arising from short-run coordination problems; (iii)
incentives and coordination of investments and cost of capital; and finally (iv) double
What are the problems?
The relative merits of these two approaches – bundling versus unbundling – depend on
several factors, including the following: (i) the degree of common expertise required to
operate in both activities; (ii) the size of the transaction costs, and in particular the difficulty
in writing contracts; (iii) effect on incentive properties; (iv) effect on risk allocation; (v)
Joskow, Paul L., Transmission policy in the United States, Utilities Policy 13 (2005) 95-115.
European Energy Markets
Abel M. Mateus
effect on control over performance; and (vi) issues related to uncertainty.31 In fact, one of
the main benefits of vertical integration—i.e. the protection of a specific investment against
ex-post opportunistic behaviour and against market risks—can be replicated by long-term
contracts. However, vertical integration can offer benefits relative to the use of contracts in
several situations, as a result of transactions costs or information asymmetry. In some
industries, including the electricity and rail industries, vertical unbundling has been
undertaken, and coordination has effectively been reproduced through contracts rather
than through ownership, and, in general, evidence has shown that transaction and
coordination costs of an unbundled system are much lower than the deadweight loss of
having a vertically bundled system. First, generation costs are almost five times higher than
transmission and transportation costs in electricity and gas markets. Thus, a distortion in
the generation market is much higher than in the national grid system. Second, the
transmission system can be effectively used to foreclose market entry reverberating higher
producer costs throughout the system and up to the final consumer.
The remainder of this section will review the arguments for and against unbundling, and
contrasting the evidence from theoretical models and empirical studies. The Council of
Regulators and Pollitt do the same exercise. According to Pollitt there is a clear advantage
for unbundling in terms of (i) impact on competition; (ii) ease and effectiveness of
regulation; (iii) facilitation of privatisation; (iv) security of supply; and (v) specialization and
focus of management. A negative net effect may come from transaction costs. He also
considers that there is a reduced risk of arbitrary government intervention.
Impact on competition and potential for discrimination
There is an extensive literature on vertical integration. For an excellent recent survey see Joskow, Paul.
Vertical Integration, mimeo. Handbook of Industrial Organization, vol.3, North-Holland, forthcoming
European Energy Markets
Abel M. Mateus
According to a detailed study just published by the ERGEG32, possibility for discrimination
is high in both electricity and gas grids, in the following decisions of the TSO/ISO: (i)
access principles object of the grid code (access and financial rights, including connection
of producers and costumers to the grid and sub-grids); (ii) market principles, including
balancing market, intra-day market, provision of energy losses, day-ahead forecast, data
exchange, metering and billing; (iii) security principles, including N-1 principle, monitoring,
coordination, restoration, black start capability, load shedding, islanding, etc., as well as
forecast of demand for winter in gas; (iv) schedule management: technical and financial
organization to match physical delivery; (v) investment planning: long-term forecasts of
supply and demand, investment planning, benefit-cost analysis of each investment project;
dedicated transport network and extensions; (vi) procurement of balancing energy
(flexibility): market model to procure energy for ancillary services (auction, merit order);
(vii) transparency: necessary information to the market; (viii) capacity calculation
(optimization over the all grid); (ix) capacity allocation through a market mechanism
(coupling, auctioning, open season and contract); (x) congestion management; (xi) tariff
methodology using methods like fair return on investment, benchmarking, allocation of
costs to specific consumers, definition of deep and shallow connection charges, definition
of entry and exit points and allocation of costs to these points; and (xii) maintenance
Another important instrument for discrimination used by vertically integrated incumbents
is the use of their large supply mix in order to strategically manipulate entry into the
Third Legislative Package …
The Spanish case is an interesting one. Endesa and Iberdrola, the two major electricity producers are able
to jointly dominate the market because they have complementary supply mixes.
European Energy Markets
Abel M. Mateus
On the empirical front, Lowe et al. cite several cases of discrimination that support the
assertion that integrated firms not only have the potential for but they do discriminate
against third party access. For example, top management of the supply branch, which are
represented at the parent level, often have access to strategic business information of the
transport company, either directly or as a result of their representation in the Supervisory
or Administrative Board of the latter, and they often receive insights about activities of
competitors. Network operators may inform the vertically integrated supply branch when a
customer is considering switching supplier, and they do provide more detailed information
on loads, outages, generation to their affiliate so they can better optimize their trading and
production portfolio.
The case of Portugal (Annex II) demonstrates that unbundling with a single supplier or
with a high generation concentration does not allow a significant impact on consumer
prices, since generation costs are dominant, and competition does not take place on any
element of the chain. There might be a positive impact from cross-border trade, but the
incumbent usually has the capability to congest the interconnector. On the other hand, if
the case in point is a large country next to a small one, opening the border will allow the
large country incumbent to exert pressure on the prices of the smaller country.
The Commission gathered information during its Sector Inquiry that one TSO grants its
affiliated supply company rebates for the transportation fees as compared to competitors.
In another instance the supply company of an integrated gas company could nominate
their capacity directly to the network’s dispatching centre while third parties with shortterm interruptible contracts still had to nominate their capacities in advance.
European Energy Markets
Abel M. Mateus
Econometric work has also confirmed the importance of unbundling for competition.
According to a study by Copenhagen Economics,34 using data for 1990-2003, electricity
prices across the EU were lower with higher levels of unbundling.
On the theoretical front, Joskow and Tirole analyse the relationship between transmission
rights ownership and market power and show that vertical integration increases the ability
of generators to exercise market power by withholding transmission capacity.
Transaction costs arising from short-run coordination problems
Once third party access is granted then there is little difference between the costs of
transaction and costs of information for regulation for an integrated or unbundled system.
The main issue is to require opening the network to competitors on an equal basis and
information sharing.
However, one has to consider the potential benefits that can arise from what is the
opposite case of perfect vertical integration –e.g. the case of competition in generation and
retail supply. Open access to the transmission network is strategic for a workable
competition framework. Once the electricity industry is open to competition the
coordination problem arises. Perfect coordination can only be upheld if no competition
Opening access to transmission networks, even in the context of vertical integration
requires regulatory solutions in order to disentangle the coordination problems that arise
with free entry. The benefits from competition at the generation and retail supply levels
may pose negative externalities to the transmission activities, which need to be internalized.
In fact, these coordination problems have been one of the main arguments against the full
ownership unbundling.
Market Opening in Network Industries, Report to DG Internal Market. September 2005.
European Energy Markets
Abel M. Mateus
The previous section described the different codes that are part of the Network Code, and
give detailed information about the different aspects that requires agreement between
contracting parties and supervision by the regulator.
Thus, most of the discussion of Pollitt35 and others seem displaced, and the conclusion that
coordination costs are higher than the benefits that can be gained by competition are
simply wrong. The question and the answer given do not address the right issue.
A preferable approach is always to find a workable system of market indicators that are not
prone for distortion. To what extent the coordination that existed in the monopoly
framework could be replicated by correct economic signals has been one of the key
problems for regulatory authorities. Designing these signals is difficult and, until now, no
common approach has emerged. Economic signals can be part of the wholesale market
design (e.g. Location Marginal Pricing) or be part of the network charges (discriminating
Generation – G – and Load – L – charges according to the externalities created by
generation and load in each grid location). Locational marginal prices induce investment in
higher priced locations. On the contrary, locational G charges, induce location where the G
charge is lower. Since there may be an inverse correlation between locational G charges and
Locational Marginal Prices, these two models can achieve in theory the same results in
what concerns location signals for new generation. These economic signals can be used for
short term operation of the transmission system and also to guide investment decisions,
considering both the expansion of transmission capacity and the location of new
generation plants and electricity-intensive consuming industries.
The empirical evidence for countries that are even starting the road of functional
unbundling shows that quite an amount of information is required to facilitate third party
access. In the UK writing the Network Code took several years and hundreds of pages.
Incentives and coordination of investments; cost of capital
See Pollitt, op. cit..
European Energy Markets
Abel M. Mateus
In the electricity industry, vertical integration, as in the former monopoly framework, had
the merit of facilitate the coordination between investments in generation and transmission
which are highly specific and highly interdependent36. As a matter of fact, in some cases
(e.g. removal of grid bottlenecks), investments in generation or in transmission capacity can
be regarded as substitutes. Therefore, within the vertical integrated firm, coordination
meant the ability to choose the least cost solution. The optimality of investments under
perfect coordination could also achieve a better planning of grid networks, minimizing the
need for new lines – often prone to resistance from environmental organizations and local
However, the EU Sector Inquiry and other empirical evidence show that making
transmission asset ownership separate from generation ownership improves incentives for
market expansion and deepening. The willingness of integrated generation and
transmission asset owning firms to propose socially beneficial investments in the
transmission networks is quite low, since they may benefit competition and invite entry or
sustain operation of competitors. This problem is also felt by ISOs that face the approval
of the supply owner in order to carry out those investments. This was the experience of
Chile and the position taken by the incumbent.
The vertically integrated monopolist, provided that there is sufficient generation capacity to
meet demand, has no incentives to replace obsolete power plants. The returns on
generation investments were guaranteed by long term Power Purchase Agreements for the
life time of the plants. The costs of poor generation investment choices – the investments
done in the past and the investments that were not made given the more efficient
technologies – were passed through the consumers. The main objective of central planning
was to guarantee the adequacy of generation capacity.
Robert Michaels, Vertical Integration and the Restructuring of US Electricity Industry, Policy Analysis,
CATO Institute, July 2006
European Energy Markets
Abel M. Mateus
With competition, the incentives of generators are modified. To compete, it is necessary to
search for the more efficient technologies – both in terms of expected demand and total
costs (short and long term, including environmental costs). The entry of new generation
swiftly puts aside the obsolete power plants. The entry decision is not conditioned to zero
risk PPAs.
Free entry37
and competition have contributed for improved operating efficiency of
power plants, increased innovation, increased transfer of risks from consumers to
generators and the ability of consumers to choose their suppliers. Transaction costs may be
higher – since the problem of dispatching generation plants in the past could be internally
solved by the vertically integrated firm, by constructing the cost merit order of power
plants, and now is dealt with by the market through a price merit order – however they are
well below the efficiency gains from liberalization. New entry with more efficient plants
and more incentives to minimize the procurement fuel costs are contributing to reduce
overall costs of electricity production. However, how these benefits are shared between
producers and consumers depends to a great extent on the existing level of market power,
which still persists in many markets.
European countries exhibit quite different approaches to define the economic signals for
long term locational signals. Only a few countries have adopted Locational Marginal
Pricing or G charges with locational discrimination. Charging new generators with grid
investments also follows different approaches. In some countries, new generators are only
required to finance the connection to the grid. In others, on the contrary, they are required
also to finance grid upgrades in order to dispatch the new generation capacity.
37 GED (GlobalEnergyDecisions), Putting Competitive Power Markets to the Test the Benefits of
Competition in America’s Electric Grid: Cost Savings and Operating Efficiencies, July 2005.
