European carbon prices fundamentals in 2005-2007:
the effects of energy markets, temperatures
and sectorial production
Université Paris X-Nanterre
Maison Max Weber (bâtiments K et G)
200, Avenue de la République
92001 NANTERRE CEDEX
Tél et Fax : 33.(0)1.40.97.59.07
Email : secretariat-economix@u-paris10.fr
Document de Travail
Working Paper
2007-33
Émilie ALBEROLA
Julien CHEVALLIER
Benoît CHÈZE
E
c
o
n
o
m
i
X
Université Paris X Nanterre
http://economix.u-paris10.fr/
UMR 7166 CNRS
European Carbon Price Fundamentals in
2005-2007: the Effects of Energy M arkets,
Temperatures and Sectorial Production
Emilie Alberola
∗
, Julien Chevallier
†
and Benoît Chèze
‡
October 17, 2007
Abstract
- This article aims at characterizing the daily price fundamentals of
Europ ean Union Allowances (EUAs) traded since 2005 as part of the
Emissions Trading Scheme (ETS). First, the presence of two structural
changes on April, 2006 following the disclosure of 2005 verified emissions
and on Octob er, 2006 following the European Commission announcement
of stricter Phase II allocation allow to isolate distinct fundamentals evolv-
ing overtime. The results extend previous literature by showing that spot
prices react not only to other energy markets and temperatures, but also
to economic activity within the main sectors covered by the EU ETS such
as proxied by sectoral production indices. Besides, the sub-period decom-
p osi tion of the pilot phase gives a better grasp of institutional and market
events that drive allowance price changes.
JEL Codes: Q40, Q48 Q54
Keywor ds: Carbon Emissions Trading, Market Price Fundamentals,
EU ETS.
∗
EconomiX-CNRS, University of Paris 10, ADEME (the French Government Agency for
Environmental and Energy Management) and the Caisse des Dépôts–Climate Task Force,
Paris, France. emilie.alberola@caissedesdepots.fr
†
Corresponding author. EconomiX-CNRS, University of Paris 10, Department of Eco-
nomics, Office G-507a, 200 avenue de la République, 92001 Nanterre Cedex, France. Tel: +33
1 40 97 59 36; fax: +33 1 40 97 77 84; jchevall@u-paris10.fr
‡
EconomiX-CNRS, University of Paris 10 and ADEME, Paris, France. benoit.cheze@u-
paris10.fr
1
1 Introduction
Since January 1, 2005 each carbon ton emitted in Europe by about 11,500
energy intensive plants has been priced. The European Union Emissions Trading
Scheme (EU ETS), which covers up to 46% of European CO
2
emissions, aims
at helping Member States to achieve compliance with their commitments under
the Kyoto Protocol during 2008-2012. While International Emissions Trading
(IET) allows trading between governments starting in 2008, the EU ETS breaks
down emissions trading to the company level. Its main objective consists in
giving incentives to ind ustrials to reduce emissions and to contribute to the
promotion of low carbon technologies and energy efficiency among CO
2
emitting
plants. Most important polluting entities manage their compliance between
their allocation and annual verified emissions by buying or selling European
Union Allowances (EUAs) to emit a ton of carbon. At the end of the first
commitment period on December 31, 2007, the European Commission (EC)
intends to provide decision makers with an allowance price to lead to emissions
abatements.
Yet the first disclosure of 2005 verified emissions on April, 2006 revealing
the net short/long position
1
of each plant was accompanied by a sudden al-
lowance price collapse, and tends towards zero thereafter (see Figure 1). This
price path therefore suggests that trading was based on heterogenous anticipa-
tions prior to information disclosure. Within 2005-2007, different fundamentals
seem to co-exist before and after the compliance break. Thus, understanding
price formation mechanisms when creating such a market appears of critical
importance. In this context, the question we address is the following: which
factors contribute to shape the price formation of this newly European Union
Allowances market?
This article analyses the EU ETS during its pilot p hase (2005-2007) by
focusing on the empirical relationship between CO
2
price changes
2
and its main
fundamentals. S pringer (2003)’s review of theoretical models and Christiansen
et al. (2005) lead to the identification of the carbon prices main drivers being
economic growth, energy prices, weather condition and policy issues. Their
potential impacts are analysed in this paper.
