German cartel clearance of Automotive Licensing Negotiation Group for SEPs won’t matter if SEP holders simply say no

The FCO’s decisions formally matter only in Germany, and even in Germany, they are court-reviewable. The FCO does not have jurisdiction over the global SEP market, and the mere fact that the four companies wh proposed the ALNG (Volkswagen, Mercedes, BMW, ThyssenKrupp) are German does not mean that the FCO has worldwide jurisdiction over what they do. For that reason alone, the FCO’s announcement of a decision to “tolerate” the ALNG subject to various conditions (while reserving the right to go after misconduct) is merely a political signal at best or PR stunt at worst.

The distinction must be made between the formal legal effect and hypothetical indirect consequences (intended or unintended).

1. Formal legal effect: no license to hold out

Based on only the FCO’s letter, those automotive companies are not in a position to tell any SEP holder that they decline to negotiate bilateral license agreements. It’s not just that a cartel, like a merger, must be cleared in every relevant jurisdiction to go forward, as no single regulator can decide for the whole world. Even if, hypothetically speaking, every relevant jurisdiction in the world adopted the FCO’s stance (and if the courts, including in Germany, didn’t overrule it), what would the letter actually say?

Here’s a redacted version of the letter (in German):

The final part of the letter establishes five general conditions (unofficial translation):

1. The ALNG’s and its negotiation groups’ activities are limited to the licensing of general mobile telecommunications standards and of other standards that are not to a heightened degree specific to the automotive sector, such as WiFI and H.265.
2. The ALNG and its negotiation groups are open to automotive suppliers requesting to participate, as they may also be entitled to a license under competition law.
3. In legal and practical/economic terms, participation in negotiations with the ALNG or its negotiation groups is voluntary for licensors (patent owners and patent pools). This means licensors must be free to enter into such negotiations and free to terminate them anytime.
4. There will be no information exchange between two or more ALNG members beyond the extent strictly necessary for the ALNG to conduct the relevant collective licensing negotiations as described in the submitted documents. The preventive measures aganst a horizontal information flow between members of the ALNG must meet at least the standard of protection described in [REDACTED].
5. Any expansion of the ALNG’s or its negotiation groups’ activities to technologies other than the ones set forth in paragraph 1 will be previously notified to the Federal Cartel Office. This also applies to any other substantial changes to the organization or the ALNG’s activities.

In addition to the conditions for clearance laid out in that section, one also has to bear in mind how the FCO summarizes the ALNG’s description of what it intends to do (unofficial translation):

“Specifically, the ALNG will offer to its members (potentially in different groupings subject to interest) the possibility to have SEP license agreements negotiated with patent owners and patent pools by a negotiating team acting on the members’ behalf. For that team, rules to safeguard neutrality and to limit the (particularly horizontal) flow of information between individual members have been described. Potential licensors will be free to enter into negotiations with the ALNG. They can also terminate any commenced negotiations anytime. However, they will have to commit, for the duration of the negotiations with the ALNG, not to enter into parallel bilateral negotaitions with members represented by the ALNG and not to take court action against them.

“If the ALNG reaches an agreement with a licensor, its members are free (but not obligated) to adopt it. The ALNG’s founders come from the assumption that in a subset of cases, there may still be, after an agreement at the level of the ALNG, a need for bilateral negotiations in order to make adaptations reflecting the particular situations of individual members, such as with a view to cross-licenses.

The above raises various issues, even though the FCO letter says that the agency merely “tolerates” the ALNG and reserves the right to conclude otherwise if warranted by new aspects or facts becoming known. It’s foreseeable that the ALNG’s members would first try to bring down the license fee through collective negotiations and would then still have an incentive to engage in hold-out by claiming that individual adjustments are needed, be it because even a negligible quantity of a company’s own SEPs may open the door to cross-licensing negotiations or because they make up some other reason.

But regardless of whether one considers the proposal good or not, the question with a view to the formal legal effect is whether the existence of such an ALNG (again, simplifying this by just assuming that other jurisdictions allow it on the same terms) (a) deprives SEP holders of any rights, (b) will encourage unlawful behavior or (c) otherwise threatens to result in market distortions.

(a) Impact on patentees’ rights

The voluntary nature was mentioned in the press release, but it’s clearer now based on the actual letter. Based on the FCO’s comfort letter, the ALNG can tell patent holders (or pools) that it stands ready to negotiate a collective license, though patent holders will find out quickly that whatever result they achieve in negotiations with the ALNG does not mean the end of the road: individual members can still come up with reasons for which they believe they should get better terms. That creates yet another asymmetry between an LNG and a patent pool: the pool makes a valid offer, while the outcome of an LNG negotiation as described above would not yet be a set of actual license agreements.