Newbery, D., Pollitt, M., The restructuring and privatisation of Britain’s CEGB – Was it worth it?, The
Journal of Industrial Economics, September 1997
European Energy Markets
Abel M. Mateus
Requiring the investors to pay for the grid upgrades needed to dispatch the new generation
plants is another solution, providing incentives for investors to search for the grid’s least
cost location. If grid reinforcement costs are correlated with the Generator charge in every
location, then these two models will provide the same locational signs. However,
differently from the Generator charge model, in this alternative model, since costs of grid
upgrades are sunk costs, they do not interfere with short term operation of the power
plants, ignoring some potential externalities of generation in different locations in
transmission losses.
Regulation EC 1228/200339 favours the use of Generation and Load charges to provide
locational signs for new investments. However, under the current debate no consensus has
emerged on how to harmonise Europe’s transmission policy. One of the main reasons for
this lack of consensus relies on the impending uncertainty of how efficient and accurate G
charge locational signals can be. In fact, investment in generation involves other sources of
costs, either than the transmission charges, which must be considered.
For example, in several markets in the United States, even though LMP is implemented,
new generation facilities are also charged with the cost of network upgrades required to
dispatch generation. As Joskow40 points out, under this framework, it is likely that these
cost obligations play a more important role in generator location decisions than do
variations in Locational Marginal Prices.
On the other hand, the issue of investment in merchant transmission lines – lines that are
built based on the expected returns related with traded transmission rights, which value is
based on LMP’s differences between the points interconnected by the line– has until now
been lower than the initially expected.
COUNCIL of 26 June 2003 on conditions for access to the network for cross-border exchanges in electricity.
Joskow, Paul L. Transmission policy in the United States, Utilities Policy 13 (2005)
European Energy Markets
Abel M. Mateus
At this point in time, the search for an optimal transmission policy that allows for better
coordination between transmission and generation investments is still underway. However,
this cannot be used as an argument against ownership unbundling.
One of the merits of ownership unbundling is to disconnect the incentives of the
Transmission Owner from generation and supply activities. Under the ownership
independence framework, the incentives of the transmission owner depend on the adopted
economic regulatory model, with the rate of return model being the most commonly used.
Although it is arguable that this model may lead to overinvestment, one has to consider the
burden to consumers that this overinvestment leads to and, on the contrary, the potential
benefits that can be derived from overinvestment in transmission.
The share of transmission costs on end customer’s price is in the range of 5 to 15 percent.
On the other hand, the cost of energy production ranges between 60 and 80 percent. If
over-investment in transmission network means lower restrictions for competition in
generation and less grid bottlenecks, allowing effectively for less energy costs, on balance it
is a welfare improving move.
For instance, congestion costs represent a significant source of inefficiency in electricity
markets performance. Congestion means that cheaper generation cannot reach consumers,
requiring, in areas where the bottlenecks exist, to call more expensive power plants to
generate. These costs may be very high, and represent a source of inefficiency far more
important than the inefficiencies derived from overinvestment in transmission capacity.
Coordination of investments in the Portuguese case has been largely solved (See Annex II)
by entrusting the TSO with the function of producing medium and long-term plans of
supply and demand for the generation system, as well as another plan for the required
network infrastructure. Then, the regulator (ERSE) will have to supervise the network
expansion plans, like the UK regulator, whereas the government grants licenses for
production based on market needs.
European Energy Markets
Abel M. Mateus
Does unbundling lead to an increase in the cost of capital to the TSO and generation
companies? Theory and empirical evidence do not give an affirmative answer. Firms may
be smaller, but management will be more focussed and rip the benefits of a higher
specialization. If some of the firms can merge with negatively correlated risks the cost of
capital may even fall. However, as Pollitt (2007) states, it is difficult to believe that with
efficient capital markets the effect may be significant. Moreover, separate market interest
rates for the different businesses is important for financial efficiency.
The European Commission Sector Inquiry shows, on the empirical front, that it is often in
the interest of the integrated incumbent not to invest in infrastructure that would bring
additional competition. As Lowe et al. point out (p. 27) “the interest in protecting the
market power and profitability of their supply business trumps their interest in increasing
their (regulated) network business. In 2006 there was an interesting case of abuse of
dominant position brought by the Autorità against ENI, the Italian gas incumbent, because
it delayed investment in a pipeline owned by one of its subsidiaries, the Trans Tunisian
Pipeline Co. The increased capacity would have improved the ability for competitors to
import Algerian gas to Italy and compete with ENI on the Italian wholesale market.41 In a
number of cases, companies have invested only in capacity if their related supply arms had
confirmed their interest.
As ERGEG42 underlines, where generation and network business are integrated, potential
for discrimination exists, namely in what concern “slowness to connect competitors,
investment decisions (e.g. failing to invest to remove bottlenecks), and scheduling of
Bolletino of the Autorità Garante della Concorrenza e del Mercato no. 5/2006 (reference A358) Decision
of 15/02/2006.
ERGEG, Fact Sheet on Unbundling, May 2007
European Energy Markets
Abel M. Mateus
Econometric evidence on this subject is more controversial, especially contrasting data
from the USA. However, it is important to say at the outset that the US is a much more
competitive market with a lower level of concentration than Europe.43 The old literature
on vertical economies in the electricity industry finds that are scope economies in the joint
operation of generation and transmission.44 Michaels (2006) reviews 12 papers on vertical
integration in electric power and finds that 11 show benefits from vertical integration,
mainly derived from separability tests.
Finally, on the investment front, there may be other barriers to an efficient market, derived
from government licensing, subsidies or tax benefits, or land use permits.
Double marginalisation
This issue can be easily disposed off in theoretical terms. According to Pollitt45 the problem
does not arise because in network service pricing the method used is a multipart tariff and
marginal prices usually equal marginal costs.
Moreover, since transmission prices are regulated the problem cannot be seen as the trivial
question of integrated monopoly maximization versus two-monopolies maximization.
The European debate
At the European level a definitive proposal for the insufficient unbundling has not yet been
defined. France is opposed because the Government wants to maintain EDF46 and GDF as
the incumbents, with state control, fully integrated, and has shown a sign that wants to
This has resulted from antitrust positions taken during the 1930s.
Nemoto, J. and M. Goto. Technological externalities and economies of vertical integration in the electric
utility industry. International Journal of Industrial Organisation, 2004, vol. 22 (1) p. 67-81.
The arguments for and against ownership unbundling of energy transmission networks, August 2007,
CWPE 0737.
The ISO for France is RTE. It has only legal autonomy vis-à-vis EDF Transport, but the management is
chosen by EDF in consultation with the government.
European Energy Markets
Abel M. Mateus
protect them as European energy champions.47 The following citation of the regulator
shows that the situation is satisfactory with an integrated single monopoly and that
regulation can solve all the problems raised above:
Intervention of Phillippe de Ladoucette during the seminar “Politique Energetique: 2007, quells choix pour la
France?”, Maison de la Chimie, January 17th, 2007.
Germany wants to preserve its large energy firms, largely controlled by the Landers,4849 also
fully integrated50. Until recently there were 12 regional TSOs, but now there are only four,
after merging their networks.51 It has even resisted until two years ago in forming a sector
regulator, and then it has given it very light powers.
The problem of the level playing field in the ownership of firms in merger situations has
been addressed recently by the Commission52 within the financial sector. For example, any
Spanish Caixa, which are regional banks controlled by the Governos Autonomicos, can buy
a private bank anywhere in the EU while the reverse cannot occur because of their statutes.
Centrica recently filled a claim with the Commission against the Continental vertically integrated firms
because they were excluding from the market the non-integrated suppliers. This behaviour combined with the
intervention of governments to regulate consumer prices for political reasons has led to the bankruptcy of
quite a number of traders in Germany and Portugal, among other countries.
E.g. E.ON is normally considered to be controlled by the state of Baviera through its Cmmunal Funds.
In Spain Iberdrola is also controlled by the state of Cataluña, through its regional bank.
On March 7th, 2007, the Süddeutsche Zeitung reported that incumbent operators Eon, RwE, EnBW,
Vatenfall and VDN planned to propose to the Commission to bundle transmission lines for electricity and
gas within a region extending from Germany to France, Belgium, Netherlands and Luxemburg.
The same measure was taken by Switzerland after the great blackout of 2003, all TSOs have been merged
into one.
See also the presentation of the Portuguese Competition Authority president to the Parliament in 2006.
European Energy Markets
Abel M. Mateus
In the last count, there are about ten countries that are against ownership unbundling. The
arguments presented by all of them are just a perusal of the points raised above about the
costs of unbundling. The resistance to structural unbundle Transmission System
Operators53, led to the study of alternative unbundling provisions, namely to separate
system operators without ownership unbundling – the ISO/TSO model.
Choice between TSO and ISO
Theoretically, ownership separation has an important impact on competition. It removes
network bottlenecks and creates independence from the suppliers. The system can to a
certain extent be replicated by an Independent System Operator that, even if it does not
own the assets, dispatches and plans the investment in the network (ISO/TSO model).
There are different types of ISOs. The Deep ISO model removes all of the functions of the
system operator from the bundled company, leaving the bundled company only with the
ownership of the assets. The ISO undertakes live network operation, arrange for network
access, undertake network planning and make investment decisions, arrange for network
connections, undertake emergency planning and levy for use of the network, maintain
direct relations with the transmission customers and manage information flows to the
outside world. The Shallow ISO entails all transmission functions remaining with the
bundle company apart from the live operation of the transmission network during and
close to real time.
Some of the regulators in countries were the ISO/TSO model was implemented agree that
conflicts between the owners of the asset and the system operator are difficult to manage,
In March 6th, 2007, in an interview to Financial Times, German Chancellor Angela Merkel stated: "I do
not think the issue of competition can be reduced to the unbundling question. Stripping the energy producers
of their networks is no guarantee that there will be more competition. What you need is a strong regulatory
system to enhance competition and the right incentives to invest. In some countries, the separation of rail
services and networks did not exactly improve the network’s quality."
European Energy Markets
Abel M. Mateus
and require a complex set of contracts. The principal/agent economic theory, that offers a
model that can be referred to for similar circumstances, point out that the ISO (principal)
will face transaction costs to ensure that the TSO(s) (agent) makes all the efforts necessary
to realize his orders, since the TSO(s) interests, as part of the vertically integrated
undertaking, are inclined to diverge from them.
In fact, ERGEG argues that ownership unbundling has the advantage over the Deep ISO
model of having the right incentives for investment: “The ISO model…adds the risk of
conflicts as regards investing and sharing the profits resulting from transmission activity
between the owner and operator of the assets, without any benefits to the network user.”
(ERGEG op.cit.).
The ISO/TSO model is the main solution adopted in the US to tackle the vertical
integration issues under the open access to the transmission networks framework.54
Although it is politically simpler than the full unbundling, since it is less likely to face a
significant resistance from the utilities, it requires extensive regulation and it is not yet a
proven solution to create an efficient framework for the operation and planning of the
electric transmission systems. As Joskow55 observes, in regional markets where the
ISO/TSO governance model was implemented, like the PJM and New York regional
markets, while congestion costs increased, the investment in transmission has not fully
responded to alleviate this congestion costs. Moreover, the US markets are less
concentrated than the European.
Ownership unbundling not only is not affected by these transaction costs, but seems to
have a positive impact, at least in terms of timing, on investment. Most of the MS
It is important to note that the ISO model of the US is not similar to the one being discussed in the EU.
