INTELLECTUAL PROPERTY
Reform of the technology licensing rules
David W Hull, Covington & Burling

While the pending changes to the Merger Regulation and the Commission’s modernisation proposal have dominated the limelight during the past year, another pending reform of importance for industry is that of the EU technology licensing rules set forth in the Technology Transfer Block Exemption (‘TTBE’).1 Because the TTBE serves as a virtual template for licence agreements, these changes are of practical significance for companies doing business in the EU, particularly those in industries that rely heavily on licensing such as pharmaceuticals, biotech, and software.
In December 2001, the Commission published a report on the TTBE which gave the TTBE low grades in many areas and suggested that a fundamental reform of the TTBE is in order.2 The Commission floated a number of ideas for reform, and asked for public comment. The Commission received around 30 responses (available on DG Competition’s website), most of which are broadly supportive of the reform. It is expected that the Commission will publish a draft regulation and guidelines by the end of 2002.

Problems with the current TTBE
In its report, the Commission identified several major problems with the TTBE. Perhaps the most fundamental problem is that it is out of line with two major reforms in the competition field undertaken by the Commission in recent years: (1) the adoption of new rules on horizontal and vertical agreements, and (2) the modernisation proposal that is currently under discussion.
The TTBE is in need of a thorough revision to bring it into line with the new rules on horizontal and vertical agreements. When viewed against the backdrop of these rules, the TTBE stands out as an anachronism from another era with its detailed lists of dos and don’ts. Considering that the current TTBE was adopted only a few years ago, in 1996, its obsolescence is a measure of just how quickly EU competition policy has since evolved. When the Commission adopted the new rules on horizontal and vertical restraints, it jettisoned the legalistic, rule-oriented approach that still characterises the TTBE in favour of one that emphasises economic analysis and that focuses more on inter-brand competition and on the possible efficiencies related to certain restrictions. The rules were simplified by providing for only a single short list of hard-core restrictions. The new approach gave companies greater freedom to structure their transactions in ways that made the most commercial sense. The Commission acknowledges that the TTBE is complex and formalistic, and is in need of a similar overhaul. As its report points out, many businesses view the TTBE as a straightjacket that discourages efficient transactions and the dissemination of new technologies.
While bringing the TTBE into line with the rules on vertical and horizontal agreements would be a welcome development, it is hoped that the Commission will be careful not to import these rules into the TTBE without carefully considering whether they make sense in the context of technology licensing. For example, in considering the appropriate scope of territorial restrictions, licensees may deserve greater protection than distributors because they often must construct manufacturing facilities in addition to incurring costs associated with marketing and sales activities.
The TTBE also has major shortcomings when viewed in the context of the Commission’s modernisation proposal, which is aimed at decentralising the application of the EU competition rules by giving national competition authorities and national courts greater power to apply these rules. Such a decentralised system places a premium on having clear, unambiguous rules in order to achieve as much uniformity as possible in their application. With its complex and detailed provisions, the current TTBE fails to provide the kind of clear guidance needed at the national level. The TTBE’s provisions are notoriously complex, so that even EU competition specialists face uncertainty in applying them. If the current version of the TTBE were applied by national judges unfamiliar with competition law concepts, this could lead to unpredictable results and a patchwork of interpretations.
A revision of the TTBE along the lines of the rules on vertical and horizontal agreements ostensibly would result in a more streamlined regulation with fewer detailed rules. However, the approach outlined in the Commission’s report suggests that, at the same time that the Commission is discarding the current approach of using detailed white, gray, and black lists, it will introduce a complicated matrix of market share thresholds. The net result of these changes could be to make the TTBE even more difficult to apply in practice, which would undermine the goal of achieving a simplified, user-friendly regulation.

Key themes of the reform
Competitors v non-competitors
Two key themes emerge from the Commission’s report. First, in line with its aim of moving away from a legalistic approach towards one based on economic analysis, the Commission plans to differentiate between licences between non-competitors and those between competitors. As a general rule, the former raise fewer competition concerns than the latter and, logically, should be subject to more lenient rules.
While few would argue that sound competition policy demands a difference in treatment of competitors and non-competitors, an issue that will certainly generate debate is how to define ‘competitor.’ In its report, the Commission puts forward a tentative definition under which companies would not be considered to be competitors in the following situations:

• where the licensor does not exploit the technology itself;
  
• where the licensee is not active in either the technology or product market;
  
• where the licensor and the licensee were producing competing products before the licence, but the licensed technology represents such a ‘sweeping breakthrough’ that there would no longer be any competition between them absent the licence; and/or
  
• where the licensor and licensee are in a mutually blocking position with respect to their IPRs.

