The European Commission's leaked draft regulation on standard-essential patents (SEPs) (initial reaction, table of contents and synopsis) is one-sided (as its thrust is an attempt to devalue SEPs), but if it got enacted in this form, it would nonetheless represent a double-edged sword.
It was drafted without a sufficient understanding of SEP licensing and enforcement, and key functions would be administered by an EU agency that has no expertise in this area. The EUIPO is actually the EUTMO: the EU Trademark Office. Now, imagine you find yourselves in a taxi. The maps are flawed and the driver doesn't have a driving license. Even if the ones who created the maps and picked the driver intended to support you, do you think you would safely arrive at your actual destination?
It shouldn't be hard to reach a procedural agreement between proponents and opponents of SEP devaluation that the current draft cannot seriously serve as the basis for a formal legislative proposal. Even if one directionally agreed with the draft regulation, it's a disaster. The Commission needs to go back to the drawing board, and it badly needs expert advice. Yes, there was a SEP Expert Group in place a few years ago, but I have yet to find a single person who wants to accept responsibility for that poorly-crafted document. Much to the contrary, some people indirectly involved with the process even reached out to me proactively in order to distance themselves from the draft regulation. On LinkedIn, IPlytics founder Tim Pohlmann stressed that "[t]he members of [a group of researchers he was part of] did not have access to the documents (draft proposal and impact assessment) or were in any way involved with the drafting [of] the documents that have apparently been leaked."
Success has many parents while failure is an orphan.
The Commission should recognize that this is one of the most complicated and specialized fields in which to legislate, and that the draft regulation proves that a completely different process is needed to discuss potential measures with stakeholders.
For the purposes of this post, let's just imagine we are all--as many of us are indeed--trying to help net licensees bring down licensing costs. And let's specifically think of small and medium-sized enterprises (SMEs), a group the Commission particularly tries--but fails--to help.
From that vantage point, we have a lot to criticize and to be concerned about, particularly when taking into account how net licensors would respond to that regulatory framework. I'm now going to raise various issues from an implementer's perspective. The order in which I do so does not indicate a prioritization. Let me start with a link list:
Incremental costs to be passed on to licensees (at least in part)
Non-patent owners would be charged for aggregate royalty determinations
Non-SEPs could account for larger part of total royalty payments and make up for SEP devaluation
Risk of huge litigation push before the regulation actually takes effect
1. Incremental costs to be passed on to licensees (at least in part)
The costs of the SEP Register and the calculation of an aggregate royalty (per standard) are going to be borne for the most part by SEP owners. But in order to maintain the same level of investment in innovation, they will be under pressure to figure out ways to increase their licensing income.
2. Valuation costs split evenly between parties
Art. 66(4) says this about the costs of FRAND determinations:
"The fees and costs of conciliators shall be borne equally by the parties, unless they agree otherwise, or the panel suggests a different apportionment based on whether a party took part in the procedure or not. All other expenses and costs in connection with the FRAND determination incurred by a party are borne by that party." (emphases added)
While the draft SEP Regulation tries to impose costs on SEP holders to the greatest extent possible, and to give SMEs preferential treatment wherever possible, the above paragraph does not--as it cannot--do any of that. Here, everyone is treated the same way, unless one opts out of the process, in which case it would be a one-sided proceeding with a potentially one-sided outcome.
Those conciliators and the parties' own experts are not going to be cheap. Even if we assume they're not going to charge millions like they would in some U.S. disputes, they're going to have to spent a lot of time at reasonably high hourly rates. A FRAND determination involving a large portfolio would probably end up costing hundreds of thousands of euros per party, and even for small portfolios the amount would be very significant--all of that for what will purely be an advisory opinion (not binding on any court). For a draft regulation that purports to also protect the interests of "micro enterprises" that makes no sense, given that opting out (or participating without experts) is not a good idea.
3. Non-patent owners would be charged for aggregate royalty determinations
This is just one example of unintended consequence of that low-quality draft regulation: even companies that don't hold a single SEP--and this could even include SMEs--might be charged fees by the EUIPO-led "competence centre" for the aggregate royalty determination (which like everything in that draft regulation is just an advisory opinion that doesn't bind anyone) for a given standard. Art. 19(6) says this:
"The fees of the competence centre and the conciliator shall be paid by the contributors that participated in the process based on the number of technical contributions to the standard." (emphasis added)
A technical contribution to a standard is not necessarily patented. It can just be a suggestion by an implementer who holds no patents, or someone who holds patents on other techniques that are part of the standard than the one on which a technical contribution is made. Also, there will sometimes be a debate over whether a contribution is technical or not.