E.g. in the PJM region the ISO is like an extension of state power and regulation into the association of
energy firms.
Joskow, Paul L., id..
European Energy Markets
Abel M. Mateus
regulators that deal with ownership unbundled TSO(s), seem to agree that these TSO(s)
reduced the time needed for planning, instructing authorization process and realizing the
investments. The Portuguese case (Annex II) shows that it was the change from legal to
ownership unbundling56 that led to increased investment in the transmission network and
increased interconnector capacity with Spain. The UK case (Annex I) also shows that
ownership unbundling led Transco to increase substantially investments. And Lowe et al.
also cite the case of Netherlands. And in the case of Italy, the ISO experience deteriorated
the efficiency of the system, and after the blackout of 2003 that affected North Italy for
part of the summer, it adopted a TSO.
Therefore, while ownership unbundling is the most effective means to ensure non
discriminatory access to networks and to induce investment, many uncertainties can be
raised about the effectiveness of ISO/TSO model.
Public versus private ownership of the energy assets
There is a wide consensus among economists and regulators that effective unbundling is
both necessary either in the context of publicly or privately ownership of the networks.57
In fact, since the high voltage or high pressure networks are natural monopolies, the
Nordic countries have decided to maintain them under public ownership. Since generators
are mostly private firms, this arrangement does not preclude, and may even accentuate the
benefits of ownership unbundling.
The decision was taken by the Government in 2000 after a black-out affected most of the Lisbon region
with serious consequences.
See ERGEG. Third Legislative Package Input. Paper1: Unbundling. June 2007 and Lowe, Philip, et al..
Effective Unbundling of energy transmission networks: lessons from the Energy Sector Inquiry. Competition
Policy Newsletter. Spring 2007.
European Energy Markets
Abel M. Mateus
The problem arises when both generation and transmission are publicly owned, because in
this case it would be difficult to have the benefits of unbundling even when different assets
are supervised by different ministries or different levels of government.
Legal problems of implementation of the unbundling
First, estimates of costs of unbundling, sometimes involving divestiture, have been
exaggerated. The one-off cost of the British Gas de-merger in 2000 was around 3.2 percent
of the company’s yearly turnover (Lowe et al., p.32).
And the European Parliament, in its Resolution of June 19th, 2007, “19, underlines that the
completion of ownership unbundling in the energy sector, alongside the dismantling of
vertical conglomerates and guarantying conditions for effective market access should be
given high priority.”
Can the Commission impose ownership unbundling without a new Directive? The
Commission is vested with the powers to impose structural remedies under Regulation
1/2003. However, contrary to the UK Competition Commission that can impose those
remedies within the context of a sector inquiry, the Commission can only impose them
under antitrust cases, where there is no equally effective behavioural remedy or where such
a behavioural remedy would be more burdensome for the undertaking concerned. In fact,
the Commission seems to interpret this power as an “atomic bomb”. Accordingly, Lowe et
al. consider that antitrust instruments are not well suited to require ownership unbundling.
In certain MS, like in the UK, there are limits in ownership links, introduced through
licensing rules that prevent e.g. the same group to hold a license for network operation and
to supply/production activities.
Various legal objections have been voiced against introducing ownership unbundling –
with the consequent divestiture – through EU legislation. One of the most frequently heard
is that it constitutes a form of expropriation, running against article 295 EC or general
principles of EC law, like the European Convention on Human Rights and Fundamental
European Energy Markets
Abel M. Mateus
Freedoms. But the argument against this position is that the property right may be limited
by objectives of general interest pursued by the Community and not constitute a
disproportionate and intolerable interference.
Lowe et al. (p. 33) conclude that “the arguments brought forward for ownership
unbundling, and the need to ensure a level playing field across an integrated EU energy
market, clearly justify action at EU level”.58
2- Cross border trade and interconnection: building regional markets
A full functioning integrated energy market requires high levels of cross-border trade
between Member-States to take into account the welfare benefits of international trade
and the impact of the rule of cost minimization. In order to do so, investment is needed
to increase interconnection capacity, since traditionally interconnections were not
dimensioned to allow significant trade patterns.
However, as Newbery notes59 intensifying trade in gas also contributes to the integration of
electricity markets. As gas transmission costs are relatively low, the price of gas should be
more uniform. Thus, new entrants into electricity generation would face similar costs and
would press for price convergence.
However the Commission has to convince two thirds of the Council. At this time there are still about ten
countries against unbundling. On the other side, seven governments have sent to the Commission a letter
pressing for the change (UK, Spain, Italy, Netherlands, Sweden, Denmark and Finland).
Market Design, 2006.
European Energy Markets
Abel M. Mateus
Source: European Commission60
Source: European Commission, Energy Sector Inquiry
Expanding interconnection capacity requires a high degree of coordination between TSOs
and depends on exogenous factors – like environmental impact studies, planning and
authorization procedures. To meet certain expansion deadlines, TSOs must be highly
committed to deal with the difficulties involving expanding existing interconnection
One the major lessons of the European blackouts have been that building and managing
properly interconnectors is fundamental for improving the security and reliability of the
THE EUROPEAN PARLIAMENT, Report on the experience gained in the application of the Regulation
(EC) No 1228/2003 "Regulation on Cross-Border Exchanges in Electricity", May 2007
European Energy Markets
Abel M. Mateus
national energy networks. Since then, there was already a Multilateral Agreement among
EU TSOs (Synchronic Interconnected EU Network) and the adoption of an Operational
Handbook. It was adopted the N-1 security rule for network simulations. However, none
of these initiatives reflects competition concerns. In fact, the balancing agent has the sole
power to use or exclude form access any domestic or foreign supplier.61
From a competition point a view, to determine the right level of interconnection capacity
to integrate markets is a complex task. In fact, the amount of required capacity needs to be
addressed by economic modelling and not by a single rule that “fits all”.62 There are quite a
number of strategic issues involved, and the market is usually one of a sophisticated
oligopoly. An efficient incentives network based system could also provide a market
solution for the problem.
The main objective of interconnection capacity is to guarantee that there exists sufficient
capacity to efficiently arbitrage price differences between price zones, effectively integrating
different zones in one single price zone, but without having necessarily a unique price.
Price differences can be explained by differences in generation mix but also by differences
in concentration levels among price zones. The higher the price differences the more
interconnection capacity is required to effectively harmonize prices. If price differences
attributable to market dominance in one price zone are higher than the perceived cost
difference between generation mixes than, ceteris paribus, the higher is the optimal
interconnection capacity to integrate markets.63
These problems are compounded by the physical characteristics of the generator. A nuclear power plant
cannot be taken out of the system, wind is intermittent, a coal plant will take 6 to 12 hours to start injecting
electricity and a gas fired about half an hour. Hydro power has the lowest connection lag.
Like the present rule of 10 percent of demand.
The interconnector between the Iberian Peninsula and France is almost permanently congested by
Iberdrola, decoupling the Portuguese and Spanish markets from the rest of Europe.
European Energy Markets
Abel M. Mateus
However, the effects of trade between markets of different scale are necessarily
asymmetrical. Integration leads to a higher external competitive pressure over the
incumbent of the small country. The contrary is less evident.
It may be particularly difficult to obtain from a vertically integrated firm the necessary level
of commitment needed to develop interconnection capacity with its neighbouring country.
Even using regulatory powers, it may be difficult to entrust the incumbent to invest, since
incentives to invest in interconnection from a vertically integrated utility are low and
interconnection capacity raises the perceived competition levels over its generation and
retail supplier affiliates.
Incumbents will always claim that do not have any incentives to abuse its dominant
position, accepting or proposing levels of interconnection capacity lower than the optimal
levels that would lead to perfectly integrated markets. Potential delays in fulfilling the
timetable for the interconnection expansion can be easily justified with exogenous factors,
like lengthy environmental authorization procedures or the lack of coordination with the
neighbouring TSO.
To put this claims to test one should analyse the Iberian example. Since 2000, full
ownership unbundling exists both in Spain and Portugal. The two Iberian TSOs, supported
by the political willingness by the Portuguese and Spanish governments to create an Iberian
Electricity market, have set an ambitious investment programme in interconnections to
more than double, by 2008, the interconnection capacity between Portugal and Spain.
Despite the several delays in creating the market – firstly foreseen for the beginning 2003
but only launched in the second semester of 2007 – the two TSO maintained the
investments schedule. Actually, the two TSO are foreseeing new lines to be constructed till
On the contrary, within the same time frame, the interconnection between Spain and
France – where RTE, the French TSO is owned by the incumbent EDF – has maintained
European Energy Markets
Abel M. Mateus
its levels, despite the commitment by RTE, after a merger decision64 to increase this
capacity. The Iberian Peninsula is stated as an Electrical Island – since the France-Spain
interconnection capacity is low.
To create an energy internal market interconnection capacity is critical, however it is not
sufficient. It is also needed to promote the use of existing capacity through market forces,
which requires the development of the market mechanisms to manage congestions.
At the European level, Regulation (EC) 1228/200365 requires that the mechanisms to
manage congestion must be market based. According with the existing regulation, two
mechanisms were accepted: explicit auctions and implicit auctions.
Explicit auction have the consequence of separating energy and transmission markets,
which limits the coordination and efficiency of this model. The European Commission
Energy Sector Inquiry questioned, in particular, this model, considering the irrational
outcomes it may provide. On the contrary, in an implicit auction interconnection capacity
is managed by the clearing house in order to provide the most efficient outcome. Energy
and transmission markets are unified. Energy price differences among zones reflect the
shadow price of using the limited transmission capacity.
Explicit auctions have poorer performances than implicit auctions, in what concerns
efficient use of limited interconnection capacity but also in what concerns market
integration. Implicit auctions have the merit of more effectively arbitraging price
differences, interfering more heavily with price formation in domestic markets.
However, in most of European borders between Member-States explicit auction is the
adopted solution. Notwithstanding, the movement to adopt implicit auctions is significant.
Market splitting can be observed in the Nordic market and in the Iberian Market, which
European Commission, merger case COMP/M.2684 - EnBW / EDP / CAJASTUR /
HIDROCANTÁBRICO, decided in 26/09/2001.
COUNCIL of 26 June 2003 on conditions for access to the network for cross-border exchanges in electricity
European Energy Markets
Abel M. Mateus
started in the second semester of 2007. In the Iberian market, the market splitting
mechanism is also combined with explicit auctions (where traded rights have use or sell it
obligations). The similar market coupling solution was implemented in the first half of
2007 in the French, Belgium and Netherlands markets.
The use of such market mechanisms will lead to a greater convergence of prices between
countries, subject to the observed constrains in interconnection capacity. The more
efficient use of existing transmission capacity will be translated in more efficient arbitrage.
Even though existing limitations in transmission capacity will limit full price convergence,
more trade between countries will influence domestic price formation.
Nevertheless, the effects of this convergence will be asymmetrical regarding small and big
markets. One same level of interconnection capacity can be distinctly interpreted
depending on the market under analysis. The Iberian market clearly shows the different
potential impacts of trade though the interconnections: by 2008, the import capacity from
Spain will represent between 15% and 20% of the Portuguese peak demand; on the
contrary, import capacity from Portugal will only represent about 3 or 4% of Spanish peak
The more interconnected markets are the more effects of trade will arise on domestic
prices. However, these effects are likely to be higher in smaller markets. The London
Economic price study clearly demonstrated that high levels of interconnection capacity in
smaller markets (like in the Netherlands), can have a discipline effect on incumbents
market behaviour in what concerns price formation. The contrary however is less evident.
