As many readers will know, Apple and Optis have been engaged in international litigation concerning Standard Essential Patents (SEPs) since 2019. On 26 February 2019, Optis filed an action against Apple in the English Patents Court for infringement of the GB designations of eight European Patents. Following a finding that at least one of Optis’ patents is valid and essential, and a dispute between the parties over abuse of dominance and the right to an injunction on an SEP, the two parties attended a trial last year to fight it out over the terms of a FRAND licence to Optis’ Portfolio. The trial spanned four months from June to October, albeit with a number of breaks and several rounds of closing arguments. By the time of the trial, Apple and Optis had both accepted that they would take a licence on the terms set by the English Court.
On 7 June 2023, Mr Justice Marcus Smith handed down judgment with his FRAND determination.[1] This is the second judgment issued by the English Courts containing a substantive FRAND determination since the pivotal judgment of the UK Supreme Court in 2020 in Unwired Planet (the first being the judgment of Mr Justice Mellor in InterDigital v Lenovo earlier this year). In the judgment, the Court re-emphasises its readiness to set the terms of a global FRAND licence and finds that as between Optis and Apple an annual lump sum of US$5.13m is FRAND for the portion of Optis’ portfolio(s) to which Apple requires a licence (just over 60%).[2]
At 283 pages long, the judgment is not a quick read and therefore this article only scratches the surface of its contents. The judgment is also currently heavily redacted, making parts of it somewhat difficult to follow. The Judge addresses this issue in a top note to the judgment, explaining that all potential confidential information was redacted so that third parties could have the opportunity to be heard at a further hearing before making the information public. However, the Judge also indicates a preliminary view that most or all of the redactions are indefensible. The hearing on confidentiality (and other consequential matters) will be listed for the first open date after 26 June 2023, following which an updated version of the judgment should be available (with fewer redactions).
Ultimately, Mr Justice Marcus Smith determined that a FRAND licence to Apple for the Portfolio was a lump sum licence, with the total sum payable by Apple being US$56.43m plus interest. The licence would be a worldwide 4G multi-standard licence covering all future Apple products (including the theorised “Apple car” costing US$100,000…). This comprised a forward looking licence running until the expiry of all of the patents in the Portfolio, valued by the Court at US$25.65m (five years’ annual fees), and a release in relation to six years’ past infringement, valued at US$30.78m. Interestingly, the Judge held that the past release should be calculated from when Optis first “asserted itself” in 2017 (seemingly on the basis that this was when Optis commenced negotiations with Apple regarding the Portfolio). This appears to be in contrast with the finding of Mr Justice Mellor in InterDigital v Lenovo that all past infringements should be paid for (even if that involves ignoring limitation periods), as well as comments made elsewhere by Mr Justice Meade that liability arises from first use of the patented technology.
The judgment is also interesting because it is the first time the court has really focused on the question of interest on past licensing fees (interest on past licensing fees is in issue in InterDigital v Lenovo, but the question of whether it is payable and in what amount was adjourned to a later hearing). Mr Justice Marcus Smith found that interest was payable as a matter of principle (it was discussed as a means to discourage hold-out). He made a “firm but provisional” finding that compound interest of 5% per annum was appropriate, but the matter has been adjourned to a further hearing (which will also consider any dispute over the precise terms of the FRAND licence, to be drawn up in accordance with the judgment, as well as the question of a 5G licence). All other ongoing proceedings between the parties involving this Portfolio should, in the Judge’s view, be compromised as one of the terms of the licence.
Turning to how the Judge reached these conclusions, before undertaking his own valuation of the Portfolio, the Judge criticised the approach of both parties. Whilst he was in favour of Apple’s general framework (and indeed went on to use a variant of it later), he felt that Apple’s proposed use of the smallest saleable patent practicing unit was “indefensible”, and that any proposed patent-by-patent assessment would be unworkable in relation to any portfolio of substance. Instead, the Judge held that all SEPs should be viewed homogeneously. Further, the Judge opined that the value in a portfolio of SEPs lay, in fact, in the right to access the Standard that a licence inherently involved. As regards Optis’ approach, amongst other criticisms (of Optis’ comparable licences, discussed further below, and any attempt to value a licence based on scaling the Unwired Planet rates) the Judge considered that Optis’ top-down cross check was untenable due to its unjustified use of ad valorem rates (i.e. based on the value of the product being sold), and the use of headline rates. He also felt that the inconsistency between the value of the entire SEP universe (referred to by him as the “Stack”) implied by the Optis comparables and the assumption that the aggregate royalty burden of the Stack was 15% meant that the top-down cross checks failed in any event.
