I
have now had a chance to read Lenovo Group Ltd. v. Telefonaktiebolaget LM
Ericsson (Publ), [2025] EWCA Civ 182, opinion by Lord Justice Arnold joined
by Lady Justice Falk and Lord Justice Newey, concluding that a willing licensor
in the position of Ericsson would enter into an interim license with Lenovo, on
terms halfway between the parties’ last offers, pending a judicially-determined
global FRAND license. Below is a short
summary of the decision, and then my initial thoughts.
The
facts, in brief, are as follows. Both
Ericsson and Lenovo own portfolios of FRAND-committed SEPs, and have spent many
years negotiating a cross-license, without success. (There is a 2014 license in place, however,
between Motorola, which in 2014 became a Lenovo subsidiary, and Ericsson. That license covers “certain Ericsson SEPs
declared essential to the 4G standard and certain Motorola devices” (para. 49),
and will remain in effect until all of the relevant patents expire. It also states that disputes “related to the
construction and interpretation of” the license shall be settled in London
(para. 51). Lenovo’s own SEPs, of
course, are distinct from the Motorola ones, but the existence of the Motorola
license will affect the eventual terms of any cross-license.) In 2023, Ericsson filed an action against Lenovo
in the United States District Court of the Eastern District of North Carolina, alleging
the infringement of four SEPs declared essential to the ETSI 5G standard, and requesting
a declaration that Ericsson had complied with its FRAND obligation or, if not,
that the court “determine a FRAND rate for a global cross-license between
Ericsson and Lenovo” (para. 58). Note,
however, that “because of the way in which Ericsson had framed their claim for
relief, there was a possibility that the EDNC would not ultimately determine a
FRAND rate for a global cross-licence between Ericsson and Lenovo. . . . [T]his
possibility could be avoided if Ericsson were to amend their claim to seek a determination
of what rate is FRAND, but Ericsson declined to make the necessary amendment.
That remains Ericsson’s position” (para. 60). Ericsson
also filed an action relating to five non-SEPs, and later an action relating to
certain SEPs declared essential to the ITU’s HEVC/H.265 standard (id.). Those cases remain pending. Also in in 2023, Ericsson initiated three ITC
investigations alleging the infringement of the patents at issue in the EDNC
proceedings. In September 2024, the U.S.
Office for Unfair Import Investigations (OUII) filed a brief in the first of
the ITC investigations, stating its conclusion that an exclusion order would
not raise public interest concerns or have adverse effects on competitive
conditions in the U.S., and that “Ericsson has and is engaged in negotiations
in good faith” (para. 66). In December
2024, the ALJ issued an Initial Final Determination in Ericsson’s favor, but the
ALJ’s decision whether to enter an exclusion order remains pending until April. Ericsson also has filed actions in Brazil and
Colombia, obtaining a preliminary injunction in Brazil (now stayed pending
appeal, see para. 75), and some in Colombia as well, though some of these have
been lifted (para. 76). For its part, Lenovo
has initiated two proceedings in the ITC targeting Ericsson, but these remain
pending are “are less far advanced than the proceedings commenced by Ericsson”
(para. 68). Lenovo has counterclaimed in
the EDNC “for a determination of FRAND terms” and requested an ASI relating to the
Brazilian and Colombian actions. The district
court denied the ASI, but the Federal Circuit remanded for further
consideration in October 2024 (para. 77; see also the relevant post on this
blog). Also, Lenovo in 2023 and 2024 filed
actions against Ericsson in the Patents Court for England and Wales, alleging inter
alia the infringement of some of its patents and requesting the court to
establish the terms of a global FRAND license (para. 69). Most relevant here, Lenovo also requested “a
declaration that willing parties in the position of Lenovo and Ericsson would
agree to a short-term licensing regime pending the determination of FRAND terms
for a final cross-license, and a declaration as to the terms of” such a license
(para. 70). (Lenovo’s subsidiary
Motorola also has commenced two actions before the UPC, see this article on
JUVE Patent.) Meanwhile, in October 2024
the EWCA held in Panasonic v. Xiaomi that Xiaomi was entitled to a
similar declaration, though the facts of that case were somewhat different in
that both parties had agreed to abide by the English court’s determination of a
global FRAND license, but SEP owner Panasonic appeared to renege on the
agreement by pursuing injunctive relief in Germany. (See here.) In any event, the Patents Court in the
present case denied Lenovo’s request for a declaration in January 2025, but
last Friday the EWCA allowed the appeal.