The larger the domestic market and the larger the concentration levels in that market the
less likely the possibility for cross-border to promote competition and discipline incumbent
European Energy Markets
Abel M. Mateus
An interesting study66 for the most advanced regional market in the EU, the Nordpool,
shows a very significant increase in consumer prices due to the exercise of market power
during certain periods of the day in which dominant positions are created. They are able to
increase peak prices by about 30 percent and non-peak prices by 10 percent, on average.
Simulations also show the detrimental impact that mergers of firms across countries can
have in the Nordpool price, even when traditional measures of concentration do not show
a serious problem.
Integrating smaller and larger markets will lead to a price convergence that will be most
likely dominated by the price level of the largest market. In this process, the market
concentration levels of the largest market are determinant for the overall price level of the
integrated market.
For small countries bordering a large one, regional integration may pose particularly acute
problems, if national markets are not competitive. The situation arises because a dominant
position in the large country will be extended to the small one. On the other hand, a
competitive market in a small country can not automatically be extended to the larger
neighbour country. As a matter of fact, creating a competitive market in the small country
requires that a competitive market be simultaneously created in the large country.
Interconnection between the small and large country will not solve the problem of creating
a two-country competitive market. Also suppose that the small country has been able to
build a competitive market, than by linking-up with the large un-competitive market will
clearly lower welfare. This example clearly demonstrates that it is not enough to build a
competitive single market by building up an interconnection system.
Simulations carried out for Netherlands regarding closer market integration with
neighbouring countries following the experience of Nordic countries show how tricky are
A Powerful Competition Policy: towards a more coherent competition policy in the Nordic market for
electric power. Report from the Nordic Competition Authorities nº 1/2003.
European Energy Markets
Abel M. Mateus
the problems, specially with strong cross-border ownership (Electrabel in Belgium and the
Netherlands, E.On and RWE spanning Dutch-German border, in both cases owning
transmission and interconnection).67
3- Merger policies at National an Community levels: questioning the 2/3 rule
The US Department of Energy stated in 2000 that “antitrust remedies are not well-suited
to address problems of market power in the electric power industry that result from
existing high levels of concentration”.68 This is why structural remedies are usually required,
as the UK experience abundantly shows.
Merger policy is a preventive tool vis-à-vis the build up of dominant positions. Its correct
implementation is thus essential. In fact, recent restructuring led to the emergence of panEuropean energy companies. But the path that is leading to this scenario shows several
inconsistencies in merger control practice, and as we show below could lead to a defeat of
the objective of building a EU-wide competitive market.
At the national level, several pleas exist to support the creation of energy “national
champions”. Large countries with large energy companies use their financial muscle to
acquire smaller energy companies in other national markets, sometimes with the banner of
building “European champions”. But aside transmission networks, which are natural
monopolies, there are no efficiencies to gain from generation concentration. Different
technologies allow efficient power plants of different sizes, as the US experience shows.
Some defend the need to balance the scale disadvantage (from a conglomerate perspective)
that some countries incumbents face against the larger countries national incumbents to
concentrate their energy companies; others have fears of foreign take-overs and defend
This highlights the importance of merger control viewed too narrowly in national terms, and according to
D. Newbery. Market Design, shows that the European Commission needs to have a deeper analysis of the
We would say the same about all network industries.
European Energy Markets
Abel M. Mateus
that a larger national company would be more difficult (more costly) for a foreign firm to
acquire. Still others support the idea that national companies are more prone to accept
domestic political influence, or claim that energy is a strategic sector that should be
sheltered from Community influence.
At the same time, vertical relations between gas and electricity are clearly developing.
Natural gas is an increasingly important source of energy for electricity production. New
entrants in electricity production often choose natural gas as a production input. From this
convergence between both industries, market players have identified merging opportunities
to vertically integrate gas and electricity businesses. Recent merger case history shows
clearly the type of competition problems that can arise from this vertical integration
process, especially when undertaken between dominant players in each of the two
industries, and that puts a barrier to a more competitive regional or EU-wide market.
Finally, recent merger case history shows differences in approach and methodologies
between the European Commission and National Authorities, and among different
national jurisdictions. Let us review some of the recent cases.
E.ON/Ruhrgas case
The mergers between E.ON/Ruhrgas, dealt with by the Bundeskartellamt in 2002, and
EDP/ENI/GDP, dealt with by the Commission in 2004, clearly show these
inconsistencies. In both cases, it was clearly showed that without appropriate remedies, the
vertical integration between gas and electricity dominant players could pose several
constraints to competition, namely the removal of strong potential competitors at retail and
wholesale levels and the potential foreclosure of access to natural gas for electricity
generation. In both cases a prohibition decision was issued by the Competition Authorities
on the grounds of its negative effects on competition.
In the E.ON / Gelsenberg (Ruhrgas) case, Bundeskartellamt’s assessment demonstrated
that the merger would have strengthened dominant positions both in the gas and electricity
European Energy Markets
Abel M. Mateus
retail supply markets. Indeed, Ruhrgas was the largest German grid gas company,
accounting for nearly 60% of total natural gas demand, and the only German gas player
with access to all the different sources of gas supplies for Germany (namely supplies from
Norway, Russia, the Netherlands, Great Britain and also the existing domestic sources). As
for E.ON, it was the second largest electricity player, vertically integrated and also present
in the supply of natural gas, namely thorough its shareholder positions in regional
electricity and gas distributors and in municipal utilities.
According to the assessment of Bundeskartellamt, the merger of Ruhrgas and E.ON would
structurally secure Ruhrgas’ sales to E.ON affiliates and holdings. The foreclosing of a
significant share of the wholesale gas market would therefore strengthen Ruhrgas dominant
position. E.ON affiliates that where located in Ruhrgas’ transmission area would also
strengthen their dominant positions in supplying large gas consumers and local gas
distributors, since Ruhrgas would be removed as a potential competitor. In electricity
markets, the merger would also produce detrimental effects, contributing to strengthen the
collective dominance of E.ON and RWE (which had also a gas subsidiary – Thyssengas),
considering the potential foreclosure of natural gas access for new entrants in generation of
The parties offered commitments in order to address the competition concerns raised by
the merger, however, they were considered to be of little significance to prevent the
strengthening of the dominant positions in the gas and electricity markets. Therefore the
Bundeskartellamt decided to prohibit the merger between E.ON and Ruhrgas.
Notwithstanding, after an appeal by the merging companies to the German Minister for
Economy the permission was granted, on the grounds that overall economic advantages
stemming from the merger had been demonstrated.
European Energy Markets
Abel M. Mateus
A number of experts have claimed the damage caused by this decision, e.g. David Newbery
claims that the European Commission should have a jurisdiction on such a case, due to its
impact on electricity and gas market in other countries of the EU.69
On the contrary, in the merger EDP/ENI/GDP, which was supported by the Portuguese
Government, with the main argument that the incumbent needed to gain scale to compete
in the wider Iberian market, the prohibition by the European Commission in 2004
effectively blocked the merger, a decision that was uphold by the Court of First Instance in
September 2005.
The Case EDP/ENI/GDP concerned the proposed acquisition by Energias de Portugal
(EDP) and ENI of joint control over Gás de Portugal (GDP). EDP is the Portuguese
electricity incumbent in Portugal, at the time, with more than 70% of the Portuguese
generation capacity and almost a monopolist in electricity distribution throughout the
country, as well as present in the Spanish natural gas and electricity markets (through its
Spanish affiliates Hidrocantabrico and Naturcorp). ENI is an Italian company active
internationally at all levels of the energy supply and distribution chain. GDP is the
incumbent gas operator in Portugal, active at all levels of the gas chain in Portugal, at the
time with exclusive rights for import, storage, transportation, wholesale supply of natural
gas and the control of five of the six Portuguese local gas distribution companies (the sixth
being controlled by EDP).70
One of the main disputes in the case concerned the geographic market definition of the
electricity wholesale market. Since the Iberian Wholesale Market was due to start in the
Newbery Market Design
The Portuguese Competition Authority (PCA) had already anticipated the problems of this concentration
in the study commissioned to Cambridge Associates of D. Newbery. See the site of the PCA:
European Energy Markets
Abel M. Mateus
near future, the parties alleged that the wholesale electricity market would be Iberian
(Portugal and Spain). However, considering the high congestion levels in the
interconnection between Portugal and Spain, which would remain high in the future,
despite the undergoing upgrade in interconnection capacity, the Commission ruled out the
possibility that the wholesale electricity market would be Iberian in the future.
The Commission assessment of the case demonstrated that the merger would strengthen
EDP’s dominant position in the electricity wholesale and retail markets in Portugal. Indeed,
it would remove GDP as a potential strong competitor in the electricity markets.
Furthermore, since new entrants in generation chose natural gas as the primary source of
energy, the concentration would have made the existing and possible new future power
producers in Portugal dependent on their main competitor, namely EDP. The
concentration would also strengthen GDP’s dominant position in the relevant gas markets
in Portugal, through the foreclosure of a significant part of the gas demand (controlled by
EDP) and the elimination of EDP as most likely entrant in the gas markets.
Endesa/Gas Natural
In 2005, in Spain, similarly, the vertical merger between Gas Natural, the gas dominant
player in the Spanish market, and Endesa, the largest electricity player in Spain, would have
allowed for strong impediments for competition in the Spanish natural gas and electricity
markets. It would strengthen Natural Gas dominant position – being Endesa an entrant in
the gas retail market with prospects of high growth of its market share. Since Gas Natural
was previously the strongest new entrant in generation, the new merged entity would
acquire a market share in the Spanish electricity wholesale market higher than the before
held by Endesa, raising the concentration levels of the Spanish wholesale electricity. The
European Energy Markets
Abel M. Mateus
merger was also likely to raise vertical effects similar to the merger cases previously
presented. 71
Notwithstanding, despite the Spanish Competition Authority recommendation to block the
merger, the Spanish Council of Ministries decided to approve the merger, based on the
recommendation of the sector regulator, subject to some mild conditions. Endesa appealed
to courts of the Government decision. In the meantime E.On made a counterbid clearly
above the one of Gas Natural. The merger was approved by the European Commission,
but the Spanish sector regulator imposed a number of substantial remedies, after the
Spanish government enact increased powers equivalent to almost a veto to its sector
regulator. E.On appealed to the Commission that initiated procedures against the Spain.
The arm twisting was lingering on, and the Spanish construction conglomerate, ACS,
increased its share in the capital of Endesa to slightly below 30 percent, making it very
difficult for E.On that required 2/3 of the capital to take control of Endesa. Two years
later, the situation was finally resolved when the Italian state company Enel bought part of
the stockholdings and jointly with ACS stroke a deal with E.On that remained with a small
part of Endesa assets and thus withdrew the bid.
