The European Commission is part way through a consultation process on a tightening up of the rules on technology transfer agreements. The proposed changes are an evolution, rather than a revolution, of the current regime which dates back to 2004 and which must be replaced by 2014. This article focuses on the proposed changes and their implications for business.
The Commission has published a draft new Technology Transfer Block Exemption Regulation ("TTBER") and draft accompanying guidelines. The deadline for responses to the consultation is 17 May 2013.
To recap, the TTBER provides a safe harbour for patent and know-how licences between two parties and meeting specified criteria. The safe harbour protects the licence from challenge under Article 101 of the EU Treaty, which prohibits anti-competitive agreements. Licences falling outside the safe harbour are not treated as automatically anti-competitive; rather, they have to be analysed carefully to identify whether their object or effect is anti-competitive.
Existing agreements and changing circumstances
In-house counsel do not need to revisit agreements already entered into before any new law based on the 20 February proposal comes into effect. That is, if pre-existing agreements fall fairly and squarely within the 2004 TTBER, they do not become anti-competitive simply by virtue of the new Regulation coming into force.
However, as under the 2004 regime, any agreements which fall within the safe harbour under the new regulation should be kept under review in case the competitive relationship between the parties changes over time. This is because different rules will apply according to whether or not the parties are competitors: not surprisingly, a stricter regime will apply to agreements between competitors. If the parties start off as non-competitors but then become competitors over time, the agreement will still benefit from the exemption offered by the Regulation unless the agreement is later "amended in any material respect". That includes not just an amendment as normally understood, but also (in a change from the existing regime) the situation where the parties enter into a new and separate technology transfer agreement relating to competing technologies.
The list of hardcore restrictions is also tougher where the agreement is "reciprocal" (i.e. where there is a cross licence). As under the current regime, a one-way licence can become "reciprocal" if the parties enter into a new parallel agreement so that, taken together, there is a cross licensing relationship.
In-house counsel therefore need to monitor the changing competitive landscape and in particular the signing of parallel agreements with the same counterparty which might remove the benefit of the exemption from a formerly exempted contract.
Market share issues
The newly published proposal follows an earlier consultation launched in December 2011, which elicited a variety of comments, most in favour of the market share threshold approach or at least regarding it as something of a necessary evil.
The introduction of that approach in 2004 was a major shift away from the predecessor regime which took a more formalistic approach, but which had the benefit for industry of certainty: if you stayed on the right side of the line, you knew that your licence would not fall foul of the Treaty rules against anti-competitive agreements, with the attendant risks of clauses or agreements being invalid and fines being levied. In 2004, the old approach was swept away in favour of one which depends on whether the parties to the technology transfer agreement had market shares above certain thresholds. That is a logical stance for the Commission to take since, after all, any anti-competitive effect of a licence should be examined in its proper context, but it does make it hard for a business to know with any certainty whether it is within the parameters of the block exemption, particularly given the notorious difficulty of defining the relevant market for the purposes of calculating market share.
Under the current regime, the safe harbour applies where the combined market share of parties who are competitors is 20% or less of the relevant technology and product market. Licences between non-competitors are protected where the market share of either party is 30% or less. Under the proposed new regime, this 30% threshold would be reduced to 20% where the licensee owns technology which it uses only in its own business and which is substitutable for the technology which is being licensed to it (since in this situation, the licensee could keep exclusive access to its own "captive" technology as well as getting exclusive rights from the licensor, and thereby foreclose access to competitors).
The hardcore restrictions which take the entirety of the licence outside the block exemption – and can also lead to fines - are provisions the Commission regards as almost inherently anti-competitive. They are therefore a useful checklist of what not to include even in agreements which are not covered by the TTBER – for example because they have multiple parties or relate to different kinds of IP rights or because the parties' market shares are too high.
The list of hardcore restrictions has changed in only one respect: the ability of a licensor to restrict "passive sales" by the licensee into a country or to a customer group which is exclusively allocated to another licensee used to be permitted for two years in an agreement between non-competitors. That two year let-out is due to be abolished, bringing the TTBER into line with other block exemptions.
No more work-arounds for excluded restrictions
Restrictions excluded from the scope of the safe harbour are less objectionable than hardcore restrictions in the sense that they generally have a less serious impact on competition and so may be severed from the licence, leaving the remainder covered by the TTBER – a hardcore restriction renders the whole agreement invalid. Two clauses commonly used by licensing lawyers to work around excluded restrictions, and explicitly permitted within the current regime, are to be removed from the safe harbour, as follows.
Under the 2004 TTBER, a licensor cannot require a licensee to grant it an exclusive licence of the licensee's improvements to or new applications of the licensed IP – unless those improvements are "non severable" i.e. cannot be used without infringing the licensed patents or breaching the licensed know-how. The common work-around (permitted under the existing rules) is to have an exclusive grant-back of non-severable improvements and a non-exclusive grant-back of severable ones. Under the new proposal, grant-backs must always be non-exclusive. (It is still the case that an exclusive grant-back might not fall foul of Article 101 where the licensee receives a payment in exchange, depending on the overall assessment of the agreement, and in particular the parties' market positions.)
