September – December 2008 IP PLUS (2)
slaughter and may
IP Plus
Intellectual Property sector supplement
The scope of protection for well known
trade marks is clarified by the ECJ:
the Intel case
Overview
On 27 November 2008, the European
Court of Justice (“ECJ”) gave judgment
in Intel Corporation Inc. v CPM United
Kingdom Limited. The case concerned a
challenge brought by Intel Corporation,
Inc. (“Intel”), on the basis of its INTEL
trade mark, to the validity of a registration
for INTELMARK owned by CPM United
Kingdom Limited (“CPM”).
clarity as to when an infringement
claim is likely to succeed in relation to
“reputation” infringement. The case
establishes that:
The case gives crucial guidance on the
scope of protection available to well
known marks.
2. a “link” alone is not sufficient to establish
infringement – unfair advantage or
detriment must also be proven
Marks with a reputation have a greater
degree of protection than other marks
and an infringement action may be
available even in the absence of any
confusion resulting from the use of the
defendant’s sign.
The ECJ has previously held that
there must be a “link” between the
registered mark and the defendant’s sign.
Infringement also requires proof of some
unfair advantage being taken of the
mark or a detriment being caused to the
mark. In the Intel case, the court had an
opportunity to examine these elements
of the cause of action in detail.
The case may not have extended the
protection of well known marks as
far as some might have hoped but it
has provided a considerable degree of
1. where an infringing sign allegedly
“calls to mind” a mark with a
reputation, that is sufficient to
establish the required “link” between
the registered mark and the sign
3. to e s t a b l is h d e t r i m e nt it is
necessary to show a change in
economic behaviour among relevant
consumers
4. it may be necessary in establishing
unfair advantage to show a change
in economic behaviour but the Court
did not decide the issue.
5. it is not necessary to show an actual
change in behaviour, it is sufficient to
show that a change is likely but there
must be a serious likelihood that use
of the mark will cause consumers to
change their behaviour.
Reputation Based Infringment
INTELMARK was registered (and was to
be used) for marketing services. INTEL
was registered and used for computerrelated products, in relation to which
it had a “huge” reputation. It was
common ground between the parties
that marketing services are not similar
to computer-related products. As such,
Intel could object to the registration of
INTELMARK only on the basis of Section
5(3) of the Trade Marks Act 1994 (the
“Act”), which corresponds to Article
4(4)(a) of Council Directive 89/104/EEC
(the “Directive”).
Article 4(4)(a) of the Directive permits
the owner of a trade mark with a
reputation to prevent the registration
of a later mark if it is identical or similar
to the earlier mark and use of that later
mark would take unfair advantage of,
or be detrimental to, the distinctive
character or repute of the earlier mark.
Section 10(3) of the Act (corresponding
to Article 5(2) of the Directive) contains
an essentially identical provision in
relation to the infringement of a trade
mark with a reputation.
The ECJ’s judgment covers both
reputation -based invalidit y and
reputation-based infringement.
It has long been clear that reputationbased invalidity or infringement can
arise even when use of the later mark
does not cause confusion with the
IP Plus (2)
earlier mark. It does not matter that
the consumer is not led to believe
there is some connection or association
between the goods sold under the two
marks. What is required, however, is that
use of the later mark would cause the
consumer to establish a “link” between
the two marks.
The nature of that link, and whether
such a link is in itself sufficient to
establish reputation-based invalidity or
infringement, was at the heart of the
dispute in Intel.
The Court of Appeal
Before the Court of Appeal, Intel argued
that if any kind of mental association
was made between INTELMARK and
INTEL, that was enough to establish
the necessary link. Further, where a
trade mark is, like INTEL, both highly
distinctive and the beneficiary of a
huge reputation, Intel argued it must be
accepted that any third party use of an
identical or similar mark will at least be
detrimental to the distinctive character
of the earlier mark. The distinctiveness
of the INTEL mark would suffer a “death
by a thousand cuts” if Intel were not
permitted to prevent the use of every
later mark to which a link would be
made.
Jacob LJ was not persuaded by Intel’s
position, regarding it as advancing a
potentially oppressive conception of trade
mark law. However, since the law was not
acte clair, the Court of Appeal elected to
refer a series of questions to the ECJ.