Key
to the EWCA’s decision is the following, regarding why it believes Ericsson wants
to obtain an exclusion order from the ITC and does not want to accede to an
interim FRAND license pending a global determination in the U.K.:
. . . Ericsson want to achieve a
better outcome than a determination by the English courts would provide . . . .
To put it at its lowest, Ericsson must perceive that there is at least a
material risk that the English courts will determine that Ericsson’s October
2023 Offer was not FRAND, whether on its own terms or with reference to the
correct interpretation of the [Motorola] Licence, and that FRAND terms for the
cross-licence are closer to those offered by Lenovo, and Ericsson must be
determined to avoid that risk. In my judgment this is indeed the true explanation
for Ericsson’s conduct (para. 124).
The
court then concludes that Ericsson’s conduct, while “not as egregious as that
of Panasonic in Panasonic v. Xiaomi,” is “to coerce Lenovo into
accepting terms more favourable to Ericsson than the English courts would determine
to be FRAND, or at the very least to avoid the risk that the English courts
would determine that FRAND terms are less advantageous to Ericsson than those
sought by Ericsson” (para. 128). Thus,
while Ericsson may simply be “exercising the legal rights which are available
to them in another jurisdiction,” it does not follow that such conduct “cannot
be contrary to their obligation of good faith.”
Specifically:
. . . the whole point of a SEP owner’s obligation
to ETSI is that it is a derogation from the patentee’s normal entitlement to
enforce its patent by means of an injunction . . . . Secondly, the purpose and effect
of an obligation of good faith is to act as a constraint upon a party’s ability
to enforce its strict legal rights solely with regard to its own interests.
Ironically, this is the very reason that English contract law, unlike French
contract law, has historically been reluctant to embrace obligations of good
faith save in limited circumstances. To put it bluntly, Ericsson’s position
amounts to saying that they are entitled to use their raw legal power to compel
Lenovo to submit. That might well have been a legitimate response to a long
period of hold out by Lenovo, but as explained above Lenovo are no longer
holding out even if they were previously. On the contrary, Lenovo have now
accepted that they must pay Ericsson whatever an independent and impartial court
determines to be FRAND plus interest. In those circumstances coercion by Ericsson
is no longer justified. Accordingly, Ericsson are in breach of their obligation
of good faith (para. 129).
The EWCA
therefore rejects the Patents Court’s conclusion that, in view of the OUII
brief’s statement that Ericsson’s October 2023 Offer “was comfortably within
the FRAND range,” the judge could not be “satisfied to a high degree of
assurance that, in exerting pressure on Lenovo to accept Ericsson’s October
2023 Offer, Ericsson were necessarily seeking supra-FRAND rates” (para. 130). In addition, the EWCA concludes that an
interim license (payments under which can “be adjusted to the extent necessary
in consequence of the determination of the terms of the final licence,” (para.
136)) “would serve a useful purpose in forcing Ericsson to reconsider its
position,” to “see the error of their ways” (para.142). Finally, in addressing comity, the EWCA
agrees with Lenovo that “[i]f the declaration does induce Ericsson to
reconsider their position and to grant Lenovo an interim licence, that would
promote comity because it would relieve the courts and tribunals of the USA,
Brazil and Colombia of a great deal of burdensome and wasteful litigation. If,
on the other hand, Ericsson decide to ignore the declaration and to pursue the
proceedings in the USA, Brazil and Colombia, it will be entirely for those
courts and tribunals to make their own assessment of the parties’ conduct,
including their conduct in the English proceedings, and to decide what, if any,
relief to grant Ericsson for any infringements they may find established in the
absence of a licence. Thus, a
declaration would not be contrary to comity” (para. 149). The court also states that, although the fact
that “the courts and tribunals of the USA . . . were first seised of the
dispute between the parties” “is an important consideration,” the English court
also is “properly seised of the FRAND dispute,” and has “exclusive jurisdiction
to determine the issues concerning the [Motorola] Licence, which have a
significant impact on what terms for the cross-licence are FRAND. In those
circumstances, the legal centre of gravity, although not the commercial centre
of gravity, of the overall dispute is in England” (para. 152). Moreover:
In the absence of a global dispute
mechanism for determining FRAND disputes, or an ad hoc agreement to
arbitration, the possibility of jurisdictional conflict is inescapable. Leaving
aside Lenovo’s point about the exclusive jurisdiction clause in the [Motorola]
Licence, the principled answer to this might be that the court first seised should
determine what terms are FRAND. In the present case, however, it is plain that Ericsson
do not want the EDNC to determine FRAND terms for the cross-licence any more
than they want the English courts to do so. If Ericsson wanted the EDNC to determine
FRAND terms in preference to the English courts, they would have made the
simple amendment to their claim in the EDNC I Proceedings which Richards J identified
as being appropriate in his judgment on the Jurisdiction Application as long ago
as 18 April 2024 and would have undertaken to accept the EDNC’s determination as
to FRAND terms. Ericsson have not done so. By contrast, Lenovo have offered to accept
the EDNC’s determination as to FRAND terms if Ericsson drop their campaign to
obtain injunctions and equivalent relief, but Ericsson have not agreed to this.
