Getting the EU SEP Regulation right — Part 1: Hold-up vs. hold-out

It’s a simple truth (and not even specific to SEPs) that patent licensing and patent litigation are inextricably linked. A patent is not a license to do something, but a license to sue. Even if you obtain a patent on something, it’s possible that the moment you actually implement your own invention, you infringe one or more patents held by others. And you don’t even know if your patent will survive a future challenge.

With copyright it’s simpler, but even your copyright isn’t worth anything if you can’t enforce it against infringers.

Real property works differently. There is rarely doubt over ownership, and in compliance with applicable laws and regulations, you can build a house on your piece of land. If someone parks a car there without permission, you have the right to have it towed away, and that is similar to suing an infringer of an intellectual property right.

Why would an IP-specialized website even state those fundamental facts? The simple justification would be that analogies with non-IP purchases were drawn in the build-up to the EP vote. The more important reason is that we must remember in the context of policies regulating enforcement that patent licenses are not “sold” in accordance with supply-and-demand dynamics. Someone who implements a standard builds a house on some other people’s property without previously having to negotiate a land purchase or lease.

Contrary to the dogmatic and unbalanced Orange-Book-Standard guidance by the Federal Court of Justice of Germany (2009), ip fray does not believe that an implementer of a standard should have the obligation to seek a license from each and every SEP holder before the first sale. No other jurisdiction ever adopted the German outlier stance anyway, and even Orange-Book-Standard immediately softened that position (even if only slightly so) after stating it. But there must be a reasonably efficient process once a SEP holder knocks at an implementer’s door and wants to get paid.

That still leaves the burden of identifying implementers on the SEP holder’s part. If one wants to take a symmetrical position on transparency and registers, it’s worth thinking about a requirement that implementers register any standard-compliant products, making it easier for right holders to identify and contact them. Such a register would arguably also have to disclose (not to the general public, but to those who have a legitimate interest in knowing) approximate aggregate sales volumes.

“Market” dynamics

The only reason someone takes a patent license (unless it’s part of a commercial agreement that involves more than licensing) is that they will otherwise get sued. If they can get away with never paying a cent, or with paying much later without significant incremental costs, then that’s what they’ll do in accordance with the economic principle.

There is no “demand generation” of the kind you see in a product market. If the right holder wants to turn the implementer’s ongoing use of a patented invention into a demand, it has to threaten with—and if all else fails, actually bring—litigation. A licensing “offer” is, in the event of already-ongoing use, also a threat of enforcement.

The above has implications for both sides in the SEP Regulation debate:

• Right holders and those sympathetic to them cannot, with the sole exception discussed below (pools), portray the SEP licensing process as taking place in a market that will determine the right price all by itself. If right holder A wants to get paid, the answer is not to take a license from right holder B. If they hold actual and valid SEPs, you will need both licenses. The substitutability of offerings is an indispensable requirement of a market.
• Implementers and their supporters must accept the fact that weakening or even just delaying patent enforcement reduces “demand” and results in fewer or at least slower transactions.

There is only one exception where limited market dynamics do come into play: if an implementer can license a patent (or, more likely, patent portfolio of whatever size) either bilaterally (i.e.. directly from the right holder) or through one or more pools. In that case, “demand” may still have to be generated through enforcement, but at least there are substitutable alternatives (one license or the other). That is a good thing as it means pools have an incentive to pass on to their licensees some of the transactional efficiencies they generate. It is one of the unfortunate aberrations of German SEP case law that implementers facing an injunction increasingly often don’t get that choice.

License fees may spiral out of control in one direction or the other

There are plenty of publicly-accessible court orders and rulings from which one can infer, and ip fray indeed has inferred, that both hold-up (overleveraging of SEPs by a right holder) and hold-out (unreasonable delays and other complications caused by an implementer) are facts of life. It will be difficult to have a constructive discussion over the next steps with flat-earthers who deny one phenomenon or the other.