GdF/Suez case
The merger between Gaz de France (GDF) and Suez, approved by the European
Commission in 2006, involved the dominant gas incumbents of France and Belgium, each
vertically integrated, in two neighbouring countries and actively entering in each other
The PCA appealed to the Commission asking for a referral of the case from the national authorities to the
Commission, which was opposed by the Spanish government. After a lengthy debate, revolving about some
detailed accounting issues, the Commission accepted the position of the Spanish government. The tenet of
the Commission vis-à-vis the request of the PCA was that the merger in Spain would not have any significant
impact in Portugal, despite the projected increase in the interconnector that was to reach 20 percent of
Portuguese electricity demand.
European Energy Markets
Abel M. Mateus
domestic market. Moreover, since Electrabel, the Belgium electricity incumbent, was a
subsidiary of the Suez group, and GDF, along with Centrica, had joint control over SPE,
the second player in the Belgian electricity and gas markets, the merger affected also the
Belgium electric market, raising also vertical effects.
As stated by the European Commission, concerns related mainly to the removal of the
increasing competitive pressure that GDF and Suez had so far exerted (and would have
exerted in the foreseeable future) on each other in both Belgium and France.
In the Belgium gas market, the merger would have removed GDF as the strongest
competitor of Distrigaz (Suez group). This merger would also remove GDF as a
competitor to supply gas-fired power generators competing with Electrabel, the dominant
Belgium electricity incumbent, therefore affecting the competition conditions in the supply
of gas to power generation. Moreover, given the control of Fluxys, the Belgium TSO, the
GDF/Suez merger would have allowed the parties to have privileged access to a very
important transit gas network, where a gas hub is located, linked with several other markets
and with an LNG regasification terminal.
In France, since Distrigaz was a strong new entrant in the French gas markets, through the
merger this competitive pressure over GDF’s dominant position would cease to exist.
Since the Suez group was the largest player on the French market for district heating and
GDF the second, the merger would raise the concentration level in this market.
In order to address this competition concerns, a set of remedies were proposed, namely the
divestiture of activities where a horizontal overlap was identified. In the Belgium gas
market, Suez would divest Distrigaz, solving the horizontal concerns in the Belgium gas
market and the vertical concerns related with access to gas for power generation. GDF
would divest its SPE shareholding, solving the concerns in the Belgium electricity market,
European Energy Markets
Abel M. Mateus
and its subsidiary Cofathec-Coriance, to address the concerns in the district heating market
in France.
Suez would also relinquish its control of Belgian gas TSO Fluxys, namely by reducing its
stake in its share capital to 45% and implementing an independent governance model. In
this model, the merged entity would only be able to nominate 1/3 of the board of
directors. A management committee, independent from the board, with full powers to
manage all aspects of Fluxys activities, would be created. The rules for the management
committee nomination would ensure its independence from the board of directors. In
Fluxys International (owning the transit pipelines and the LNG terminal) the merged entity
stake would be restricted to 60%, however, granting Fluxys the requisite rights to use
installations and equipments of Fluxys International.
The parties also presented the commitment to carry out a series of investment projects in
Belgium and France in order to increase infrastructure capacities and facilitate entry.
Through commitments, the Zeebrugge hub would also act as an aggregator and therefore
increase market liquidity.
As far as the Belgium market is considered, through the remedies, since Distrigaz and
Fluxys would remain under the control of different entities there would be a structural
separation of the dominant gas market player and the gas TSO.
However, being France and Belgium both transit countries for Norwegian gas that flow
from North to South (the Benelux Italy Axe, as identified in the European Commission
Sector Inquiry), this will lead to the fact that two alternative transit routes (the one that
crosses France and the one that crosses Belgium and Germany on the way to Italy) will
cross two gas TSO with similar shareholders. Being transit flows strategic for the gas
internal market construction, this framework might lead to competition problems in access
to transit networks and new investments regarding transit pipelines.
European Energy Markets
Abel M. Mateus
It may be argued that the governance model implemented with the remedies for Flexys
may solve the potential competition concerns in gas transit flows that might result from the
merger. In fact, the commitments offered by the parties were designed in way to impede
(or make it difficult) the future merged entity to block the independent management
committee investment decisions over Fluxys and Fluxys international. Furthermore, some
commitments were offered regarding gas transit.
However, the adopted model for relinquishing the control over Fluxys are far from the
ownership unbundling of the TSO model adopted in several countries, where market
players are restricted to a maximum of 5% of the TSO’s share capital. In fact, the
GDF/Suez group would still be able to appoint 1/3 of board of directors in Fluxys. And,
as far as the Belgium gas market is concerned, will remain a market player (through GDF
former market share) with a stake of 45% of the TSO’s capital. The standard of
independence that this governance model presents has not yet been properly tested and
shares the same concerns as referred elsewhere in this paper.
European Energy Markets
Abel M. Mateus
Problems with promoting European champions
The flurry of cross-border acquisitions in EU markets that we have witnessed in the last
decade, taking advantage of fragmented national markets is a serious obstacle to the
functioning of competitive regional markets and an integrated EU-wide market. The
problem can be illustrated by the following example with a large (X) and a small country
(Y), and suppose the two countries have a limited volume of interconnection. In country X
there are two major electricity generators and in country Y a dominant player. Suppose
now that firm A from country X acquires firm C from country Y. Since it is a cross-border
merger it goes to the Commission. Since firm A does not have business in country Y and
the increase in its share in country X is low, the merger may be approved.
Country X Firm A
Firm B
Firm C
Country Y
Firm A+C merge
Country X Merged
Firm B
Firm C
Country Y
However, a crucial aspect is that the Commission will consider two separate national
markets, since the level of interconnection is low. Now, let us consider two situations. If
the level of interconnection is substantially increased and the market is integrated there is a
serious problem of concentration: the HHI jumps from 4229 to 4950. But if the level of
interconnection is reasonable and might even be high (the present required level is 10
percent of demand, but obviously 10 percent for country X is quite different from 10
percent of country Y), most of the time and at least in peak hours the interconnector
would be congested, which means that the merged firm not only has the capability to
European Energy Markets
Abel M. Mateus
control the price in country Y but also either individually or in conjunction with firm B
control the price in country C. In fact, in the regional market there was a reduction from 3
to 2 big players.
The problem arises because we still have fragmented markets that require geographical
national dimension and regional integration requires already a different approach. The
problem is that the sequential analysis leads to inconsistencies.
This is not purely an abstract example, simulations for the Nordic market show that a
merger of a large Finish generator with a large generator in Denmark, even without any
overlap of geographical markets could generate a substantial increase in prices in their
home countries. Figure x shows an increase of about 50 percent in peak hours in Denmark.
Source: Nordic Competition Authorities. A Powerful Competition Policy, op.cit..
The 2/3 merger rule for case allocation
Analysing these cases one has to conclude that inconsistencies in merger control may
undermine the efforts in smaller neighbouring countries to construct competitive
frameworks. In effect, even though wholesale electricity market geographic definition may
rule out that the markets are all the time integrated, if one accepts the geographic market
European Energy Markets
Abel M. Mateus
definition adopted by the European Commission and the Nordic Competition Authorities
in the Nordpool72, where geographic markets are defined according to the market splitting
congestion mechanism – e.g. wholesale market can be Nordic in some hours, with a single
price for all the grid zones and regional in other hours, defined according to the price
zones that are defined – the effects of a merger in one country might not be confined to
the market where most of its effects are created. In larger countries/markets, the creation
or strengthening of a dominant position might have effects, through cross-border trade, in
its neighbouring countries.73
Allowing concentration levels to increase at the domestic level might have strong
detrimental effects on internal energy market construction. Therefore, it is necessary to
guarantee that merger control in the energy sector, whether appraised at the European level
or at the national level, follows the same principles. The share of competences between the
European Commission and the National Authorities in what concerns merger control
should not allow for different interpretations. This share of competences should not serve
the way to approve mergers that lead to the creation of energy “national champions”.
Sorgard74 has also raised the problem of the “beggar-thy-neighbour” regarding national
champions. Suppose a country forms an energy champion, then its neighbour also wants to
form an equal energy champion in terms of countervailing power. Due to the externality
involved he also favours a revision of the 2/3 rule.
The EU energy internal market needs more infrastructures to integrate markets. But it also
needs a coherent merger control practice within all member states.
See European Commission decisions on the cases M.3867 VATENFALL / ELSAM and E2 Assets,
decided in 2005/12/22, and M.3868 DONG/ELSAM/ENERGI E2, decide in 2006/03/14.
Among EU energy firms the ones that have been buying most of the other EU countries companies are the
French, German and to a less extent the public owned Italian incumbent.
Lars Sorgard. The Economics of National Champions. European Competition Journal, June 2007. He also
refers a case where the champion might have some monopoly power on a larger market than the national
market, and thus be able to appropriate consumer welfare from the foreign consumers.
European Energy Markets
Abel M. Mateus
The 2/3 Rule for allocating merger cases is in need of modernization. In fact, we
can deduct the proposed new rule from a purely abstract model, in the best
tradition of conceiving laws and regulations, without referring to any case in
particular. No single case has been raised in the tradable sector. The problem
arises in nontradables, with strong network and scale economies (infrastructure),
which are usually regulated markets. In these markets, mergers have a local and
national impact, but they can have also a strong cross-border effect, and will have
an ever stronger impact as the European integration proceeds – as we all wish.
Let us suppose we have a given undertaking, with a given Minimum Efficient
Scale, that is relatively large, when compared with the size of a small economy
market, due to scale, network and scope economies. Now, it is probable that the
large country undertaking has most of the operations within the large country,
and the small country undertaking has assets in other EU countries and/or a
controlling part of its assets is owned by a firm of another EU country. And this
probability will increase with financial and economic integration. However, a
merger within two undertakings in a large country most probably would fall under
the NCA, and the merger of two undertakings in a small country would go to the
There would be no problem as to what jurisdiction decides on the case, if there
was full harmonization in merger rules among EU countries. But, in contrast with
antitrust rules, this is not the case. This is there are still countries whose merger
control regime allows for decision criteria based on “national interest”, as an
extraordinary appeal to the Minister of the Economy may be lodged in case of a
negative decision by the PCA.
We badly need some type of harmonization of merger rules, and this is the core
of the problem. Let me say that there is now a clear imbalance between merger
European Energy Markets
Abel M. Mateus
cases and Regulation 1/2003 dealing with articles 81 and 82, where the
Commission can take a case from a NCA in order to ensure coherence of the EC
competition rules. Accordingly, the 2/3 Rule should be qualified, giving the
Commission the right to avocate the case if it is a network sector, and the case is
important for community policy reasons. Any MS competition authority or
government should also have the power to ask the Commission to take the case
or have the right of referral.
This position has already been echoed by the European Parliament. In its
Resolution of June 19, 2007, concerning Competition Policy, it states: “24. Recalls
the Commission’s commitment to review the “two-thirds rule” as a threshold for
finding a Community impact in regard to merger proposals, suggests that progress
in this area and a more consistent approach in the evaluation of comparable
merger operations would be welcome whenever decisions taken at national level
could have a strong impact on the market structure of neighbouring Member
Methodologies used for merger analysis and its impact
Mergers in electricity markets cannot be analysed with the traditional measures of
concentration. The US FERC uses a screen consisting of two tests: a market share and the
pivotal supplier test. The merger passes the pivotal supplier test if, assuming the merged
entity is removed from the market, the remaining suppliers have sufficient capacity to meet
the demand at a similar price. If it fails the test can still rebut the case offering evidence
based on a delivered price test or filing a mitigating proposal accepting e.g. cost-based
regulated rates.