4. Costs of "balkanization" of SEP portfolios
One of many things the Commission has not properly considered here is that the envisioned regulation would strongly incentitivize--and almost dictate--the "balkanization" of large patent portfolios through the assignment of patents to different parties. The Commission's well-meant but poorly-considered SME fetish makes it a very attractive option to have SMEs enforce those patents. The assignment agreements could still enable the assignor to get the lion's share of the license fees generated.
Those small and medium-sized assignees would pay lower fees under the draft regulation and have more control over the samples used for essentiality checks. Now, imagine you're an implementer, such as an IoT startup. You'll get royalty demands from far more entities. And if multiple such entities start FRAND determination procedures against you (followed by potential infringement actions), it will soon become so costly that you're not even going to want to go down that avenue.
5. Costs of valuation of NON-essential patents
The draft regulation treats entire patents as (potentially or actually) standard-essential. But as ETSI and other standard-setting organizations recognize, essentiality is actually a per-claim question. A patent may have a broad independent claim (typically claim 1 is always the broadest one) that is essential, but any or all of the derivative claims may be non-essential due to their additional limitations. Also, many SEPs come with claims that relate to end-user devices, which read only on the user equipment (UE) parts of standards, while other sets of claims read on the network infrastructure side. It's possible that one set is essential while the other is not.
The FRAND determinations envisaged by the draft regulation would, however, involve the valuation of all claims of all of those patents.
6. Non-SEPs could account for larger part of total royalty payments and make up for SEP devaluation
Smaller licensing firms often have only SEPs, but major patent holders always license their entire portfolios to implementers: SEPs as well as non-SEPs (also called "implementation patents"). The only exception I've seen in practice was more than ten years ago: Nokia sued HTC over non-SEPs though HTC had a license to Nokia's SEPs.
Even if a FRAND determination under the envisioned regulation resulted in a devaluation of SEPs, patent holders could simply make their non-SEPs more expensive. Those are not restricted or encumbered. In current and recent disputes, companies like Nokia and Ericsson are/were asserting a mix of SEPs and non-SEPs. While non-SEPs can often be worked around in the event of injunctions, the cost of such workarounds may be high.
7. Damages awards could get costly
The draft regulation would result in protracted periods of time during which no injunction could issue. At first sight, implementers may like that, hoping that SEP holders would lower their royalty demands. But licensees could face a hefty bill: some courts will not limit the amount of damages awards for past infringement to a FRAND rate. The draft regulation mentions "damages" only twice, and just for the purpose of precluding damages for the period prior to registration of a SEP.
8. Advisory opinions may prove useless in some venues
There isn't much in the draft regulation in terms of hard and fast rules other than precluding enforcement unless certain requirements have previously been met. Royalty determinations (be it for a standard or for a particular implementer's use of a portfolio) and essentiality assessments are non-binding. In U.S. law this is called an "advisory opinion" (and a reason for which Article III courts would not have jurisdiction).
We have previously seen certain dynamics between "competing" venues. There is a possibility of some courts becoming or remaining popular SEP enforcement venues after the passage into law of the proposed regulation simply because they would minimize its impact. There are two things they can do with the wiggle room they enjoy:
They can focus on an implementer's conduct, as German courts currently do in this post-Sisvel v. Haier era. The FRAND determination under the draft regulation would sort of take Huawei v. ZTE into account, but it would always arrive at a set of terms and conditions, particularly at a royalty rate. There is no reason why German and other courts couldn't still identify negotiation conduct on the part of an implementer that warrants an injunction.
Courts could simply reject low-ball FRAND determinations that are not binding.
Just look at what happened to that German patent injunction "reform": it's been about 20 months since it entered into force, and "proportionality" has not made an impact in a single case.
9. Increased enforcement in non-EU jurisdictions
The draft regulation would only affect patents valid in the EU and their enforcement in the EU. It would not put global disputes to rest. SEP holders could decide not to enforce in the EU at all, or treat the EU as a lower priority (because of the delays) and make more of an effort to gain leverage elsewhere:
There would likely be more USITC complaints.
