Post-expiry royalties: what’s the problem?

There are some weird terms in US licence agreements. Let’s leave aside the general peculiarities of US contract wording. Examples such as “indemnify, hold harmless and defend”, “represents, warrants and undertakes”, “successors and assigns”, and a host of other excrescences, appear in many types of commercial agreement and not just IP licences. Instead, let’s focus on wording that deals with the duration of royalties in licence agreements. This issue came into sharp focus last week, with the decision of the US Supreme Court in the case of Kimble v Marvel Entertainment, LLC.

More on that case later. The general issue, in the US and internationally, is whether it is appropriate to require a licensee of IP to pay royalties after the IP has expired, been revoked, or otherwise ceased to exist. A generation or two ago, there seemed to be a consensus among legislators and the courts that it was not appropriate. This attitude could be seen, for example, in:

• the US Supreme Court case of Brulotte v Thys Co, a 1964 decision that was discussed and followed in the Kimble case linked above. In Brulotte, the court decided that a contractual obligation to pay patent royalties after the patents had expired was “unlawful per se“.
• the UK Patents Act 1977, which included a provision in section 45, since repealed, that: “Any contract for the supply of a patented product or licence to work a patented invention, or contract relating to any such supply or licence, may at any time after the patent or all the patents by which the product or invention was protected at the time of the making of the contract or granting of the licence has or have ceased to be in force, and notwithstanding anything to the contrary in the contract or licence or in any other contract, be determined, to the extent (and only to the extent) that the contract or licence relates to the product or invention, by either party on giving three months’ notice in writing to the other party.”
• the 1984 EC Block Exemption Regulation for patent and know-how licences, which black-listed a provision whereby: “the licensee is charged royalties on products which are not entirely or partially patented …without prejudice to arrangements whereby, in order to facilitate payment, the royalty payments for the use of a licensed invention are spread over a period extending beyond the life of the licensed patents …” Recital 22 to the Regulation clarified that this spreading of payments referred to “spreading payments in respect of previous use of the licensed invention” – ie use during the period when the patents were in force.

A possible solution to this issue is to grant a mixed patent and know-how licence, in which royalties can be charged for use of know-how in circumstances where there are no patents, eg because they have expired or not been applied for in a particular country.

While this solution may work in many countries, there has clearly been a strand of opinion that, in the USA, a more nuanced approach to royalty terms is required. It seems to be thought by some that the licence agreement should state separate royalty rates for use of patents and for use of know-how. Presumably this makes it easier to show that there is no disguised patent royalty after the patents have expired. This approach is consistent with a comment from Kagan J in the Kimble case. She said:

That means, for example, that a license involving both a patent and a trade secret can set a 5% royalty during the patent period (as compensation for the two combined) and a 4% royalty afterward (as payment for the trade secret alone).

IP Draughts has seen some very strange royalty terms that try to finess this issue, eg providing separately for X% for use of patents and another royalty of X% for use of know-how, but stating that for as long as both patents and know-how protect the product, only the patent royalty applies. After the patent expires, only the know-how royalty of X% applies. Hey presto, X% applies both before and after the patent expires! IP Draughts has severe doubts about the effectiveness of this type of legal engineering.

More conventional, in IPDraughts experience, is a clause that sets the royalty at X% and reduces it to 50% of X in any country where there is no valid patent.

Ley lines

IP Draughts’ impression is that economists and competition (or in the USA, antitrust) authorities are no longer as concerned about post-expiry royalties as they once were. For example, the European Commission’s 2014 Guidelines on Technology Transfer Agreements state, at paragraph 187:

Notwithstanding the fact that the block exemption [for technology transfer agreements] only applies as long as the technology rights are valid and in force, the parties can normally agree to extend royalty obligations beyond the period of validity of the licensed intellectual property rights without falling foul of Article 101(1) of the Treaty. Once these rights expire, third parties can legally exploit the technology in question and compete with the parties to the agreement. Such actual and potential competition will normally be sufficient to ensure that the obligation in question does not have appreciable anti-competitive effects.

Kimble

In Kimble, the parties had settled patent litigation on terms that the inventor, Kimble, assigned a patent to Marvel in return for royalties. The parties set no end-date for the payment of royalties. Some years later, Marvel “stumbled across Brulotte” and sought and obtained a declaratory judgment that it could cease paying royalties at the end of the patent term. On appeal, the Supreme Court upheld the award of the declaratory judgment. In passing, one wonders how such a defective settlement agreement could have been drafted. Presumably the parties were advised in their patent litigation and settlement negotiations by lawyers who held themselves out as specialists in US patent law.

The majority of the justices in Kimble appeared to recognise that the current thinking of economists (and therefore antitrust authorities) does not object to post-expiry royalties. As the majority judgment put it:

A broad scholarly consensus supports Kimble’s view of the competitive effects of post-expiration royalties, and we see no error in that shared analysis.”

However, that consensus was irrelevant, according to the majority, as the issue before them was one of interpreting statutory patent law, and not antitrust law. The Supreme Court was bound by the principle of stare decisis to follow the decision in Brulotte. If the case had been properly considered as an antitrust case, they might well have been prepared to decide Kimble differently.

IP Draughts found this part of the Kimble decision surprising. Though he has no expertise in US laws, he had always understood the general issue, at least as it is understood in the UK and Europe, to be one of competition (antitrust) law.

The 3 minority justices in Kimble also saw things differently. They commented that the earlier Brulotte case “was an antitrust decision masquerading as a patent case”.

Good old stare decisis. IP Draughts remembers being taught about the English version of the rule in his first term as an undergraduate law student, in 1979. Courts are sometimes bound by earlier court decisions on points of law. The English rule is not so constraining as the US one, it seems. The UK House of Lords (now the UK Supreme Court) simply announced in 1966 that it would no longer consider itself bound by its previous decisions.

IP Draughts is left feeling perplexed by the decision in Kimble. It is concerned only with a narrow point on the duration of patent royalties. But on that narrow point, US licensing practice and to some extent (because of the strong, international influence of the US) non-US licensing practice, is frozen in time by the opinions and decisions of an earlier generation of US judges. It matters not whether the decision is based on statutory interpretation or antitrust laws, the practical effect is the same.

Practitioners advising on licence agreements that have a US element to them must consider carefully how the royalty duration is expressed. Many of IP Draughts’ licence agreements provide for royalties to be paid, on a country-by-country basis, for the longer of (a) the duration of the licensed patents, or (b) in the case of know-how, for a period (often 10 years) from the first commercial sale of licensed products. At first glance, this would appear to address the issue. What is troubling IP Draughts is whether the agreement needs to go further, in light of this US case law, and have separate royalty rates for patents and for know-how, as some US templates for patent licence agreements seem to prefer. Readers – do you think this is necessary?