Each party relied on 14 licences they said were comparable, though Apple elected to rely primarily[3] on its own out-licences rather than Optis’ licences. The Judge held that these two sets of allegedly comparable licences were categorically different from each other. He therefore did not form a single pool of comparables for use in the valuation. In his view, the counterparties in Optis’ comparables were too small and the inequality of bargaining power rendered them “worse than useless” as comparables, with there also being a concern that the licences were entered into primarily for the purpose of establishing comparables for litigation. The Judge considered Apple’s comparables to be more helpful in that they could be used to value the Stack, although the Judge emphasised that caution was still necessary when dealing with them.
When determining the value to Apple of the Portfolio, the Judge first calculated the value to Apple of the Stack and then scaled this using an estimate of Optis’ share of the Stack.[4] In doing this, the Judge originally found that the Stack comprised 26,600 patent families. However, to take into account the fact that the Stack had increased in size over time he used the reduced value of 22,000 patent families in his calculations. Optis’ share of the Stack was calculated to be 0.61% (based on 135 patent families). He attributed a value to Apple of US$8.235m per annum for that proportion of the Stack. As regards how that figure was derived, first the Judge used the unpacked rates from each of the Apple comparables to derive an implied value of the Stack in each case. Having established this possible range of values for the Stack, he excluded some of the possible values as outliers and modified others before averaging the remaining values to arrive at an estimate for the total value of the Stack per annum (the intricacies of the calculation are somewhat difficult to make out at this stage in light of the heavy redactions in this section). As foreshadowed this proportion was reduced to 0.38% (for reasons that were redacted), making the value US$5.13m per annum.
A lot of consideration was given to the question of whether the licence should be on an ad valorem rate, per unit rate, or lump sum, in the context of potential discrimination between implementers. It is worth noting that Optis favoured ad valorem rates whereas Apple favoured lump sum licences. The Judge felt this question to be highly nuanced, but in this context, a lump sum licence specific to Apple was not discriminatory. In this sense, he made it clear that he was “pricing the value to Apple of the Portfolio”.
Further, the Judge felt that unpacking Apple’s (lump sum) licences to determine an ad valorem rate was intrinsically unreliable. He explained that the more ‘unpacking’ required to make a licence appear comparable, the less likely that any useful inferences could be taken from it.
On the question of an abuse of dominance, or hold-out and hold-up, the Judge criticised both parties for “wasting valuable time and money” on these “entirely irrelevant” allegations. Nevertheless, the Judge ruled on the questions of (i) whether Optis had abused a dominant position by delaying negotiations and refusing to conclude a licence on FRAND terms (i.e., engaging in hold-up); and (ii) whether Apple had acted in bad faith by holding out in negotiations. On the first issue, the Judge held that although Optis’ conduct in negotiations was “inept” and “not conducive to achieving a negotiated outcome”, it was not abusive. In any event, Optis was not in a dominant position by virtue of, in particular, its FRAND undertaking. On the second issue, whilst Apple had made “missteps” in the negotiations this did not amount to “illegitimate Hold Out”; Apple had engaged with Optis throughout and had negotiated in good faith.
The authors await the outcome of the consequentials hearing with interest, as well as a slightly less redacted judgment.
[1] The citation is [2023] EWHC 1095 (Ch). The Judgment is dated 10 May 2023, and the reason for the significant delay before handing down of this judgment is not immediately clear.
[2] It should be noted that Optis’ portfolio comprises three separate portfolios, although due to the current redactions, it is unclear whether Apple ultimately required a licence to all three (though it is clear Apple only required a licence to a proportion of Optis’ patents). The portfolio is focused on 4G and earlier standards, and does not include any patents declared solely for 5G. For convenience, the term Portfolio is used in this article to cover all of Optis’ SEPs.
[3] It appears from the Judge’s comments that each party submitted solely licences in which they were one of the parties, although this is difficult to confirm given that the parties to the licences are currently redacted.
[4] The Judge used the data provided by Innography, the consultants used by Apple. The parties used different third party consultancies to determine data concerning the Stack and Optis’ share; Optis used PA Consulting, Apple used Innography. The Judge chose to disregard PA Consulting’s data as he felt that the qualitative data was more ambitious and therefore intrinsically more unreliable.
________________________
To make sure you do not miss out on regular updates from the Kluwer Patent Blog, please subscribe here.