The questions referred to the ECJ
were:
1) “For the purposes of Article 4(4)(a)
First Council Directive 89/104 of 21
December 1988, where:
(a) the earlier mark has a huge
reputation for certain specific
types of goods or services,
(b) those goods or services are
dissimilar or dissimilar to a
substantial degree to the goods
or services of the later mark,
(c) the earlier mark is unique
in respect of any goods or
services
(d) the earlier mark would be brought
to mind by the average consumer
when he or she encounters the
later mark used for the services
of the later mark,
are those facts sufficient in themselves
to establish (i) “a link” within the
meaning of paragraphs [29] and [30]
of Adidas-Saloman AG v Fitnessworld
Trading Ltd, Case C-408/01 [2003] ECR
I-2357 and/or (ii) unfair advantage
and/or detriment within the meaning
of that Article?
(2) If no, what factors is the national
cour t to take into account in
decide whether such is sufficient.
Specifically, in the global appreciation
to determine whether there is a
“link”, what significance is to be
attached to the goods or services in
the specification of the later mark?
(3) In the context of Article 4(4)(a), what
is required in order to satisfy the
condition of detriment to distinctive
character? Specifically, does
> Can establishing a link (whatever
that may require) ever suffice in
itself to show that unfair advantage
has been taken of, and/or detriment
caused to, the earlier mark?
Had the answer on both points been “yes”,
it would have afforded owners of wellknown trade marks a wide-ranging ability
to prevent the use of identical or similar
marks, irrespective of the goods or services
for which those marks might be used.
In the event, the ECJ’s answers were,
respectively, “yes” and “no”.
The nature of the “link”
The ECJ held that there will be a link
between the earlier mark and the later
mark if (1) use of the later mark creates
a likelihood of confusion or (2) the
earlier mark will be brought to mind by
exposure to the later mark.
Whether there is a likelihood of confusion
or a bringing to mind of the earlier mark
must be decided on the basis of a global
appreciation, taking all relevant factors
into account.
Relevant factors are, in all cases, likely
to include:
(i) the earlier mark have to be
unique
(i) the degree of similarity between the
two marks;
(ii) is a first conflicting use sufficient
to e s tablish detriment to
distinctive character and
(ii) the nature and (dis)similarity of the
goods or services sold under the two
marks;
(iii) does the element of detriment
to distinctive character of the
earlier mark require an effect on
the economic behaviour of the
consumer?”
(iii) the strength of the earlier mark’s
reputation; and
Those questions ultimately boil down to
two key inquiries:
Other things being equal, the more
similar the two marks are, the more
likely it is that exposure to the later mark
will prompt the consumer to call the
earlier mark to mind. However, this may
be offset by the dissimilarity between
the respective goods or services. If
there is no real overlap between the
> What is required to establish a “link”?
In particular, if the earlier mark is
merely brought to mind by the later
mark, is that in itself sufficient to
establish the necessary link?
(iv) the strength of the earlier mark’s
distinctive character.
IP Plus (2) 3
consumers of INTELMARK services and
the consumers of INTEL products, it will
be difficult to show that a link is likely
to be drawn, no matter how similar
the marks might be. Conversely, if the
goods sold under the two marks are very
similar, it is more likely that the later
mark will call the earlier mark to mind.
Similarly, the stronger the earlier mark’s
reputation, and the greater that mark’s
inherent or acquired distinctiveness,
the more likely it is that it will be called
to mind by a consumer when exposed
to the later mark. In some cases,
the earlier mark’s reputation may be
sufficiently strong to reach out beyond
those who actually consume the goods
and services sold under that mark. In
such cases, the reach of that reputation
can potentially overcome the lack of
any overlap between the two sets of
consumers and revive the possibility of
a link being established.
A link is not sufficient
The ECJ’s decision that a mere “calling to
mind” of the earlier mark is sufficient to
establish a link between that mark and
the later mark is a departure from the
prevailing view taken by the English courts
to date. The English courts have typically
demanded that a link be something
more than this. Notwithstanding the
need to make a global assessment in all
the circumstances of each case, the ECJ’s
approach means it will be significantly
easier for the owner of a well-known
trade mark to establish the necessary
link.
