Case Notes
| Published date | 01 December 2009 |
| Date | 01 December 2009 |
| DOI | 10.1177/1023263X0901600408 |
| Subject Matter | Case Notes |
16 MJ 4 (2009) 507
CASE NOTES
Services of Genera l Economic Interest and the state aid paradox in BUPA
Case T-289/03 (12th February 20 08)
§1. INTRODUCTION
In recent years, Services of Genera l Economic Interest (SGEI)1 have gained import ance
in elds such as transport or postal services as they open a valuable possibility to derogate
from the stric t dictate of EC internal market and competition law. Yet there are two
crucial problems wit h this concept. In the rst place, neither leg islation nor case law
currently prov ide a clear-cut den ition.2 Secondly, an academ ic and judicial battleeld
exists as to the limits wit hin which such serv ices can actual ly be nanced by t he state.3
It is problematic that governments, and especially loc al municipalities , have to decide
which s ervice they can bri ng under t his concept. is also eects how much fundi ng
they c an give to a market operator for rendering services . If a mistake is made, a nd a
competitor brings the case to court, the ma rket operator will have to repay the money.
is ca n cause nanc ial problems for compan ies, both duri ng and beyond t he current
economic crisis.4
In the 20 03 Altmark judgment, the European Court of Just ice (ECJ) developed a
simple four step test under Article 107(1) TFEU (ex Art icle 87(1) TEC).5 If it is f ullled,
1 Services of Gene ral Economic Interest are, in gener al terms, those serv ices which are considered by the
citizens of a Member State to be economical ly in the general interest. Exa mples are transportati on lines
in remote are as. ese may not be economically v iable but they are provided since they are desired by
citizens. O ther prominent examples i nclude telecommunicat ions, electricit y, gas and postal serv ices.
2 See for ex ample Case T-289/03 Briti sh United Provi dent Associat ion Ltd (BUPA), BUPA Insurance Ltd
and BUPA Ireland Ltd v. Commissio n of the European Communitie s [2008] ECR II-00081, para. 165.
3 Outside of thes e limit s, fu nding con stitutes state ai d and therefore is not allowed. See for exa mple
the dierenc es between the Ferring judgment C-53/00 Ferring SA v. Agenc e centrale des or ganismes de
sécurité soc iale (ACOSS) [2001] ECR I-9067 and the Altmark judgement Cas e C-280/00 Altmark Trans
GmbH v. Nahverkehrsgese llscha Altmark GmbH [2003] ECR I-7747.
4 Only if a measure constit utes state aid does it have to be notied to the Commission. On avera ge it then
takes 5,3 months for the Comm ission to take a de cision. If a measu re is not notied a nd is later found
to be unlawf ul (be cause it is st ate aid), it has to be recovered. European Commission, Vademecum
Community rules on state aid, elec tronically available at: http: //ec.europa.eu/competition/state _aid/
studies_ reports/studies _reports.cfm ( last visited 10.01.10), 13.
5 Case C-280/00 Altmark. See p aragraph 95 for the test .
Case Notes
508 16 MJ 4 (2009)
the funding of an SGEI does not constit ute state aid.6 e question rema ins whether
this test should also be applied if a Commission decision pre-dates the judgment and
if a compensation mechan ism does not correspond to the Altmark criter ia. Moreover,
how should the Court ac t if its judgment has the potential to st rike down the legislation
enacting the health insurance system of a Member State and, even worse, if that judgment
concerns a state (Ireland), which at that ti me was decisive to the success of the Li sbon
Treaty?
In the BUPA (British United Provident As sociation) judgment,7 the General Cour t
(GC, formerly Court of First Instance) sought to face the challenge by scruti nizing a
Commission decision on the Irish health insurance system.8 Perhaps due to its leng th
the judgment has gained l ittle attention. Yet its signicance is not limited to the fact
that the Cour t stepped into the role of the C ommission and argued as t hough the latter
had used t he Altmark test, e ven though the Commission did not bring forward any
arguments in this rega rd.9 It is also the rst t ime that European Courts have prov ided
an excellent insight i nto the criteria which determine an SGE I. Equally important is t he
fact that t he GC demonstrated that, in its view, the Altmark test is not a straight jacket.
Unfortunately, the Altmark ‘delux’ test which it developed blurs the dist inction between
Article 107(1) TFEU and Article 106(2) TFEU (ex Article 86(2) TEC) and brings elements
of the Article 106(2) TFEU test into Article 107(1) TFEU.
§2. THE RISK EQUALIZATION SCHEME RES IN THE IRISH
HEALTH INSURANCE
Contrary to a situation such as the one in Altmark, the R isk Equalization Scheme is not
a mecha nism in w hich the s tate compensates t he direct costs incurred by the insurer,
which would be the case, for example, i f an insurant were to submit a bill for a surger y.
e Scheme provides for a risk por tfolio for every insurer. In a nutshell, th is means that
an average for all insu rers on t he market is ca lculated based on t he risk of insurants
becoming ill a nd the claim costs submitted by t hem during a reference period. If an
insurer’s own portfolio lies above this average, he will re ceive compensation from an
insurer whose por tfolio is below the average. us t he latter insures people t hat are less
oen ill (statistically) and therefore it incurs fewer costs. In consequence, the compensation
6 e test e ssentially en sures that a company does not receive more fu nding than the cost of providing
the SGEI, and th at it works eciently.
7 Case T-289/03 British United Provident Assoc iation Ltd (BUPA), BUPA Insurance Ltd and BUPA Ireland
Ltd v. Commission of the Eu ropean Communities [2008] ECR I I-00081.
8 Commission Decision N46/20 03 – Irela nd (Ri sk equ alization scheme in the Irish health insurance
market).
9 e Com mission had based its decision on the Ferring test. By the time t he GC was ca lled upon to
decide on B UPA, the ECJ had already deve loped the Altmark test . e GC dec ided apply the Altmark
test and therefore ha d to act as though the C ommission had used th is test.
To continue reading
Request your trial