On the contrary, Ericsson have vigorously pursued such relief, in particular in
the ITC. This demonstrates that Ericsson’s stance is not driven by
jurisdictional preference with respect to FRAND determination. It is driven by
a preference for the exclusionary power of a national injunction (or equivalent
relief) over FRAND determination by any court. This is hold up.
The second argument is that making
the declaration sought by Lenovo would promote forum shopping. . . . . [F]orum shopping
by both SEP owners and implementers is equally to be deprecated. Regrettably,
however, the potential for forum shopping is an inevitable feature of the present
ETSI IPR Policy. . . .
In conclusion, I entirely accept
that, as counsel for Ericsson submitted, jurisdictional imperialism is to be
eschewed. As I have explained, however, it is common ground in this case that a
FRAND cross-licence would be global. [Unwired Planet] establishes that,
in such a case, the English courts have jurisdiction to determine what terms
are FRAND on a global basis. A critic might argue that, to that extent, a
degree of jurisdictional imperialism is already hard-wired into the English
courts’ approach to these issues. The declaration sought by Lenovo is less
intrusive into the jurisdictions of foreign courts and tribunals than a global
FRAND determination (paras. 153-55).
The
court concludes:
For the reasons given above I
conclude that: (1) Ericsson are in breach of their obligation of good faith
under clause 6.1 of the ETSI IPR Policy by pursuing claims for injunctions and
equivalent remedies in foreign courts and tribunals despite Lenovo having
undertaken to enter into a licence on the terms determined by the Patents Court
to be FRAND (subject to adjustment on any appeal) and having offered to submit
to determination of FRAND terms by the EDNC; (2) a willing licensor in the
position of Ericsson would enter into an interim licence with Lenovo pending
that determination, and FRAND terms for that licence would be those set out in
the preceding paragraph; (3) making the declaration sought by Lenovo would
serve a useful purpose; and (4) the declaration should not be refused on the
grounds of comity. I would therefore allow the appeal (para. 157).
So
what are the implications of the decision?
I’m not altogether sure, but am trying to think it through, so here are
my initial thoughts.
First,
if Ericsson does not reconsider its position, and agree to an interim licence on
the terms stated in the decision pending the Patents Court’s determination of a
global FRAND cross-license, what happens?
I think that would mean that Lenovo could argue to the Patents Court that
Ericsson is an unwilling licensee of Lenovo’s SEPs, and therefore can be
enjoined from practicing those patents in the U.K. That would give Lenovo some leverage to
compel a global settlement. As for
whether it would affect any of the litigation occurring anywhere else, I don’t
know. If courts in other jurisdictions
recognize the EWCA judgment, and consider themselves bound by its finding that Ericsson
is an unwilling licensee, again that would increase Lenovo’s leverage to enjoin
Ericsson and thus compel a settlement.
But will other countries recognize the judgment, or consider it to be
against public policy insofar as it is intended to prevent Ericsson from
pursuing the litigation it is otherwise permitted to pursue under the laws of
the U.S., Colombia, Brazil, and the UPC?