However, while it may have set a
relatively undemanding test for the
establishment of a link, the ECJ has
now clearly and unequivocally stated
that a link is never sufficient in itself to
establish unfair advantage or detriment
for the purposes of reputation-based
invalidity or infringement. Rather,
having established that the necessary
link is drawn between the two marks, a
further analysis is required to determine
whether use of the later mark does any
damage to the earlier mark.
In other words, the ECJ has reduced the
significance of a “link” between the marks;
what is important is whether injury to the
earlier mark has been caused.
reached the Court of Appeal, Intel’s case
had become focussed on allegations of
dilution and this focus was reflected in
the questions put to the ECJ.
As with the “link” analysis, this “injury”
analysis is to be conducted on the basis
of a global assessment, taking account of
all relevant factors. Many of the factors
relevant to the “injury” analysis will be
the same as those that were relevant to
the “link” analysis. But the question to
be answered at each stage is different:
Following the ECJ’s judgment in Intel, it
is clear that a trade mark cannot be said
to be diluted unless consumers of the
goods sold under that mark are caused
to change their economic behaviour as a
result of the use made of a later mark.
> at the first stage, the question is “will
the earlier mark be called to mind by
exposure to the later mark?”
> at the second stage, the question
is, “will use of the later mark cause
a relevant change in consumers’
economic behaviour?”
This seems to be a more logical balance.
The requirement of a “link” is not
specified in the legislation. Other case
law of the ECJ stresses that infringement
must have an effect on the “essential
function” of the mark. In the case of
reputation based infringement that is
done by proving the detriment or unfair
advantage. The concept of a “link” is
perhaps better confined to a “causal”
test – ie it is in effect a check that the
required injury is in fact caused by the
use of the defendant’s sign.
Dilution and the need for a change in
economic behaviour
In stating that establishing a link is not
sufficient to prove any relevant injury,
the ECJ was explicitly covering all
forms of injury covered by reputationbased invalidity and infringement
– namely, unfair advantage, detriment
to distinctive character (“dilution”) and
detriment to repute (“tarnishment”).
When it came to consider what else
might be required over and above the
link to prove a relevant injury, the ECJ’s
remarks were confined to cases of
dilution. The reason for this limitation
lies in the fact that by the time the case
The ECJ does not suggest what sort of
economic behavioural change would
be required to manifest dilution, but it
has stated that economic benefit to the
user of the later mark is irrelevant for
these purposes. As such, it would seem
there must be an economic behavioural
change that adversely affects the earlier
mark. Presumably this must ultimately
involve a reduced willingness on the
part of consumers to purchase goods
or services sold under the earlier mark.
and/or consumers’ views of the mark,
goods or services declining.
It might be said that by the time
dilution is made manifest in changes in
consumers’ economic behaviour, it is too
late. The damage has been done and
the previously undiluted distinctiveness
of the mark cannot readily be restored.
This is important since the risk is often
that someone entering the market using
the same or a similar mark may cause
little or no damage at first and this will
increase slowly but that this will be
exacerbated as other infringers follow.
Fortunately, the ECJ accepts that the
owner of a mark with a reputation
should not be compelled to wait until
the harm is done before taking action.
However, the ECJ insists that, as a
minimum, the trade mark owner must
be able to adduce evidence of a serious
risk of injury. This means there must be
evidence of a serious likelihood that use
of the later mark will cause consumers to
change their economic behaviour.
The ECJ does not attempt to quantify
what makes a risk or likelihood “serious”.
4 IP Plus (2)
Nevertheless, the ECJ evidently intends
to require that the prospect of harm
is “real and tangible” (to use Jacob
LJ’s words). Something more than a
mere likelihood or risk of economic
behavioural change must be proven
before a trade mark owner can make out
a case for reputation-based invalidity or
infringement on the basis of dilution.
Tarnishment and unfair advantage
What of tarnishment (detriment to
repute) and unfair advantage?
It is certain that a link alone is no more
sufficient to prove either of these forms of
injury than it is sufficient to prove dilution.
It is also certain that the trade mark
owner must prove at least a serious risk of
tarnishment or unfair advantage if he is to
make out his case. However, the ECJ has
not specified that either of these forms of
injury requires proof of any change in the
economic behaviour of consumers. As
such, it is conceivable that these forms
of injury can be demonstrated without
showing that there is a serious likelihood
of any such change.