The FCO’ wording concerning patentees’ voluntary participation is waterproof. It will be voluntary in legal, practical and economic terms.

If any automotive company on the receiving end of a SEP enforcement action argued that it’s a willing licensee because the SEP holder could have negotiated with an ALNG of which that implementer is a member, it would be easy for the judges to see that the SEP holder had every right to insist on bilateral dealings.

The starting point in SEP enforcement is that a patent is an exclusionary right. An affirmative defense based on an entitlement to a FRAND license still requires the implementer to be a willing licensee. By pointing to a comfort letter that explicitly says (even twice) that patentees don’t have to deal with the ALNG if they don’t want to, no one can establish a willingness to take a license. If a SEP holder acts in accordance with Huawei v. ZTE, it will give an infringement notice and simultaneously or further to the implementer’s request offer bilateral terms. It could then clarify that it will decline an invitation to negotiate with an ALNG, or it can decline if and when the invitation issues.

Where SEP holders take a bit of a risk is if they do embark on negotiations with an ALNG, but abandon them and start to enforce. But that’s their choice.

(b) Encouragement of unlawful behavior

Given that any exchange of information betwee the ALNG’s members must be strictly limited to what is necessary for the ALNG to carry out its work, an ALNG with which no SEP holder is willing to negotiate can’t engage in a group boycott without breaking the law.

The question is, however, whether the FCO letter increases the risk of unlawful conduct anyway. Unfortunately, that is the case: even if, in a hypothetical example, Ericsson declines to negotiate with the ALNG, Nokia may accept to do so. That means the licensing executives of major automotive players will coordinate negotiations with Nokia, and once they are in the same room, there is a heightened risk of them also talking about how things are progressing with Ericsson. It might be unlawful, but if they had inhibitions to even sit in the same room (which could also be a video conference) and talk about any SEP licensing negotiations, the risk would be much lower.

They could even try to come up with an excuse: given that Nokia and Ericsson are similar companies in some ways, they felt they couldn’t meaningfully discuss the Nokia negotiations without also talking about Ericsson at a high level.

It’s a can of worms, especially when an industry has such a “rich” tradition of forming cartels. And the FCO was actually fully aware of it when nevertheless issuing that comfort letter (unofficial translation):

“That said, there is always an inherent risk with structures entailing frequent contact between competitors that the dynamics of such interactions will in reality overstep the envisioned scope.”

The letter merely refers to the FCO’s “discretion” but does not address those concerns in detail. Other regulators and the courts of law should not content themselves with someone else’s “discretion.”

(c) Other market distortions

Effects on the market do not depend on whether something is binding, but on whether market power is aggregated and subsequently abused.

What the FCO letter enables automotive comapnies to do (again, if this were to become the worldwide situation) is to subtly indicate to SEP holders that there’s no enthusiasm for bilateral negotiations and they should talk to the ALNG. In that case, the SEP holder would know that negotiations with the ALNG would start quickly (though they might be drawn out and ultimately unsuccessful, and even if they succeeded, each member would reserve the right to bring down the royalty rate once again). The SEP holder would also realize that bilateral talks are going to be slow and one may have to go to court.

The FCO’s condition that SEP holders must be free in legal, practical and economic terms to accept or decline an invitation to negotiate with the ALNG is one thing. The reality of how that requirement can be enforced is another. It’s like in other legal contexts where a party may threaten subtly, possibly even tacitly (such as by not responding in due course), and can’t held liable for it. Simply put, no automaker will be so stupid as to write in a letter or email that the only way to negotiate with them is through the ALNG because otherwise they’re not really going to engage.

This is an even bigger problem than the one outlined in the previous section (information exchange with respect to licensors who declined to negotiate with the ALNG) because there would be market distortions even if no single company could be convicted of an outright violation of cartel law.

The legal answer would be for SEP holders to sue someone who subtly or tacitly threatens with hold-out. But SEP enforcement is costly, and most car makers are not as big as VW or Toyota but sell maybe a million cars a year. It would take many years of royalty payments to recover the litigation expenses of suing such a company (though by suing one or a few of them, one might dissuade others from hold-out).

2. Overbroad market definition

The comfort letter makes references ot the European Commission’s horizontal guidelines, according to which purchasing cartels with a market share below 15% do not generally raise competition concerns.

Obviously, the three large German automakers are well above that threshold. In fact, Volkswagen alone is not far from it. And they leave the door open to others to join them.