However, it is possible to prove that those tests are still unsatisfactory. The best merger test
is the Competitive Residual Demand (CRD) analysis based on a simulation model of
residual demand (the wholesale market demand less the aggregated uncommitted supply
European Energy Markets
Abel M. Mateus
from all the other firms), to test if the merged firm has an incentive to reduce output or
raise prices relative to pre-merger levels. Obviously, market demand will differ at different
points in time.75
There is quite diversity in the methods used in the EU. The Italian and Spanish authorities
used the pivotal supplier test. The Commission uses a qualitative analysis that presupposes
some of these methods, namely when it looks at its order in the supply schedule. However,
in one case referring to the Swedish and Danish markets they used the MARS model, used
by the Nordic competition authorities, to test its qualitative analysis.76
NMa has also used a sophisticated simulation model to study the Nuon-Reliant merger.
Most of the applied case studies show that in each relevant market, for passing the tests of
Newbery strategic supply function criteria and CRD we need to have at least four
producer firms of similar capacity. This is clearly much more stringent than traditional
market share analysis used by most of the competition authorities.
Do mergers of electricity companies have a significant impact on improving firm
efficiency? The track record is not encouraging. John Kowka and Michael Pollitt77
conducted an extensive study of 75 mergers in the US from 1994 to 2002, involving half of
the customers of all investor-owned electricity companies. They test for the impact of the
merger in efficiency using data envelopment analysis. Their conclusion is that are usually
less efficient firms that buy more efficient firms and after the merger seller’s efficiency
declines towards the norm. This is clearly not evidence that mergers in electricity increase
D. Newbery. Electricity Merger Policy in the Shadow of Regulation, October 2006.
COMP/M.3867-Vattenfall/Elsam and E2 Assets.
Industry Restructuring, Mergers and Efficiency: Evidence from Electric Power. CWPE 0725. April 2007.
One specific example in Massachusetts, where Nstar bought some of the neighbour distributors promising
a reduction in operating costs of 13.7 percent, that never materialized. (See Kowka and Pollitt, op. cit.)
European Energy Markets
Abel M. Mateus
4- Competition policy in an environment of global concentration in natural
The trend for concentration of oil and gas supply in an ever smaller number of countries is
a well known fact. According to the IEA proven oil reserves are concentrated in the
Middle East and North Africa (62 percent of the world total). As a result, OPEC oil
production will increase from 40 percent of world production in 2006 to around 50 percent
in 2030, with Russia, Central Asia, Latin America and Canada the only regions to increase
production. Although reserves of natural gas are more widely distributed, 56 percent are
concentrated in Russia, Iran and Qatar. By contrast coal and uranium are more widely
distributed. Besides this concentration of resources, there are factors that can limit
investment and trade in the future: (i) resource nationalism, with the State assuming an
increasing intervention in order to restrict or discriminate access to foreign companies; (ii)
“national champions”, State or State controlled companies or even private companies
protected by the State, that control the supply of energy or natural resources used for
production of energy. These dominant companies lack a structure of governance that
presses them to be efficient and responsive to the markets, leading either to under
investments and/or overpricing. They are also more prone to political influence; (iii) abuse
of market power, which can be used over reserves, use of pipelines for transport of energy
or supply of electricity at peak hours, that enables States or enterprises to exert significant
influence over production or prices for some period of time. Coordination or cartelization
of suppliers about production, prices or investments can also increase their market power,
reducing market effectiveness; (iv) threat of terrorism, war or accidents, can increase the
likelihood of supply disruptions and industry risk, particularly when energy is being moved
across large distances, hampering investment and increasing production costs.
European Energy Markets
Abel M. Mateus
Two other factors can compound those risks: (i) lack of information about market
variables, like production, consumption and stocks of fuels that can exacerbate price
volatility; and (ii) regulatory uncertainty, that can undermine confidence of investors.
EU’s energy dependence will increase from 50 percent of total energy consumption today
to 65 percent in 2030.79 In comparison, US imports of energy will increase from 30 percent
in 2005 to 32 percent in 2030.80
Let us look at the concentration in gas production and how it will affect Europe.
According to some estimates in the next 10 years Russia will increase its market share in
total EU consumption from 26 to about 30 percent, Norway will remain at about 13
percent and Algeria will increase from 11 to 15 percent. It is possible that Russia plus
Algeria will be supplying close to half of gas consumption by 2016.
Commission, An Energy Policy for Europe, Communication from the Commission to the European
Council and the European Parliament, Brussels, January, 10th, 2007.
Reference scenario of the Department of Energy, Annual Energy Outlook, 2006.
European Energy Markets
Abel M. Mateus
International trade should be based on the functioning of open domestic market
economies and the free movement of goods and capital. These are principles that
economic theory has demonstrated for a long time that lead to maximization of world
welfare and largely reflected in the charter of WTO. It is well grounded the fact that the
exchange of goods between countries is a positive sum game. The period of protectionism
between the two World Wars shows how detrimental to developing countries and the
wealth of all nations the use of external trade as a political instrument can be.
Energy is an input used in almost all economic activity, and we live in a society that is
largely dependent on affordable energy resources. So energy is recognized as a strategic
resource by all governments. So the use of energy as an arm of influence in international
relations is an instrument that can lead quickly to the rise in political tensions and the threat
of serious conflicts.
How should the EU lower the risk of these threats? The Commission Green Paper on
Energy establishes the need for a “coherent external energy policy” to enable EU “to speak
European Energy Markets
Abel M. Mateus
with one voice”.81 In this regard it points the need to (i) define a clear policy on diversifying
the sources of energy, like new independent gas pipelines from the Caspian region, North
Africa and Middle East to the centre of Europe, and new oil pipelines from the Caspian
region through Romania, Bulgaria and Ukraine. LNG terminals in regions without an
effective competition on gas; (ii) energy partnerships with major producers/suppliers, in
particular a new EU-Russia partnership and cooperation agreement,82 efforts to secure
Russia ratification of the Energy Charter Treaty83 and conclusions on the Transit Protocol,
developing a pan-European Energy Community84 with strategic partners Norway, Turkey
If gas and oil are commodities they should be included in the external commercial policy of the
Total trade between the Eu and Russia rechead 166 billion euros in 2005, with trade surplus for
Russia representing 8 percent of its GDP. Russia is the third most important trade partner of the
EU (7.3 percent of the total) and the EU is the most important trade partner of Russia. The EU is
also the most important foreign investor in Russia.
The Energy Charter Treaty of 1994, entered into legal force in April 1998, is a multilateral
framework for energy cooperation that is unique under international law. It focuses on “four areas:
(i) protection of foreign investments, based on the extension of national treatment or mostfavoured nation treatment and protection against key non-commercial risks, (ii) non-discriminatory
conditions for trade in energy materials, products and energy-related equipment based on WTO
rules, and provisions to ensure reliable cross-border energy transit flows through pipelines, grids
and other means of transportation, (iii) resolution of disputes between participating states and /or
investors, and (iv) promotion of energy efficiency and attempts to minimize the environmental
impact of energy production and use”. For more information see Twenty six
countries have signed the Charter, besides the EU Member States, with Russian Federation and Belarus with
a provisional status. The EU should aim for a full application of these two countries.
The Energy Community Treaty, signed in Athens on October 25, 2005, aims to extend the EU internal
energy market to the South East Europe Region, with the following “goals: (i) to create a stable and
European Energy Markets
Abel M. Mateus
and Ukraine, Algeria and Caspian countries, and promoting long-term investment in transEuropean energy networks and their extensions to third partners; (iii) develop new and
targeted instruments to react to external crisis; (iv) intensify the dialogue with major energy
consumers, in particular with the US, China and India, on global issues on research and
development of new technologies, global market access and investment trends, widening
the geographic scope of the EU Emissions Trade Scheme and promote an international
agreement on energy efficiency; and finally (v) use energy to promote development by
developing micro-generation and renewable resources (e.g. only 7 percent of Africa hydropower has been developed).
The Commission is taking the right approach: it “should speak at one voice” in
negotiations with major supplier States, to establish the cooperation framework. Thus,
negotiations with Russia, Algeria or OPEC should be conducted by the Commission and
not by individual Member States (MS). It increases bargaining power, but also because what
is at stake is the supply of energy to several countries and what the Commission can offer,
in a bilateral trade agreement85 is much more than MS can. Moreover, multilateral initiatives
like some of the ones cited above can also improve the world welfare. The IEA, OECD
and WTO could provide an important contribution to the solution of some of the
problems of security and competitiveness of supply. Linking energy problems with
development is also an essential policy for the World Bank.
regulatory market framework capable of attracting investment, (ii) to create a single regulatory space for trade,
(iii) to enhance security of supply; and (iv) to improve the environmental situation and to develop electricity
and gas market competition, and in particular prohibition of cartels, abuses of dominant positions and
distorting public aid, on a broader geographical scale.” More information is provided in The following countries have signed the Treaty: Albania, Bosnia and Herzegovina, Croatia,
FYR Macedonia, Montenegro and Serbia.
It is important that these States understand that bilateral and multilateral trade is a positive sum game.
European Energy Markets
Abel M. Mateus
Second, in the same vein, a country to country approach should be avoided, since it
weakens bargaining position of MS. Only after the cooperation framework is established
then individual MS and their companies should negotiate contracts with external suppliers.
This strategy has not been followed and sometimes large MS have rushed to establish
contracts with major suppliers, either foreclosing the role of other smaller companies or
MS, or weakening the position of the EU.
Third, the policy advocated by some large MS to build national champions in the field of
energy is flawed from several viewpoints. This approach invites the same response from
supplier countries,86 that will claim discrimination if they cannot have the same rights as the
national champion either in that country or in a third country within EU. Second, a
national champion policy by giving or protecting monopoly rights to an incumbent creates
a monopoly rent that makes the company more prone to a foreign take-over.87 88
Since the EC Energy Charter Treaty establishes that foreign investors should have a
treatment similar to national investors, respect of competition rules and open capital
markets, it would be difficult for some MS to claim a “national champion status” for its
Russia, eg., has recently introduced a bill prohibiting foreign ownership above 49 percent in 31 strategic
sectors, including the energy sector. This has been criticized by the EU Parliament (see its Resolution of 19th
June, 2007).
In a more colourful note: “fattening the pig to entice other’s appetite”.
This position does not preclude blocking powers for national security reasons, established on an
institutionalised and accountable basis as the US President and the Committee on Foreign Investments in the
US (CFIUS) have. Germany is proposing a similar EU body. Although the UK is opposed to any controls,
despite its merger law also established a veto by the secretary of state based on national security reasons; it
also calls for reciprocal access to the investor’s capital markets. This might amount to blocking power for
China, which is largely closed to foreign buyers, and Russia, where large strategic investments are difficult.
Obviously the problem is particularly serious when handing the control of “critical infrastructure” like energy
and banking to state-owned non-EU interests. There are some activities that are still protected by the public
security exception.