The total number of allowances is determined by Member States negotiating
with industrials and after validation by the EC. As soon as the first National
Allocation Plans
3
(NAPs) were drafted, there was a concern of allowance over-
supply during the EU ETS pilot phase. Academic and market agents usually
agree that the information revelation by simultaneous countries of lower than
expected 2005 verified emissions is the main reason behind the fall of CO
2
prices
by more than 50% that occurred on April, 2006.
2
As pointed out by Ellerman and Buchner (2007), this allowan ces oversupply
argument shall be balanced by the analysis of net short/long positions at the
installation level. To the best of our knowledge, empirical studies have not
yet studied the effects of sectoral economic activity covered by the EU ETS
on CO
2
prices. As industrials are able to influence the market price by their
abatement decisions, we intend to include their potential effects in our analysis.
The inclusion of the sectors production variables is all the more important as the
April, 2006 break in the EUA spot price may be explained by wrong projections
at the sectoral level (Grubb and Ferrario, 2006).
Compared to previous literature, our contribution is threefold. First, we
show statistical evidence of structural changes following the d isclosure of new
information about the net short/long position at the installation level during
Phase I and the EC decision to enforce stricter NAPs during Phase II. Second,
the role played by sectoral ec onomic activity in the EU ETS is highlighted. By
doing so, this article extends, among other contributions, Mansanet-Bataller
et al. (2007) by emphasising other determinants of carbon prices than energy
prices and climatic events. Third, we find that those fundamentals vary between
periods, and that EUA spot prices react to energy and weather variables d uring
some time periods whereas during other p eriods , institutional decisions seem to
have more influence than the expected drivers. This evidence leads us to the
conclusion that allowance prices react to distinct fundamentals within this first
commitment period.
The remainder of the paper is organized as follows. Section 2 reviews the
main drivers of EUA prices. Sec tion 3 estimates the relationship between the
daily carbon price changes and energy commodities, meteorological factors and
industrial production. Section 4 presents the results. Section 5 concludes.
2 Main drivers of EUA prices
New commodity markets generally need time to achieve real price discovery. As
shown in Figure 1, the EUA price pattern experienced a strong volatility during
the first two years. Beginning at 8
=
C on January 1, 2005 EUA prices increased
to around 30
=
C on July 2005, fluc tuated during the following six months in
the range of 20-25
=
C, then rose to 30
=
C until the end of April. On the last
week of April, 2006, prices collapsed when operators disclosed 2005 verified
emissions data and showed the scheme was oversupplied. After this considerable
adjustment by 54% in four days, EUA prices moved in the range f rom 15 to
20
=
C until October, 2006. From this date, the EU ETS is sending two price
signals responding to different dynamics. Phase I prices are declining towards
zero whereas Phase II prices are increasing to 20
=
C primarily due to the EC
3
which has reaffirmed its will to enforce tighter targets. On April, 2007, verified
emissions were again below the 2006 yearly allocation. The EUA spot price
seems to react to this new information by moving towards zero. Phase I EUA
futures and spot prices are strong correlated whereas EUA Futures prices for
delivery in the Phase II are totally disconnected since October, 2006.
Figure 1: EUA spot prices from July 1, 2005 to April 30, 2007
Source: Powernext carbon
While allowance supply is fixed by each Member States through NAPs, al-
lowance demand is function of the level of CO
2
emissions whose production
depends on a large number of factors such as fuel (brent, coal and natural gas)
and power (electricity) prices, weather conditions (temperatures, rainfall and
wind spee d) and economic growth. Until now, the empirical literature focused
only on the first two factors.
According to previous literature, energy prices are the most important drivers
of carbon prices due to the ability of power generators to switch between their
fuel inputs (Kanen, 2006; Christiansen et al., 2005; Bunn and Fezzi, 2007; Con -
very and Redmon d, 2007). This option to switch from natural gas to coal in
their inputs represents an abatement opportunity to reduce CO
2
emissions in
the short term. High (low) energy prices contribute to an increase (decrease)
of carbon prices. This logic is described by Kanen (2006) who identifies brent
prices as the main driver of natural gas prices which, in turn, affect power prices
and ultimately carbon prices. Power operators also pay close attention to dark
and spark spreads and the difference b etween them. The dark spread is the
theoretical profit that a coal-fired power plant makes from selling a unit of elec-
tricity having purchased the fuel required to produce that unit of electricity. The
4