In order to be able to issue a comfort letter, the FCO seeks to downplay the problem in two ways:

• The distinction between standards that are of particular interest to the automotive industry and standards that are used more widely (such as cellular connectivity) is made in order to state the following:
  “The department comes from the assumption that on a licensing market for SEPs relating to general mobile telecommunications technologies, the demand share of current and potential future members of the ALNG, [which will be recruited only] from the automotive industry will collectively be certain to remain below the 15% threshold defined by the Horizontal Guidelines (para. 291). By contrast, if standards are specific to automotive applications to a higher degree, there is a possibility of the collective market share of the ALNG’s members exceeding those thresholds.”
• With a view to the downstream market of connected vehicles, the letter acknowledges that the 15% threshold may be exceeded, at least in certain segments (they may mean passenger cars, for instance). “But the costs of general mobile connective SEP licenses account for only a very small part of the total costs of a vehicle. Therefore, anticompetitive effects from the coordination of upstream [meaning the licensing market] purchasing on the downstream markets for vehicles are, therefore, not likely.”
  The press release referred to the fact that those costs are “below 1%” (in reality, a lot less).

It’s true that even if some automakers hypothetically got a better deal as a result of participating in the ALNG, that cost advantage is not going to make a material impact on competition in the downstream market. Even if they passed those savings on to customers, cars cost so much more that a reduction by, say, $10 would not result in a shifting of demand.

There is a serious problem, however, with how the FCO defines the upstream market.

In order to depress the market share of an ALNG that could theoretically involve the entire automotive industry (!), they look at the totality of all implementers. Of course, far more smartphones and certain other categories of devices are sold per year than cars. ip fray has doubts that an ALNG representing all automakers would actually remain below a 15% of all 5G SEP licensing revenues at a point where virtually every vehicle will come with 5G. But the bigger problem is the overbroad market definition.

The FCO either failed to or (for a desire to “break new ground” on the global stage) didn’t want to understand that licensing cellular SEPs to the automotive industry is not the same as licensing them to smartphone makers or IoT companies. Based on the FCO’s approach, even those making cheap asset trackers would be buying from the same upstream market. That can’t be right. A relevant antitrust market must have reasonably comparable conditions.

There’s also a huge rechargable-battery market of which the batteries incorporated into electric vehicles constitute only a part. There’s a huge market for “seats.” Of course, in those cases it’s easy to see that those are different products. A car seat looks different from an office chair. Batteries of electric vehicles are much larger than then ones you find in a handheld device. But a correct market definition is not limited to size or shape. It’s about the overall conditions of a market.

There is empirical evidence that the market for automotive SEP licensing is distinguishable from other SEP licensing markets. There are different prices, and there are reasons for that. There are different types of applications. And those differences are actually becoming more and more significant as technology evolves.

What the FCO would like to focus on is that no SEP holder will largely depend on the automotive industry because it can license its SEPs to other categories of implementers. But smartphones are not like asset trackers, and neither smartphones nor asset trackers are like cars. An asset tracker isn’t sensitive to network latency, while autonomous driving (whatever the degree of autonomy may actually be) very much requires low-latency connections. There are many SEPs that enable low latency. There are others that enable low-energy connections made by asset trackers.

The same SEP portfolio can be implemented in different ways, for different purposes and with different economic effects based on the specific characteristics of a product category.

There is a distinct market for automotive SEP licensing. The FCO chose to ignore that fact in order to be able to argue that no relevant thresholds are met.

3. Indirect ramifications

The FCO’s press release quotes its president as saying that with this decision, they are “breaking new ground in the area of competition law, a development which is significant far beyond Germany.”

Without an objective need to open the door to potential market distortions, the FCO apparently wanted to make a big splash.

In a world in which

• no regulator or court of law goes beyond what the FCO has authorized in this context,
• no court of law making the final decision on cartel law claims against automotive companies abusing their collective market power adopts the FCO’s market definition,
• no court of law denies an injunction over a SEP only because a patentee exercised its right to decline an invitation to negotiate with an ALNG, a right that is stated explicitly (even twice) in the FCO’s comfort letter, and
• 100% of all SEP holders decline to negotiate with the ALNG,

nothing will change. Then the ALNG won’t ever have a basis for any exchange of information. It will be an empty shell. But that’s a number of conditions.

ip fray cannot imagine that there would be problems with SEP enforcement (by companies declining to negotiate with an ALNG) in the U.S., the UK or the EU. And if a court or regulator was to point to the FCO letter, then the same court would probably have wanted to support a car maker’s hold-out anyway and the question was just what pretext would be chosen. The FCO letter would then just come in handy by pointing to a large Western economy’s national competition authority.

After years of lobbying competition regulators in multiple jurisdictions, the automotive industry players behind the ALNG idea have now achieved one second-class clearance decision: their idea is tolerated, for the time being, subject to conditions that leave it to SEP holders to determine the ALNG’s fate. Given that no regulator previously wanted to give them even that kind of comfort letter, it may look like a breakthrough. But they’re not going to get this in each and every one of the relevant jurisdictions.