European Energy Markets
Abel M. Mateus
incumbent and thus oppose an undertaking of a non-EU Energy Charter signatory to buy a
stake or control of a its national energy company. Can EDF be treated in a different way by
the French government than Gazprom? Or the other way around, can the Russian
government89 block the acquisition of a stake or control of EDF in a Russian energy firm?
There are currently some important clauses of anti-competitive practices in international
trade: one is the territorial exclusivity in gas contracts that limit distribution within national
boundaries, which is a serious limitation to cross-border trade. This is the case of long-term
gas contracts between Sonatrach and some EU incumbents. The Commission reached an
agreement on July 14, 200790 with Algeria covering all existing and future LNG and
pipeline contracts including contracts where the gas transits another MS. The settlement
includes the deletion of all territorial restrictions in existing and future contracts. As regards
profit split mechanisms (PSMs) for both pipeline gas and LNG, the settlement is based on
the passing of title and risk to the buyer. Once this occurs PSMs are no longer allowed.
Sonatrach is free however to renegotiate a different delivery point in the contract. For new
contracts the settlement takes effect immediately. For existing contracts, Sonatrach has to
renegotiate and remove restrictive clauses within a reasonable period of time. This is an
important achievement in promoting an integrated energy market.
The overall conclusion is that the best response for security reasons and to build a
cooperative framework with our energy partners is to build a competitive integrated EU
market91 combined with a common approach to foreign matters.
Once it becomes a full member of the Charter.
This also the view of the Commission Green Paper and UK White Paper.
European Energy Markets
Abel M. Mateus
Contrasting the energy scenarios in the USA and EU
Crisis periods are usually the more appropriate for political action. However, the problem
of building competitive markets may be swamped by the “command and control” decisions
of the governments about energy supply. Government intervention may, in given
circumstances, outweigh the benefits of competition. Let us analyse the present energy
scenarios of the USA that do not imply any additional cost to the consumer and the EU´s
that raise significantly the price for users. This also raises the problem of the level-playingfield between the European and American firms in terms of competitiveness.92
It is a fact that we need to prepare for a world with scarce oil and prevent global warming.
However, the two largest economic blocks in the world are pursuing radically different
strategies. The current US energy outlook from 2007 to 2030 is based on world oil price of
50 to 60 USD, at constant 2005 prices.93 According to the base scenario US energy
dependency increases from 30%(2005) to 32%(2030); imports of oil products will increase
from 22 in 2005 to 28.6 Quadrillions Btu (QBtu) in 2030. And natural gas imports will
increase from 4.4 to 6.5 QBtu. The structure of supply changes in favour of solid fuels.
Increased electricity generation is mainly based on coal (addition of 156 GW).
Consequently, there is an increase in pollution: a 1.5% annual increase in CO2, reaching 8
billion tons a year. Non-hydroelectric renewables represent 3.5 percent of total energy
This is a problem for making these policies politically acceptable. At the time that energy prices will
increase due to the switch to renewables – in the case of Portugal there are simulations that show an increase
of 30 percent to the households – it is very difficult to politicians to show the benefits of introducing more
competition: in fact, prices will keep rising.
In nominal terms would be about 90-100 USD. One important argument to consider is that if oil prices
increase more than this reference scenario there would be quite a number of alternative technologies to
produce oil. One possibility is the exploration of extensive reserves of tar sands in Canada. OPEC is certainly
aware of those alternatives, although may not be able to control completely short-term prices.
European Energy Markets
Abel M. Mateus
production in 2005 and it is supposed to increase to 7.5 percent in 2030, with about 80
percent coming from biomass.94 Coal prices are assumed to remain constant in real terms,
and natural gas prices to decrease at 1 percent per year (2005-2030). Thus, average
delivered electricity prices are assumed to remain constant at 81 USD per MW. Both US
and the EU assume the same increase in efficiency in production and use of energy: energy
efficiency is due to increase by 1.5% per year.
EU energy dependency in the current policy scenario of high use of renewables will
increase from 52 percent in the present to 58 percent in 2030, about 6 percentage points
below the “business-as-usual” scenario. In the structure of supply, renewables are supposed
to increase from around 7 percent in 2007 to 20 percent in 2020, according to EU Council
and Parliament targets, and a further 4 points until 2030.
Use of solid and liquid fuels
decreases dramatically (from 17 and 37 percent in 2007 to 10.7 and 31.4 percent in 2030).
Nuclear is also reduced by half (from 14 to 7.5 percent). The increase comes from natural
gas (from 24 to 26 percent) and above all from renewables (from 8 to 24 percent). The
share of renewables in electricity production increases from 23.5 percent in 2010 to 45.6
percent in 2030. As a result, CO2 emissions decrease by 11 percent in 2030 relative to the
1990 level, and about 9 percent below its present level. Wind generates already half of the
non-hydro electricity production (above 40 GW), and its production is expected to more
than double until 2030.96 The other large contribution comes from biomass and waste.
Recently, the Commission has studied the costs of this scenario. Assuming that oil prices
stay at around 48 USD per barrel, and a 25 euro per ton of CO2., the extra cost to reach a
20 percent renewables target, instead of the 12 percent in a business-as-usual scenario,
DOE. Annual Energy Outlook, 2007.
European Commission. Scenarios on Energy Efficiency and Renewables. European Energy and Transport,
July 2006.
Representing about 15 percent of total electricity generation.
European Energy Markets
Abel M. Mateus
would be 24 to 31 billion euros per year.9798 Thus the total additional cost of this renewable
intensive policy is a 40-60 billion euros a year, equivalent to 0.5 percent of GDP, which is a
substantial burden on the competitiveness of European firms. However, Europe will be
contributing to the reduction of global pollution.
We need to build an integrated not a single market – let us be clear about the objectives.
And the process will take a long time, may be more than 20 years. The EU is starting with
regional initiatives for integration and progressively will establish the conditions for an EUwide market. In the end, energy prices should reflect costs and we need to avoid multiple
charges throughout the transmission networks (peage multiples).
What is required for full liberalisation of energy markets in the EU? The European
Commission Energy Sector Inquiry concluded that insufficient unbundling, market
concentration and lack of interconnection are undermining the energy internal market
construction. Further intervention is needed to tackle these paramount issues. However,
MS are at very different stages of the process to build competitive markets that normally
involve privatization and liberalization, with structural concerns of competition. One of the
lessons that governments forget time and time again, and that has been very costly for
consumers, is that privatization should be undertaken in tandem with an appropriate
industry market structure. Trying to solve structural problems with antitrust is at best a
quixotesque exercise.
Commission Staff WP. Renewable Energy Roadmap, SEC(2006) 1720/2 using the PRIMES and GREEN-
X models.
Another study by IFIEC. Promotion of Renewable Energies in EU Member States. October 2003, also
calls attention for the decrease in the level of competitiveness of European industry due to the dramatic push
for renewables. E.g. in Germany electricity costs are supposed to increase 8.9 euros/MW in 2010.
European Energy Markets
Abel M. Mateus
In fact, privatization can contribute to an increase in efficiency; the problem is to pass
those gains to consumers. In 1997, Newbery and Pollitt99 have conducted a cost-benefit
analysis of the liberalization of the England & Wales electricity market done in 1990,
concluding that liberalization was beneficial for generation efficiency and for lower
emissions. Producers benefited from efficiency gains, through higher profits. However, at
the other side of the equation, consumers did not benefited from the cost savings in
electricity generation in the prices that were charged. And the main reason was that
competition was far from satisfactory.
This is the present stage of market liberalization in most EU countries. Liberalisation is
beneficial for efficiency and for fuel diversification, where natural gas is becoming an
important source of energy for power production which is also environmentally positive.
However, market power is precluding that this benefits are shared with consumers. More
importantly, market power, in the sense that a full functioning open access regime to
networks has not yet been put in place in most MS, is precluding further efficiency gains
and passing those efficiency gains to consumers.
The evidence for the UK is again revealing. In the late 1990s the government and
regulators, dissatisfied with energy prices, forced the two largest producers to divest part of
the generation capacity in England and Wales, leading to a decrease in the strategic HHI
from 0.306 to 0.189. An econometric study100 based on supply functions shows a shift in
those functions towards the marginal cost curves between 1998 and 2003, leading to a
decrease in electricity prices of about 20 percent, ceteris paribus.101
Let us now conclude our analysis, giving our answers to the questions raised.
Newbery, D., Pollitt, M. (1997).
Evans, Joanne and Richard Green. Why did British electricity prices fall after 1998? Memo. July 2005.
Surprisingly they found that the introduction of NETA did not have any significant effect on the short-
European Energy Markets
Abel M. Mateus
(1) Is ownership unbundling really necessary, and do the costs outweigh the benefits?102
We have shown that ownership unbundling is necessary but not sufficient. First we have
shown, using the study of ETSO, that there are numerous ways for an integrated
incumbent to discriminate against third party access. Opening up access to a network
requires a long process of regulation. Generally incumbents use long-term contracts,
priority of access, pricing strategies, including margin squeeze and all instruments to make
it difficult for entrants to have access and to compete in retail markets. Unbundling is
usually not enough and requires a detailed Network Code that regulates and controls
actions by the incumbent.
We have also seen that it is impossible to align investment and shareholders incentives of
the network operator and the generator/supplier. Moreover, the evidence shows that by
decoupling the network from the generators and suppliers the regulation is able to get
significant increases in efficiency from the TSO, resulting from specialization and a more
focussed management. Investment levels in the network have increased and quality levels
have also improved in the cases of unbundling – in the Portuguese and Italian cases were
even measures taken to solve blackout crisis. There are coordination problems and costs.
But these were low and solvable. The costs were low because any increase in transmission
costs are largely outweighed by the benefits that can be gained in increased competition in
generation – these are 4 to 5 times the first in global terms.103 In practice, coordination of
We had already defended most of the points expressed here in Portuguese Competition Authority,
Refocusing the vision - towards a European energy market, 2004. See also João Lopes e João E. Gata, A
Comparative Overview of the Progress Achieved to Date in the Construction of the EC Internal Energy
Market, Portuguese Competition Authority for the ECA Working Group, 2005
More evidence is provided by the UK case. North Ireland is a clear example. In 1996 prices were about
23% higher than in mainland, and the gap increased to 42% by 1998 because of the lack of competition in
generation and the protected long-term contracts with the generators, despite unbundling in 1993: NIE that
just contained transmission and distribution was sold to the public in 1993. Four power stations were sold to
European Energy Markets
Abel M. Mateus
investments have been solved by two mechanisms: (i) system planning that is a joint work
of the TSO, regulators and government; and (ii) a market based system of nodal or zonal
pricing that has proved to be an efficient scheme in cases that have been implemented. The
opinion expressed by the Commission, to accommodate the opposition of Germany and
France, is to accept an ISO. Our opinion is that this solution will hardly improve the
present situation of legal unbundling due to the arguments just referred.
2) Is there a need to increase interconnector capacity and to regulate their operation?
At the European level, the central role of cross-border trade in internal market
construction puts an emphasis on the future developments of interconnection capacity and
market design. Regional initiatives, like the Iberian market, Nordpool and the trilateral
market coupling in France, Belgium and Netherlands are certainly defining the path
towards the creation of the EU’s internal energy market. Regional integration, with more
interconnection capacity and more efficient rules regulating the use of interconnections, is
strategic to create an internal market. But this has to go hand in hand with more effective
unbundling. The evidence submitted in this paper proves that incumbents do not have
incentives to increase interconnector capacity.