Lawsuits in U.S. federal court would also become attractive. At the moment, those proceedings are too slow compared to the ones in Europe, and it's too difficult to obtain an injunction. But damages awards can be huge, and if enforcement in the EU takes longer as a result of an EU SEP regulation, one might as well sue in district court.
Just last month, the Delhi High Court essentially issued an invitation to SEP holders to sue in India.
Latin America is increasingly interesting: when Ericsson settled with Apple, it had already enforced a preliminary injunction in Colombia and was about to enforce in Brazil. Regardless of what the EU does here, I expect Latin America to be of increasing relevance to my litigation and antitrust commentary.
One doesn't even have to look that far. There are non-EU jurisdictions in Europe that could become more attractive. In the UK, it remains to be seen whether InterDigital v. Lenovo (to be appealed) will go down in history as an outlier. But places like Switzerland (small, but affluent, and with a culture that is rather unforgiving of even minor infractions) and Turkey (similar population size as Germany, though lower average income) could attract more SEP litigation.
All of this could result in increased litigation costs without ultimately saving license fees.
10. Conciliators will have incentive to set high royalty rates
If we ask ourselves what could go wrong, we also have to wonder whether those FRAND rate-setting procedures under the envisaged regulation would actually bring down royalties. The Commission wants to devalue SEPs, but would that draft proposal have that effect?
There will be a small circle of qualified "conciliators" to work on those FRAND determinations (and some of them will work as experts for parties in other proceedings). It's going to be a nice income opportunity for them. The moment they think about what will actually grow that business, they're inevitably going to conclude that high royalty rates are in their interest.
That's because SEP holders will increasingly rely on non-EU jurisdictions (previous item) if those time-consuming EU FRAND determinations are too low. And to the extent those same concilicators advise parties, the budgets for such services will be greater if the royalty amounts in dispute are high.
11. Incentives for over- and/or underdeclarations
The draft regulation oddly creates incentives for overdeclarations as well as for a few strategic underdeclarations.
The purpose of the essentiality checks and published essentiality ratios per SEP owner is to disincentivize overdeclarations. But SEP holders may nevertheless conclude that overdeclarations are in their interest because of the top-down approach (first setting an aggregate rate for a standard, then making case-specific determinations).
What has me even more concerned is that contributors to standardization may strategically select patents for which there is some reasonable justification not to declare them as essential, but which would give leverage in litigation. That's due to the sixth item: non-SEPs will have to make up for potentially reduced SEP royalties. In order to bring an infringement allegation over such patents, SEP owners wouldn't be able to map the patent to a standard, but they could intercept messages or employ other technical methods to demonstrate infringement.
Under the draft regulation, registration is up to SEP owners. Third parties can report all sorts of observations of non-essentiality to the EUIPO, but you can't register a SEP that belongs to someone else.
12. Validity not assessed in essentiality checks
As an implementer, I would have a problem with the fact that validity is simply not assessed as part of those essentiality checks. I can't remember a single SEP case in which there was no invalidity challenge except one in which a company with hardly any presence in the German smartphone market (Viewsonic) didn't have room for a nullity complaint in its German litigation budget.
The Commission ignores the interplay between non-infringement and invalidity defenses: a challenge to the validity of a patent may be dispositive, either by taking the patent down or by narrowing its scope to the extent it becomes non-essential. As an implementer I would be worried that SEP owners may go to courts and seek preliminary injunctions over patents that passed an essentiality check.
13. Risk of huge litigation push before the regulation actually takes effect
The draft regulation is centered around a SEP register and other mechanisms that it will take 24 months to set up. And even before that 24-month period begins, there would be a point at which SEP holders would already begin to prepare.
There will be an incentive to file as many SEP infringement actions in EU courts prior to the end of the 24-month period as possible. The way the draft regulation is designed to work, already-pending cases would not be affected. Maybe some court dockets would be clogged for a while, but those timely actions would be certain to be unaffected by the new law.
This list here is non-exhaustive. There is probably more--or even a lot more--that implementers could or should criticize. The draft regulation is utterly deficient and clearly nowhere near the point at which a legislative process could be initiated.