What is often overlooked, but must be considered particularly before lawmakers put a thumb on the scales of SEP licensing, is the fact that an imbalance of one kind or the other will perpetuate itself and may even exacerbate because of FRAND dynamics. When courts have to determine what constitutes fair, reasonable and non-discriminatory terms, the practical reality is that it mostly comes down to comparables, to the ND part of FRAND. So if you have an environment in which patent royalties are inflated by hold-up, “market” prices go up for the next implementers to be sued, and if hold-out forces right holders to lower their rates just to get paid in a reasonable time frame, it will sandbag royalty rates.

If the starting point of the above-mentioned Orange-Book-Standard decision applied, implementers would have to pay extortionate license fees because they couldn’t launch a product without being enjoined on a permanent basis. The other extreme would be what Judge Richard Posner proposed in Apple v. Motorola: an injunction would come down only if someone refused to pay, after the exhaustion of appeals, a court-determined rate. Other than that, right holders would have to content themselves with infringement damages awards. In that case, right holders would have to make huge concessions to get deals done in the near term. Public companies with a strong dependence on licensing income owe it to their shareholders to do deals. All companies that use licensing income from their past innovations to fund the next round of innovations depend on license fees as the lifeblood of that business model.

Mainstream and outlier positions on access to injunctions

Across the world’s jurisdictions in which anyone ever asserted a SEP, there is a clear consensus that hold-up must be prevented by denying injunctions in certain circumstances, but that hold-out must also be deterred. In U.S. federal court, patent injunctions are hard to obtain under the eBay v. MercExchange standard, but damages enhancements are a major deterrent. In the U.S. International Trade Commission, the equivalent of an injunction is avoided only on public-interest grounds. Outside the U.S., the availability of injunctions is subject to different criteria. No other jurisdiction focuses as much on implementers’ conduct as Germany where almost every SEP case gets decided on that basis, and ip fray does not defend that approach. But that doesn’t mean that a particular measure to redress the balance deserves to be supported.

If the proposed SEP Regulation was adopted in its current or any similar form, the EU would be the first jurisdiction worldwide to afford implementers an additional year (nine months plus preparations) of hold-out without any counterbalancing threat that would discourage them from causing such delay.

Separately from the question of whether the status quo should be sustained, the question must be asked: is the rest of the world just not as smart as the EU, or is the EU not as smart as the rest of the world?

There isn’t any objective justification for that delay. If the concern was that the FRAND opinion delivered by an EUIPO-appointed “conciliator” would otherwise come too late because the right holder might obtain an injunction and force the implementer into a license, then there would be much more appropriate solutions:

• In practice, no SEP case will be decided during that period unless the right holder obtains a preliminary injunction. The regulation could rule out PIs over SEPs. That would mean an abundance of caution as it hasn’t happened in the EU. It does happen in some Latin American jurisdictions, and there’s nothing the EU can do about that.
• The regulation could also say, regardless of whether it’s a preliminary or permanent injunction, that a court should not order an injunction while the FRAND determination is underway.

If policy makers do not opt for a less restrictive, yet effective alternative, then the collateral damage must be intended (because of some agenda, in which case it isn’t even “collateral”) or they incredibly don’t understand that those alternatives exist.

The proposed regulation envisions a provisional measure of a purely financial nature, but in most EU member states that option does not exist. In any event, interim payments or deposits do not have the effect of deterring hold-out.

A litigation ban for an extended period raises serious issues. High-ranking French judge Nathalie Sabotier said at a conference hosted by the Fair Standards Alliance in 2023 that French courts don’t believe they can lawfully stay a case for more than three months during mediation.

Members of the European Parliament in favor of the proposal just made the conclusory assertion that the proposal was proportionate. And a former EU General Court judge paid by implementers suggested so last year. As did the European Commission’s Directorate-General for the Internal Market (DG GROW), where there must be major misconceptions concerning patent law and SEP enforcement, judging by certain objective flaws of the Commission proposal.

If there was going to be a formal allegation of a violation of fundamental rights, the proportionality question would inevitably also involve the consideration of less restrictive alternatives, just like in antitrust law, which the judges of the European Court of Justice are intimately familiar with.

The end doesn’t always justify the means. As this entire series of articles will show, the EU Council has rightly raised a number of questions not only about the evidentiary basis of all of this, but also about the proposed means.