Despite the doubts of the effectiveness of the ISO/TSO model, its implementation could
lead to Regional Transmission Organizations, covering several countries, like in the US.
Congestions costs can certainly be higher but improved coordination between TSOs would
be beneficial to trade, since trade opportunities from the existing transmission capacity
would be maximized. In fact, Europe has finally realized the importance of this measure
for security reasons. Following the November 4, 2006 blackout the following was decided:
(i) adoption of legally binding operational security rules by all network operators; (ii)
improvement of the cooperation between EU electricity grid operators, presently under the
three different companies in 1992. But there was no supply competition as the electricity is sold to the Power
Procurement Business of NIE under long-term contracts. (Newberry chapter 6, p. 204).
European Energy Markets
Abel M. Mateus
ETSO umbrella, which should be publicly accountable for their actions; and (iii)
development by the Commission of a framework for the electricity network as part of its
energy strategy.
Moreover, generation mixes among European countries are quite different – with at one
extreme Norway with its full reliance in hydro power and at the other extreme France,
where nuclear power is dominant. Following different incentives schemes to renewable
sources of energy, and wind in particular, with its unpredictability, there would be
additional reserve capacity required. To the extent that generation mixes will not fully
converge among EU countries, prices might not fully converge in the long run. However,
the mere fact that generations mixes are different underlines the welfare benefits of trade
among EU member states. In this sense, reinforcing interconnection capacity among
countries is strategic for the success of the internal energy market.
However, the amount of interconnection required is not an engineering problem. It is
above all a game played by the operators that are able to influence the flows in the
interconnector. That is why we need to construct an oligopoly model with strategic
behaviour. E.g. in the case of the interconnector Portugal-Spain, a capacity below 30
percent of Portuguese electricity demand was deemed insufficient for an approximate
competitive solution. A one size fits all rule in this case is not a solution for the EU-wide
market. The rules defined for regulation of interconnectors, based on an auction system
and market coupling need now to be effectively enforced.
(3) Is there a need to increase structural competition of electricity generators in national
The evidence shows that it is not enough to unbundle and create third party access with
consumer defranchising. Final energy prices will not decrease without creating enough
competition in the segment of generation of electricity. Suppose new suppliers come into
the market with a TSO and with consumer defranchising. What we have documented here
European Energy Markets
Abel M. Mateus
is that traditional incumbents are able through they long association with the network to
maintain a privileged relationship and continue to dominate the market. Besides, in all
network industries we have witnesses that antitrust rules, like cases against abuses of
dominant position, have been helpless to solve these access problems.
Once again the British case shows with clear evidence that it was only after structural
measures of divestiture and new entrants came into the market that consumer prices
started to decrease.
In this regard, an important policy to consider is to encourage electricity companies to start
trading gas and gas companies entering into electricity production. This policy doubles
immediately the number of competitors. Offering dual packages is a powerful competition
instrument, especially if there are other competitors in the market.
(4) Liberalization of electricity and gas markets have to go hand in hand
Electricity liberalization has been boosted by new entry using natural gas as the primary
source of energy and the combined cycle gas turbine technology. Entry opportunities were
created by the inefficiencies that prevailed in the generation mixes inherited from the
monopoly framework. However, after the first entry boom, many new entrants are facing
financial losses. Many argue that new investment is lagging, and ask about the sustainability
of the market framework, namely in what regards security of supply. Energy Regulators are
developing regulatory solutions, like capacity markets, to induce investment. Some argue
that “energy only” markets (e.g. with prices only for energy) do not efficiently guide
investment decisions – only price spikes periods provide clean incentives to invest. This
discussion clearly reflects the Newbery statement that “the central problem of energy
liberalization is the tension between the desire for efficient, competitive and unregulated
wholesale and retail markets, and long-term investment and security of supply”104.
Newbery, David. Regulatory Challenges to European Electricity Liberalisation. The Cambridge-MIT
Institute, WP 12. September 2002.
European Energy Markets
Abel M. Mateus
Closely related is the question of long-term contracts in gas and its duration. The
Commission has opened proceedings against Distrigas,105 EDF and Electrabel concerning
downstream gas contracts with industrial consumers and electricity generators for the first,
and contracts with industrial users of electricity in the rest, under article 11(6) of Regulation
1/2003 and article 2(1) of Regulation 773/2004. Distrigas106 has offered commitments,
under article 82, ensuring that 70 percent of gas supplied each year would be subject to
contestability. And no contract with industrial users and electricity generators would have a
duration over 5 years.107
(5) EU regulators should dedicate more attention to design the appropriate wholesale
market at national and EU-wide levels.
The English experience shows that marginal single-priced pools have advantages in
providing a reference price facilitating contracts and hence entry, and allowing scarcityresponsive capacity payments, although their transparency and repeated auction structure
facilitate collusion. Newbery claims rightly that problems of gaming and collusion fall as
the number of participants increase and the length of time for which bids must hold also
Efficient trade requires efficient pricing and allocation of transmission, best achieved by
nodal pricing along the PJM standard market design. The next best solution is market
coupling as practised in the Nordic market. Local power exchanges would then send their
aggregate bid-offer curves to an international clearing stage, which would allocate
transmission between countries along the synchronized auctions proposed by ETSO. Local
Memo 06/197 of May 2006.
IP/07/490 of April 11, 2007.
If these commitments were broken the Commission may impose a fine of 10 percent of total worldwide
turnover without the need to prove an infringement of antitrust rules.
European Energy Markets
Abel M. Mateus
power exchanges would then schedule the corresponding international flows and clear local
markets. Zones can be subdivided further if internal congestion levels are high.
(6) Do we need new merger policies and merger regulation?
Merger control, under the current merger wave that is leading to a significant restructuring
of the European energy sector, is critical to the success of the internal market. A coherent
approach should be put in place. First, criteria and methodology based on CRD and
simulation models need to be implemented due to the complexity of energy markets.
Second, the analysis should be tested with different geographical markets: national, regional
and EU-wide markets, and if it fails in any of those should be rejected. Third, the 2/3 rule
should be qualified for network industries, and ensure a consistent approach both at
Commission and National Authorities level.
It has also been widely proved that a policy of national energy champions, like all types of
national champions, lowers social and consumer welfare. Moreover, there is a serious levelplaying-field between mergers of firms controlled directed or indirectly by some MS and
private firms in other MS.108
We also proved that it is a mistake to create European energy champions, and that this
policy will be a major step-back in building a EU-wide competitive market.
(7) Should we create European energy champions in order to counteract the rising of
energy giants in supplier countries?
If it is wrong to build European energy champions from an internal perspective, it is even
worse from an external perspective. This is one of the main points this paper makes.
(8) Should there be a EU regulator?
Are we ready to adopt a decision that a public enterprise or enterprise controlled directly or indirectly
cannot acquire a private enterprise in another MS? This rule would solve the problems of sovereign
controlled funds and also third countries enterprises controlled by their respective states.
European Energy Markets
Abel M. Mateus
The Commission is proposing the creation of a EU-wide regulator which will have the
mandate to regulate interconnectors, and specially the interconnection between regional
markets, with a view to build a EU-wide integrated market. It could also take some of the
functions of the US FERC.
(9) What is the role of the Commission?
The Commission is currently working in the Third Energy Package. However, the road to a
political consensus is becoming very treacherous. Again, reaching such a consensus will
lead to a “minimum common denominator”, which would be far from even a second best
solution. But sometimes creating the foundations at the technical level can be even more
important. E.g. regional integration of power exchanges (PXs), agreements among TSOs to
integrate balancing markets to increase liquidity and enhance security, are effectively
contributing more to market integration than political consensus-building.
On the other hand, enforcement of antitrust rules could have an important signalling
effect. The Commission seems inclined to let each country choose its own way: TSO or
ISO, and then try to enforce in a more vigorous way articles 81 and 82 rules.109 But to go
even further it can also use article 86, and apply “structural remedies” given to the
Commission by Regulation 1/2003.
(10) Competition can be increased by introducing suppliers and traders and enabling
consumers to choose freely among suppliers. However defranchising consumers
presuppose prices determined by the market mechanism, and the governments should
refrain from regulating the final price. Switching mechanisms have to created and switching
costs reduced to a minimum.
Competition through market access puts pressure on consumer prices as the experience of
UK shows. In the UK where more than 50 percent of the consumers have switched
After the Sector Inquiry the Commission launched a series of unannounced inspections at the premises of
major energy companies in Germany on May 2006 and several antitrust cases have been initiated.
European Energy Markets
Abel M. Mateus
supplier, the data shows that incumbents with the highest prices have indeed experienced
the largest switch of consumers to competitors.
(11) Consumers should be charged the market price, hopefully in a competitive market, but
in real-time
The California crisis has clearly shown that politicians and regulators have to let the market
work.110 The recent supply crisis in the UK after 2004 and in contrast with the Italian
experience also prove this point. However, we are aware that European politicians with a
long tradition of government intervention are prone to fix final consumer prices creating
A number of famous economists has rightly put the problem about the California
Manifesto on the California Electricity Crisis. January 30, 2003.
European Energy Markets
Abel M. Mateus
large “consumer budget deficits”, instead of the traditional public sector budget deficits,
that sooner or later the users would have to pay painfully. Besides, the distortions have led
to market disruptions (when prices do not adjust then rationing has to come in) and
bankruptcies of traders making the liberalisation process more difficult.
(12) We need a comprehensive and sequence approach for energy reform.
A strategy based on establishing open access to networks and trade liberalization would
establish, at the minimum, the following comprehensive policy package for integration:
(i) Ownership unbundling between generation and transmission, in order to open access
for any generator or trader; (ii) Establish independent regulators in every country.
Tarification and regulation harmonization, in order to integrate the markets throughout the
EU; (iii) Establish appropriate regulation for interconnector management and congestion
management in all networks in order to enable trade throughout the networks; (iv)
Guarantee conditions for competition at retail level in all Member States; and
(v) Establish good Governance in the access by writing Network Codes for all Member
This is largely the position adopted by the UK government in its White Paper:
European Energy Markets
Abel M. Mateus
We differ in a fundamental way. We have demonstrated that all those measures are
necessary, but we also need (a) to adopt, at national and Union levels, structural measures
to create a more competitive market in the generation of electricity and in the import of
gas. Divestitures, VPPs, gas releases and other measures would need to be implemented;
(b) a stricter merger policy to buttress these structural remedies; and (c) to mandate
minimum capacity of interconnectors based on economic models of strategic behaviour.
And a final thought about coherence and timing. We have seen that due to the political
decision to increase dramatically the share of renewables in the energy supply, energy prices
are going to increase substantially in the EU. It would be difficult to explain to consumers
the benefits of competition when they see prices increasing. Antitrust policy once again is
only part of the picture: we need a comprehensive overall energy policy that increases social
welfare for the present and future generations.
They also advice the establishment of the Office of the Energy Observatory to collate and monitor data
on energy supply and demand across the EU, in the short and medium term, and identify the potential need
for future investment
European Energy Markets
Abel M. Mateus