This paper – which can be found here – provides a good (and straightforward) overview of the various sides of the debate on how to deal with the standardisation of intellectual property rights, standard essential patents (SEPs) and their licensing. It covers a bit of the same ground as the paper just reviewed in the post below, but in a more detailed fashion.
The relevance and recent prominence of SEPs derives from the fact that end-products in a number of economic sectors now require the incorporation of numerous patented technologies (resulting in a so-called multi-component product). The production of multi-component products requires manufacturers to manage patent thickets (i.e. networks of overlapping IP rights owned by multiple patent holders) and to obtain licences from multiple rights holders.
The alleged patent thicket problem is linked to the hold-up problem because the threat of an injunction by a right holder against a patent implementer who has already invested heavily in the manufacture of allegedly infringing products forces the implementer to either withhold the commercialization of the product or to agree to pay excessive royalties. A closely related concept to patent thickets and hold-up is that of royalty stacking, which refers to the situation where the end product incorporates many patented technologies leading to multiple royalty burdens on the implementer, all of which may have an interest in holding up the manufacturer in turn.
This paper seeks to present an overview of various arguments and of key developments occurring in this field. It does so as follows:
A first section is devoted to explaining the theory behind hold-up and royalty stacking. It begins with a description of Cournot’s theory of complements, before moving on to Shapiro and Lemley’s ground-breaking work on patent thickets and their 2007 proposal to prohibit the grant of injunctions in patent litigation. The reason for this is that the award of such injunctions as regards multi-component products creates incentives for innovators to extract excessive royalties (by comparison to social optimum) and for patent owners to wait until an alleged infringer has incorporated a patent into its product to raise a claim. Shapiro and Lemley also argued that the hold-up phenomenon is exacerbated by royalty stacking – i.e. the co-existence of multiple patents in a single product, which creates a risk for downstream firms of having to face claims from multiple patent holders, and thus incurring prohibitively high production costs after they have already invested heavily in manufacturing the multi-component product.
This section then reviews the main criticisms of this proposal, namely: (i) that it does not take into account the impact that prohibiting injunctions may have on dynamic efficiency (i.e. innovation); (ii) that it would distort the contractual arrangements of standard setting organisations, and could lead to patent holdout or ‘reverse patent hold-up’ (see bullet below); (iii) that there is no empirical evidence of patent hold-up or royalty stacking being an issue in the real world, or that existing IP and contractual mechanisms are unable to deal with patent hold-up or royalty stacking when they do occur.
The third section deals with patent holdout, also known as ‘reverse patent hold-up’. This occurs when a product manufacturer knows he cannot be forced to stop producing or distributing a product containing someone else’s patented input – e.g. because courts cannot award injunctions regarding that IP. The manufacturer can then refuse to agree terms and wait for the IP right holder to start proceedings, in which case the worst case scenario for the manufacturer is that it will have to pay the royalties originally requested. The logical consequence of this is lower than optimal royalties, and concomitant lower investment by innovators.
The paper reviews a number of Indian decisions where hold-out was found to have taken place. It is also observed, by reference to some German case law, that the problem of patent holdout is particularly relevant in the context of portfolio licensing – with the result that: ‘patent holders engaged in portfolio licensing may find it more convenient to break up their portfolios into smaller, relatively better enforceable ones, thus taking away the efficiency achieved through large portfolio licensing in the first place.’
Another section looks at this debate from a competition law perspective – and particularly at: ‘whether a SEP holder who seeks an injunction commits an antitrust violation in the light of FRAND commitments’. Simplifying the matter immensely:
- In the US, various bodies have issued conflicting statements on the matter. For example, the DoJ and the and US Patent and Trademark Office issued a joint policy document stating that SEP holders may have market power; that F/RAND commitments may influence the choice of a remedy in patent litigation, and may render injunctive orders inappropriate; but that there may be instances where injunctions are appropriate. The International Trade Commission twice issued injunctions against the sale and import of products that infringed Samsung patents, and in one of the cases the US Trade Representative vetoed this decision while in another it did not. Finally, the courts have reiterated that injunctions are available for patent infringements subject to the balancing test of interests that normally applies to injunctions. As a result, the availability of injunctions for patent infringements is assessed on a case by case basis.
- In Europe, the European Commission initially adopted a licensee-friendly position in respect of SEPs in the Samsung and Motorola cases – in both cases the Commission concluded that the patent holders had abused a dominant position by seeking and enforcing an injunction against a SEP licensee. This took place at the same time as the German Federal Court of Justice decided that an implementer who is subject to an injunction suit may raise a FRAND defence, i.e. the implementer may defeat an injunction request if the SEP holder is abusing a dominant position by refusing to conclude a contract on FRAND terms. However, the EU and German approaches differed in a number of points, including what the licensee was required to do in order to be able to claim it had been a victim of an abuse. As such, a preliminary reference was eventually made to the European courts in the Huawei v. ZTE As described in much greater detail in my emails of 28 April and 26 May 2017, the CJEU enumerated a number of duties for both patent holders and licensees when negotiating FRAND – breach of which may, depending on who breaches these duties, either facilitate or defeat findings of abuse on the patent holders’ part.
A last section argues that there is a need to balance the interests of patent holders and licensees. The author argues that innovators need to be motivated to continually engage in intensive R&D in order to contribute to technological standards. While hold-up and royalty stacking may occur in some isolated cases, it is no justification for the creation of a system that punishes innovators, in particular when experience has shown that courts (and antitrust authorities) are well positioned to deal with these cases.
In the light of this, a rule that prevents good faith patent holders from obtaining injunctive relief against potential licensees is seen to not be appropriate. Instead, the EU approach seems to identify a balance between the interests of originators and implementers.
While I found the structure of the paper slightly odd (I did not follow it above), and would have liked a more prescriptive suggestion about how to balance the interests of patent holders and implementers in practice, I nonetheless think that this paper provides a very good overview of the interests at stake and the way different jurisdictions have tried to balance them.