The addition of these last two categories represents a significant narrowing of the definition of competitor when compared to the definition currently contained in the TTBE. As usual, the devil will be in the detail as it will be difficult for the parties to know when they are dealing with a ‘sweeping breakthrough’, particularly when the breakthrough has yet to be made.
Limited licences
A second key theme that runs through the Commission’s report is the notion that the TTBE should be changed to make it easier for a licensor to grant a limited licence of its IPRs, ie that the licensor ought to be able to grant the licensee exclusive rights for a given territory, field of use, or customer group, but retain the residual rights for exploitation by itself or by others. Under the current TTBE, such limited licences are only possible in a few situations. Arguably, this restrictive approach may discourage licensors from licensing their technology at all for fear that they will be creating competition in their own markets. Moreover, it is difficult to justify the current approach that allows some forms of limited licences, but not others.

The Commission’s proposed framework
In its report, the Commission provided the broad outlines of a revised TTBE. In order to bring the TTBE into line with its approach to horizontal and vertical agreements, it is likely that the Commission will adopt a wide, umbrella-type block exemption in combination with an accompanying set of guidelines. The Commission might adopt market share thresholds as a way of limiting the block exemption. Moreover, it appears poised to proceed with a simplification of the current regulation.
The recommended framework divides licences into two categories: those between competitors, and those between non-competitors. As to non-competitors, a revised TTBE could treat restraints unrelated to the licensed IPR, such as non-compete clauses and tying obligations, in a manner similar to the Vertical Restraints Block Exemption (30 per cent market share threshold, prohibition of certain hardcore restrictions, and other conditions). Restraints related to the exploitation of the licensed IPR, such as territorial, customer, and field-of-use restraints, would be limited by a dominance threshold, which has yet to be determined. The block exemption would include a list of hardcore restrictions, such as price restrictions and possibly certain territorial restrictions. The Commission notes that the hardcore restrictions would be more limited than under the existing TTBE. The block exemption might also prohibit certain restraints the presence of which would not invalidate the entire licensing agreement.
As to agreements between competitors, the Commission proposes a market share threshold of 25 per cent. The hardcore list of restrictions would include price fixing, output or sales limitations, and territorial or customer allocations. The regulation might also prohibit certain restraints the presence of which would not invalidate the entire licensing agreement. The Commission recommends a more nuanced approach to the following agreements between competitors: pooling arrangements; cross-licensing agreements; licences connected with joint ventures; and restraints that do not concern the IPR itself. Territorial restraints between competitors would receive less protection. In the case of both competitors and non-competitors, the Commission would issue guidelines to be used in assessing situations where the parties exceeded the relevant thresholds.
The Commission’s proposal regarding the use of market shares is problematic in at least two respects. First, there is the usual difficulty of defining relevant markets for the purposes of applying the thresholds. This difficulty is compounded in the case of the TTBE because defining technology markets is notoriously difficult. During the last revision of the TTBE in the mid-1990s, the Commission eventually abandoned the idea of using market share thresholds due to industry opposition sparked by a fear that the introduction of such thresholds would result in an undesirable degree of uncertainty. It is unlikely that the Commission will drop the idea of using thresholds this time around because thresholds have been introduced in the context of the rules on horizontal and vertical agreements, and are gradually gaining acceptance as a means of injecting more economic analysis into the assessment of agreements. To the extent that the Commission decides to keep some or all of the proposed thresholds, the guidelines accompanying the revised TTBE will need to include an extensive discussion of the issues related to the definition of technology markets in order to minimise the uncertainty that the use of thresholds introduces into the analysis.
Second, there are simply too many thresholds. While there may be sound economic reasons to use market share thresholds in analysing the competitive effects of an agreement, this threshold-for-every-occasion approach is questionable, at best. Market shares are but a rough indication of market power because market definition is more of an art than a science and there are other factors that enter into the analysis. Consequently, an approach that applies different rules depending on relatively small differences in market shares would seem to place undue reliance on market shares. It would seem preferable to use a single threshold that provides a safe harbour for agreements between parties that are below that threshold.