Second, the old regime forbade an outright ban on the licensee challenging the validity of the licensed IP rights (since that would put it in a worse position than third parties and lock it into the licence). Here, the common work-around is to transform this into a right for the licensor to terminate if the licensee chose to mount a challenge. Under the new proposal, even a termination-on-challenge clause relating to patents falls outside the safe harbour and therefore needs individual assessment. No-challenge and termination-on-challenge clauses relating just to the licensed know-how (i.e. where the licensee claims that the know-how is not confidential) are still in scope, although less likely to be useful in practice.
The new draft guidelines (but not the TTBER itself) now explicitly deal with "pay for restriction" clauses in settlement agreements (also known as "pay to delay" in the context of settlements between innovator pharma companies and generic competitors). They note that these clauses "may under certain circumstances" be caught by Article 101 and must be scrutinised especially if the licensee is agreeing to delay or restrict its entry into the market in exchange for payment or some other inducement, where it would not have accepted as restrictive a settlement purely based on the strength of the patents.
The policy underlying these brief references to pay to delay agreements becomes more apparent when viewed in the context of the Commission's enquiry into the pharma sector in 2009 and its subsequent investigations into a number of players in the sector. The wording of the guidelines may change as those investigations reach an end or are settled.
The EU language corresponds roughly to the position that the U.S. Federal Trade Commission has taken with respect to pharmaceutical patent infringement litigation settlements. For over a decade now, the FTC has maintained that a pharmaceutical patent owner's provision of anything of value to a generic alleged infringer in a patent settlement creates a rebuttable presumption that the payment was for the purpose, and had the effect, of inducing the generic firm to agree to a licence that took effect later than it would have agreed to otherwise.
The majority of U.S. federal appellate courts have rejected the FTC's theory. Instead, they instead have adopted a "scope-of-the-patent" test, pursuant to which, absent evidence that the patent infringement suit itself is a sham (either because the patent was fraudulently obtained or because there is no objective basis for alleging that the accused product infringes the patents in suit, and the plaintiff subjectively intended only to tie the alleged infringer up in litigation rather than to obtain a judgment on the patent), the settlement and licence are not subject to anti-trust scrutiny as long as the conduct restricted is within the scope of the patent asserted. Thus, if a settlement does not incorporate an agreement not to launch the accused product until sometime after the expiration of the relevant patents, and does not restrict non-infringing conduct, such as the manufacture of non-infringing products, it is effectively immune from antitrust scrutiny under the scope-of-the-patent test.
Recently, the United States Supreme Court decided to resolve this issue; on March 25, it will hear oral arguments in Federal Trade Commission v. Actavis, Inc., concerning the settlement of patent infringement litigation involving the pharmaceutical product AndroGel. The Court’s decision in this case is likely to finally determine the proper mode of analysis for infringement settlements and licenses granted pursuant to those settlements.
No-challenge clauses within settlement agreements are generally permissible in that they are unlikely to fall within Article 101 in the first place. The exception (a change in the guidelines) is where the licensor knows or could reasonably be expected to know that the patent in question is not valid, for instance where misleading information was provided to the patent office, or where the licensee is paid or otherwise induced to agree to the clause.
Technology pools (often used to create industry standards such as those in the telecoms industry) do not benefit from the block exemption since the licences into the pool do not themselves relate to the production of licensed products (and since they usually involve more than two parties). However, they are covered by a reorganised and revised section within the guidelines.
The principal changes are as follows:
- licences granted by the pool are no longer automatically exempt, although the guidelines set out the principles which the Commission will follow;
- the definition of essential technology has been revised by referring to the absence of "commercially or technically viable substitutes" rather than the absence of any substitutes (and the definition distinguishes between standard-setting and other contexts); and
- as with the draft TTBER itself, no-challenge clauses are not permitted.
The guidelines provide that the creation and operation of the pool will generally fall outside Article 101 where:
- the process is open to all;
- there are sufficient safeguards to ensure that only essential technologies are pooled, and that sensitive data is only shared to the extent necessary;
- licences into the pool are non-exclusive and on FRAND (fair, reasonable and non-discriminatory) terms;
- as with the licences granted by the pool (above) there are no no-challenge clauses;
- pool members and licensees are free to develop competing products and technology.
Checklist for in-house counsel
- Monitor the progress of the new TTBER and guidelines (especially if settlement agreements involving patent licences, and/or technology pools, are of interest);
- Review precedents for work-arounds and bans on passive sales which will not be permitted from May 2014;
- Ensure there are systems in place to track new or amended agreements with the same counterparties, in case this results in the relationship becoming reciprocal, and to track changes in the landscape which might make the parties competitors.