There is no obvious logical basis on
which to hold that dilution alone
demands evidence of such change. If
dilution must ultimately be manifested
in a reduced willingness of consumers
to purchase goods sold under the
earlier mark, the same may be true of
tarnishment. If economic benefit to the
user of the later mark is irrelevant in
cases of dilution, again the same must
be true of tarnishment.
As far as unfair advantage is concerned,
the situation must be reversed. It would
appear that what would matter here is
that use of the later mark changes the
economic behaviour of consumers of
the goods sold under that later mark
(and not consumers of goods sold under
the earlier mark). The change would
need to benefit of the user of the later
mark and any behavioural change that
is adverse to the earlier mark would be
irrelevant.
The approach of the English Courts
Prior to Intel, the English courts already
considered an impact on the economic
behaviour of consumers to be the acid test
for at least some forms of injury covered
by reputation-based invalidity and
infringement. (They typically considered
this to be a required characteristic of the
necessary “link”, which following Intel
must be wrong, but the same question of
an impact on economic behaviour caused
by use of the later mark still arises.)
While some doubt has been expressed
by English courts as to whether an
economic behavioural impact is always
required, the prevailing view has been
that unfair advantage cases are those
in which an economic impact is most
obviously required.
When the Intel case was before the High
Court, Patten J was confident that unfair
advantage “requires some evidence that
some real economic advantage has been
or will be obtained” by the user of the
later mark. In contrast, he considered the
position in cases of detriment – whether
in the form of dilution or tarnishment – to
be “more complicated”.
The ECJ has now confirmed that, at least
in cases of dilution, a change in economic
behaviour is the defining characteristic
of that form of injury. Nothing the ECJ
has said suggests that the English courts
are wrong to have considered an impact
on economic behaviour to be at the
heart of unfair advantage.
Conclusion
Concerns have been expressed that
proving reputation-based invalidity
or infringement will be a considerable
challenge in all but the clearest cases
and it could be argued that this was
the ECJ’s intention. The insistence on
proving a change in economic behaviour,
and the refusal to allow that relief should
be granted for anything less than a
“serious” risk of injury, suggests that the
ECJ was concerned to confine the reach
of these causes of action.
However, it remains to be seen how the
courts will apply this requirement in
practice. It is likely that for the “right”
mark in the “right” circumstances
(e.g. where the parties operate in
connected sectors), the owner of a
trade mark with reputation will still be
able to make out a case under Article
4(4)/5(3) and the courts will apply a
pragmatic approach to proving a change
in economic behaviour.
The ECJ’s underlying message is
that reputation-based invalidity and
infringement represent the outer reaches
of trade mark law. They permit a trade
mark owner to prevent the use of
similar marks in circumstances that
would not be touched by “mainstream”,
confusion-based trade mark law, or even
by passing off.
However, there remains potential in
the right case for the owner of a well
established mark to take action to
stop dilution and unfair advantage at
an early stage and prevent it losing
distinctiveness. The ECJ has left the
national courts with considerable
discretion to decide which facts will
support this greater protection and it
will be interesting to see how they use
that discretion.
Intel Corporation Inc. v CPM United
Kingdom Limited, Case C-252/07 (27
November 2008). Questions referred by
the Court of Appeal [2007] EWCA Civ 431
(15 May 2007), on appeal from the High
Court [2006] EWHC 1878 (Ch) (26 July
2007), on appeal from the Intellectual
Property Office O/037/06 (1 February
2006).
For further information on any Intellectual
Property related matter, please contact
your usual Slaughter and May contact, or
any of the following:
London
Susie Middlemiss
+44 (0)20 7090 4038
Cathy Connolly
+44 (0)20 7090 4020
Robert Sumroy
+44 (0)20 7090 4032
Published to provide general information and not as legal advice © Slaughter and May, 2008
One Bunhill Row, London EC1Y 8YY, United Kingdom T +44 (0)20 7600 1200 F +44 (0)20 7090 5000 http://www.slaughterandmay.com
Brussels
John Boyce
+32 (0)2 737 94 00
Hong Kong
Neil Hyman
+852 2521 0551