Scope of the TTBE
Types of IPRs
The current TTBE applies to patent and know-how licensing agreements; other IPRs, such as copyrights and trademarks, are covered only to the extent that they are ancillary to patent or know-how agreements. Apart from the uncertainty as to which IPRs are ancillary and which are not in an agreement that includes several kinds of IPRs, the current approach means that copyright licenses, trademark licences, and other forms of licence commonly used in business are not covered by the TTBE. The Commission recognises that the issue of coverage is of particular importance for the software industry, which relies on a chain of copyright licences for manufacture and distribution.
This situation creates uncertainty as to the enforceability of restrictions contained in these agreements. For example, a ‘plain vanilla’ trademark licence typically grants the licensee the exclusive right to use the trademark in a particular territory, and imposes restrictions on the licensee’s right to use the trademark outside its territory. As trademark licences fall outside the scope of the TTBE, the enforceability of such restrictions is questionable because they may well fall within the scope of Article 81(1), and, without an exemption, would be unenforceable by virtue of Article 81(2).
The Commission intends to investigate whether other IP agreements, particularly software licence agreements, should be included in a revised regulation. In considering whether to expand the coverage of the TTBE, the Commission will need to consider carefully whether this would make the TTBE too complicated, due to the need to address specific issues raised by the inclusion of other kinds of IPR. The Commission will also need to consider the extent to which copyright and other kinds of IPRs involve licensing practices or raise public policy issues that may make it inappropriate to include them within the scope of the TTBE.
One approach would be to leave these other kinds of IPRs outside of the scope of the TTBE, but to deal with them in guidelines accompanying the TTBE. This approach would allow the Commission to address the unique features of these other kinds of IPRs and provide greater legal certainty for licences of these IPRs than is available under the current regime without undermining the goal of streamlining the TTBE.

Exclusivity
There are two major problems with the TTBE’s treatment of exclusive licences. First, the TTBE only exempts exclusive licences where the exclusivity is linked to a territory, and not when it is linked to a field of use or a customer group (although a field-of-use restriction may be exempted when contained in a licence granting territorial exclusivity). As exclusivity provisions have similar positive and negative effects regardless of whether they address territories, customers, or fields of use, the current TTBE’s emphasis on territorial exclusivity to the exclusion of other kinds of exclusivity is difficult to justify in economic terms. In all three cases, the exclusivity serves to protect the licensee against competition from the licensor or other licensees, which enhances its incentive to accept and exploit a licence. Similarly, in all three cases, the exclusive licence allows the licensor to retain residual rights outside the scope of the licence, which it could exploit itself or license to other licensees, thus giving it greater freedom to divide up its IPRs in a way that makes the most commercial sense. Because it only covers exclusive territorial licences, the TTBE dilutes the incentives that limited licences give companies to engage in innovation because they are unable to extract the maximum value from their inventions.
Second, the TTBE fails to distinguish between exclusive agreements between companies that do not compete and those between competitors. In the case of non-competitors, such an agreement will bring competition to the market that probably would not exist in the absence of the licence. Moreover, the negative effects of such agreements are limited largely to situations in which the licence limits competition between licensees by imposing restrictions on the licensee’s ability to make sales outside of its exclusive territory or where the licensee holds significant market power. In contrast, licence agreements between competitors can raise serious competition concerns. They could lead to market sharing through allocation of territories or customers. In certain cases, such as joint venture licensing, exclusivity may lead to a loss of inter-brand competition and a reduction in innovation by taking away licensees’ incentives to invest in alternatives. Whether the negative effects of an exclusive licence between competitors outweigh the efficiencies generated by the licence depends largely on the market shares of the parties and the structure of the market.
To address these concerns, it appears likely that the Commission will revise the rules on exclusivity to focus more on (i) whether the parties are competitors, and (ii) their position on the market in assessing whether a given arrangement is eligible for exemption, which will mean that the type of exclusivity will become less important to the analysis. This development would seem desirable in that it should make it easier for IP owners to carve up their IP rights among territories, fields of use, and customer groups so as to ensure that they are fully exploited.