Second,
if Ericsson does reconsider and agrees to an interim licence, then Lenovo (and
Ericsson) are both licensed on an interim basis, and there are no longer any grounds
for injunctive relief or exclusion orders anywhere else; the cases elsewhere
would be moot (though perhaps the parties could agree to let the EDNC decide
the terms of the global FRAND license, as opposed to the Patents Court—assuming
that a U.S. court has jurisdiction to do that with consent of the parties, an
issue I’m not 100% sure about myself, however, after TCL v. Ericsson, as
I indicated in my article Is Global FRAND Litigation Spinning Out of Control,
2021 Patently-O L.J. 1, at footnote 10).
Third,
I agree for the most part with Lord Justice Arnold’s statement in Panasonic
v. Xiaomi, quoted in para. 106 of the present decision, that
SEPs differ in a key respect from
other patents. Normal patents are monopoly rights, and the primary remedy for
infringement is an exclusionary injunction so as to preserve the monopoly. This
is not true of SEPs, because they are subject to the SEP holder’s obligation to
grant licences to any implementer who desires a licence on FRAND terms. An implementer
is entitled to such a licence as of right. Thus SEPs are not property rights of
the same status as other patents. In effect, the SEP regime is a liability
regime in which the SEP holder’s remedy is a financial one. The only role for
an injunction in this regime is to enforce the SEP holder’s entitlement to that
financial remedy.
See
also id.
para. 21 (stating that “any SEP owner is a willing licensor at a high royalty
rate and any implementer is a willing licensee at a low royalty rate, but the
real question is whether the parties are willing to license at a royalty rate
which is in fact FRAND. Thus to decide willingness one
first has to determine what rate is FRAND, and then find out who is willing or unwilling
to license at the FRAND rate”). This is
very different, however, from the current German and UPC approaches, see,
e.g., my post on the UPC decision in Panasonic v. OPPO here.
Fourth,
I am concerned about the “jurisdictional imperialism” point. To be sure, it may be reasonable to infer, as
the court does, that the only reason Ericsson doesn’t want the Patents Court or
the EDNC to determine a global FRAND license, and instead is pursuing injunctive
relief and exclusion orders in other forums, is to force Lenovo to capitulate
into agreeing to above-FRAND terms. It
also may be reasonable to assume that a global FRAND cross-license determined
by either the Patents Court or the EDNC would, in fact, be FRAND. Nevertheless, we also should consider two
other logical possibilities: (1) that Ericsson and/or Lenovo expect that a global
FRAND cross-license to be determined by either the Patents Court or the EDNC
would be below-FRAND, and (2) that Ericsson would not use the leverage an injunction
or exclusion order would provide to extract above-FRAND terms (that is, to
practice holdup). I’m not sure it is
fair to assume these two possibilities away, notwithstanding my own “priors”
that the English and U.S. courts generally can be expected to do a reasonable
job in determining appropriate royalties, and that corporations tend to be
rational profit-maximizers. Others, such
as Judge Peter Meier-Beck are more skeptical than I about the courts’ ability
to determine FRAND royalties accurately (see here), after all; and how do I know that my
priors are right? On what basis can I be
sufficiently confident that I would be willing to penalize a litigant for
enforcing its rights as permitted in another jurisdiction? In more extreme cases, such as Microsoft v.
Motorola and Panasonic v. Xiaomi, where a litigant is alleged to have reneged on a
commitment to have a court establish a global license by pursuing litigation elsewhere
in the hope that this will compel the counterparty to give in, I think antisuit
injunctions or interim licenses probably make sense; but is this such an
extreme case?
Fifth,
and related to the preceding point, if the U.K. approach is right, then what is
to stop other countries from pursuing the same practice? There is a lot of grumbling right now about
the Chinese courts’ royalty rates being too low, as for example in the 2023 SEP
decision in Nokia v. OPPO. Assuming
for the sake of argument that such complaints are justified (I take no position
on the matter here), then should we be pleased nonetheless if a court in China were
to enter a declaration like the one in Lenovo, intended to nudge a SEP
owner to abandon litigation elsewhere in deference to the Chinese court’s
rate-setting? If not, then are we saying
that we trust the U.K. courts to do the right thing, so that they can make declarations
about interim licenses, but others cannot?
That seems . . . uncomfortable.
Maybe it’s best if nobody does this, except in extreme cases.
Sixth, perhaps the Lenovo decision,
assuming it stands, will be limited to its facts. By that I mean, maybe it will be seen as
appropriate for a U.K. court to enter an interim license declaration in a case
like the present one in which the SEP owner resists having any court
establish the term of a global (cross-)license, but not where it is willing to
let another court do so (though what if the parties disagree about which other
court is appropriate?). On that reading,
there would be no grounds for an English court to declare an interim license if
a SEP owner like Ericsson consented to having, say, the EDNC establish a global
royalty (and perhaps enter an interim license during the pendency of
proceedings), assuming that court has jurisdiction to undertake the task.