Multiparty licences
The TTBE only covers bilateral licence agreements, thereby excluding multiparty agreements such as licensing programmes, multilateral pools, and licence packages. As technology grows increasingly complex, these types of arrangements are becoming more important to industry.
The Commission acknowledges that multilateral licences may be pro-competitive when involving companies that are not competitors. For example, they may allow the parties to bring together complementary assets, avoid infringement actions, clear blocking positions, and reduce transaction costs. At the same time, a multilateral licence could lead to market foreclosure or reduce incentives for independent research and development. The Commission explains that anticompetitive risks are greater when multilateral agreements involve competitors who would have competed on the relevant technology or product market but for the agreement. In such cases, licensing agreements could facilitate price fixing, deter entry, and/or dampen innovation.
If the Commission eventually decides to extend the TTBE to cover certain kinds of multiparty licences, it will have to cope with a procedural difficulty. Under the relevant enabling legislation, Council Regulation No. 19/65, the Commission may not use the block exemption mechanism in the case of licences between more than two parties. Thus, if the Commission wished to extend the scope of the TTBE to cover multiparty licences, it must first either amend the Council Regulation – a time-consuming process that could delay adoption of a revised TTBE by a year or more – or address such licences by means of non-binding guidelines. The solution will undoubtedly depend largely on the input the Commission receives from industry concerning the degree of legal certainty that is deemed to be desirable for this issue.

Specific clauses
Territorial and customer restrictions
The TTBE permits certain restrictions on active and passive sales by licensees into each other’s territories. Restrictions on passive sales by licensees into the territories of other licensees are exempted for a period of five years from the date on which the licensed product is first put on the market within the common market by one of the licensees. The Commission points out that the exemption for restrictions on passive sales is inconsistent with the approach adopted in the Vertical Restraints Block Exemption, which qualifies as ‘hardcore’ any restriction on passive sales in the context of distribution agreements.
The Commission appears inclined to move towards an approach that denies exemption to any restriction on passive sales. It notes that the concept of IPR exhaustion does not furnish a justification for a restriction on passive sales, noting that ‘sales made in response to unsolicited orders imply that the buyer searches and purchases the product in the licensee’s territory, which leads to exhaustion of the licensed IPR, even if such intellectual property is based on national law.’ Furthermore, the Commission questions whether investments incurred by licensees for the use of licensed technology are of such a nature and magnitude as to justify greater restrictions than in the case of distribution agreements. As noted, it would seem possible that licensees generally invest more than distributors because they often must construct manufacturing facilities in addition to incurring costs associated with distribution activities. If this is indeed the case, licensees are perhaps deserving of more territorial protection than distributors.

Output restrictions
The current TTBE does not exempt clauses that restrict quantities that the licensee may manufacture or sell, except in cases where the licence is granted in order to give one of the licensor’s customers a second source of supply or where the licensee is subject to a use restriction. This treatment of output restrictions exemplifies the overly restrictive approach of the TTBE towards limited licences, and its tendency to group licence agreements involving competitors in the same category as those involving non-competitors. As the Commission recognises in its report, quantity restrictions may lead to efficiencies, such as in the case of a licensor with insufficient capacity who finds it more efficient to license the technology to another firm rather than expand its own production capacity.
The Commission notes that quantity restrictions agreed in a licence agreement between competitors are much more likely to raise concerns than those agreed between non-competitors. If a licensor imposes a quantity restriction on a competitor, this could lead to a straightforward output restriction. In the case of quantity restrictions agreed between non-competitors, the Commission states that they could restrict competition to the extent that they restrict the ability of licensees to compete with one another. This concern seems misplaced in that, without the licence, there would have been no competition between licensees in the first place.

Non-compete obligations
The current TTBE contains a broad prohibition of non-compete clauses that applies regardless of the competitive position of the parties or their position on the relevant markets. Moreover, the TTBE does not distinguish between non-compete clauses that prevent licensees from distributing competing products from those that prevent the licensee from engaging in research and development.
The Commission recognises that a more nuanced approach might be preferable. First, non-compete obligations between competitors are generally much more problematic than those between non-competitors. With competitors, the non-compete obligation gives rise to various negative competitive effects such as limiting products on the market and the ability of firms to engage in independent research and development. With non-competitors, the negative effects are more speculative as the main issue is to what extent the non-compete may foreclose competitors from the market, which depends on various factors such as the licensor’s market share, entry barriers, and the duration of the non-compete obligation.
Second, non-compete obligations covering the use of competing technologies or the sale of competing products should not necessarily be treated in the same way as those relating to research and development activities. Preventing a licensee from using competing technologies or from selling competing products can help to prevent free-rider problems and provide greater incentives for the licensee to exploit the technology. The justification for preventing a licensee from engaging in competing research and development is weaker, as the efficiencies are not as obvious as in the foregoing circumstances and such a restriction could inhibit innovation.
Invoking the example of the Vertical Restraints Block Exemption, the Commission finds that when, in the case of licences between non-competitors, the licensor’s market share does not exceed a certain threshold, non-compete restrictions placed on the licensee ‘are not likely to create a significant foreclosure effect on either the technology or product market’. The Commission goes on to conclude that ‘contrary to the current approach, [such restrictions] could generally be covered by the block exemption.’ Thus, the revised TTBE may well carve out some type of exemption for non-compete clauses among non-competitors. As the Commission points out, the issue of market share thresholds will be important in determining the limits of the exemption.

Grant-backs
The current TTBE takes an uneven approach to grant-back obligations. Non-exclusive, reciprocal obligations are exempted, but non-reciprocal obligations on the licensee or exclusive grant backs for severable improvements are not exempted. The TTBE prohibits any obligation on the licensee to assign any improvements or new applications of licensed technology to the licensor.
The Commission recognises that grant-backs offer some economic advantages. They enable parties to share the risks and costs of innovation based on the licensed technology, which can help to promote innovation and subsequent licensing of new technologies. Grant-backs, however, can harm competition if they reduce the licensee’s incentives to develop improved technologies. The Commission notes that exclusive grant-backs are more likely to affect incentives to innovate than non-exclusive grant-backs that allow licensees to exploit their improvements through licences to third parties. The Commission is not certain of the extent to which reciprocal obligations are indispensable to protect the licensee’s incentives to innovate.
Like other provisions of the TTBE, the rules on grant-backs make it difficult for the licensor to control its technology once it grants a license and, thus, as an initial matter, may create a disincentive to grant a licence. Moreover, the current rules do not fit well with licensing practices that are becoming increasingly common. For example, pharmaceutical companies are increasingly licensing technology to small biotech companies for the purpose of allowing the biotech company to carry out specialised research and development on behalf of the pharmaceutical company. The pharmaceutical company will be reluctant to enter into such an arrangement if it runs the risk that it will lose control over its technology in the event that the biotech company makes an improvement to the technology.

No-Challenge Clauses
Clauses that prohibit the licensee from challenging the validity of the licensor’s IPR are currently placed on the gray list, although the licensor may terminate the license in the event of a challenge. This treatment recognizes that no-challenge clauses may be pro-competitive in that they may remove a disincentive to licensing, i.e., the fear that the licensee will challenge the licensor’s rights. As the Commission recognizes in its report, small licensors dealing with large licensees may be particularly concerned about this risk. On the other hand, no-challenge clauses can have negative effects insofar as they could allow licensors to charge royalties for IPRs that are invalid, thus leading to higher prices for the licensed products.
In its report, the Commission suggests that it may treat no-challenge clauses in the same way that they are treated in the R&D Block Exemption, i.e., they would be prohibited, but the licensor would be allowed to terminate the agreement in the event of a challenge.

Conclusion
A revision of the TTBE along the lines suggested in the Commission’s report would be a welcome development. To move away from a legalistic, clause-based approach to IP licenses towards one based on economic analysis would not only serve to bring the TTBE into line with the EU rules on horizontal and vertical restraints, thus injecting a greater degree of coherence and consistency into the EU competition rules, but would also help achieve greater convergence with the treatment of IP licenses in the United States. As IP licenses are increasingly granted on a global basis, such convergence would be a positive development.

Notes
1 Commission Regulation (EC) 240/96 on the application of Article 81(3) of the Treaty to certain categories of technology transfer agreements, OJ (1996) L31/2.