Relaxing whilst doing Competition Law is not an Oxymoron

HLS Seminar discussion on refusals to licence.

with 6 comments

Continuing with the series of posts written by Harvard Law students for the seminar on Antitrust, Technology and Innovation, you can find part of the discussion on refusals to licence in the comments to this post.

As explained earlier, some of these comments refer to the readings available in the syllabus.

Written by Alfonso Lamadrid

16 June 2010 at 11:46 am

Posted in Uncategorized

6 Responses

Subscribe to comments with RSS.

  1. Here are a few questions:

    * There seem to be two types of innovation that are of concern here. One is innovation at the primary level of the technology that is the target of the demand for licensing. The other is innovation at the secondary level or market which uses the technology as an input. Is there a trade-off between the two levels from the aspect of innovation? In other words, by facilitating rivals’ innovation at the secondary level, are we inadvertenly dampening their (rivals’) incentives to innovate at the primary level (e.g., inventing around the dominant firm’s IP or creating alternative / substitute technologies) and herding them to compete within the boundaries of the dominant firm’s underlying technology? Is this an optimal result?

    * There is another related, perhaps more fundamental question. Refusal to license generally involves (often patented) “technology” or “know-how”. Does the “technology” factor change our outlook on the relevant doctrine? As some have noted, the targeted assets (facilities) in previous essential facility cases like Aspen, Otter Tail or Trinko were virtually impossible to replicate in a physical sense (e.g., ski slope) or an economic sense (e.g., electric power system, network infrastructure). For example, having multiple competing power systems that are costly to build, manage and operate may not be efficient, particularly if the market exhibits characteristics of a natural monopoly. This fact contributed to the formation of a doctrine based on the “essential” or “indispensible” nature of the asset being targeted.

    o But does the same always apply to “technology”? Is it unreasonable to require rivals to invent around the patented technology or create alternatives that can substitute the “technology” rather than granting them with access to the “technology”? When would such an undertaking become so infeasible, thereby making the “technology” indispensible in the sense of Aspen, and thus warrant antitrust to step in and force access to rivals (Some might arguably point to the Microsoft case as coming close to such a situation. See CFI Decision Para. 684. See also Glazer, 1203 regarding ECJ ruling in IMS)? In this regard, do Otter Tail / Aspen / Trinko provide the right guidance for refusal to license cases compared to other traditional refusal to deal cases?

    Written by: Yong. L.

    Alfonso Lamadrid

    16 June 2010 at 11:48 am

  2. I really like Yong’s distinction between innovation in the primary essential IP element and the secondary collaborative element. He’s right to point to a possible tradeoff between the two. Should antitrust enforcement want to strike a different balance in “new economy” markets than in more traditional markets, such as the pharmaceutical industry? I would think that characteristics of tech markets such as network effects would lead to a greater emphasis on rewarding collaborative innovation.

    How useful is the subjective intent test that nearly all of the DOJ/FTC panel opposed? The test is difficult to administer, as it may be difficult to reliably measure a business’s intent based on employee statements. However, the Supreme Court seemed to draw a conclusion about the defendant’s subjective intent in the Aspen Skiing case, arguing that the defendant’s “unwillingness to renew the ticket even if compensated at retail price revealed a distinctly anticompetitive bent.” The existence of a “previous relationship” does seem to make it easier to determine subjective intent, as in Aspen Skiing. Also, I thought it was funny that one of the DOJ/FTC panelists said that the subjective intent test “makes it difficult to explain to business people how to ensure their activities are lawful” on pg. 13. It seems like if regulators are just looking for subjective intent, a smart and unethical general counsel would just advise businesses to document and talk about legitimate-seeming business justifications to mask any anticompetitive motive.


    16 June 2010 at 11:49 am

  3. As a former student of bioethics, the first thing that caught my attention was the discussion in Skitol of whether there is/should be a valid distinction between (1) rejecting overtures to cooperate/share an essential facility and (2) ceasing a prior course of cooperation. There is a suggestion that the Sherman Act recognizes a distinction between stopping and not starting, which sounds intuitively plausible but as Posner points out may be difficult to cash out and leads to perverse incentives. There is a very similar debate in bioethics wrt the removal of life sustaining treatment. Does this constitute killing a terminal patient, rather than allowing them to die? Is it worse to try such a machine, and then turn it off when it becomes clear that recovery is impossible, than to never try in the first place? I wonder if any of the moral principles and intuitions which inform that debate can be transplanted (sorry, couldn’t resist) into the Aspen context?

    Paul B.

    16 June 2010 at 11:50 am

  4. I think two main points you made are very interesting and reveal the complicity of interrelation between intellectual property rights and antitrust issues. My answer to your second question, whether it is unreasonable to allow right holder to refuse to license, would be very lawyery: “It depends”. I think there will never be a “one size fits all” answer to the question whether rivals should be granted access to “the” technology or could be directly refused to have access. It really depends on the features of the particular technology and of the possible innovations in the secondary level. So, if the purpose is to increase social welfare by establishing a balance between allocative efficiency role of antitrust and promotion of innovation, as elegantly explained in Mackie-Mason’s article, critical questions would be how much secondary innovation the use of first level innovation would bring, what the welfare effect of these secondary innovation will be (so I am taking a more qualitative approach than a quantitative one) and how “essential” is the dependence of this secondary innovation to the use of first level innovation. These questions are not easy to answer in the frame of an ex ante review but essential in determining anticompetitive effects of particular refusals to license.

    A last point that I would like to make is that readings reminded me of the multi-dimensionality of competition and its complex effects. In our evaluation of anticompetitive effects of refusals to license, we should be concerned not only about the potential decrease of secondary innovation efforts of rivals but also potential decrease in the innovative efforts of the right holder itself. If we apply an inflexible rule that allows all refusals to license, intellectual property owner might also loose its incentives to innovate more, and enhance its product. Continuing competitive pressure is what makes the innovation go further.


    16 June 2010 at 11:50 am

    • I found your 2nd question very difficult to answer. Like Selen, I have to say “it depends”. Nonetheless, we could generalize by highlighting some factors which make techology essential. Following are my comments, some of which are not related to other pps comments, so I apologize in advance.

      1. Compulsory licensing under antitrust law and patent law

      In relation to the discussion in Trinko, I am wondering how remedies in each
      legal framework work out. More specifically, I am curious how complusory licensing under antitrust law should differ from complusory licensing under patent law. Which law should we use to impose compulsory licensing to minimize negative impact on innovation? Is this just a matter of form? or more substantial question?

      I understand that there is a compulsory licensing as a remedy in the patent law,
      but if I happen to be wrong, please correct me.

      2. Innovation and Patent

      Selen’s comment on innovation and Mackie-Mason’s piece (and some argument in Damien Geradin’s piece) reminds me of issue relating to innovation and patent.
      When innovation is discussed, it seems that people focus on incentive
      of rightholder or potential rightholder to further innovate. There is nothing
      wrong with it. But, I think there are other dimensions that we might want to
      consider. First, at least in theory, strong patent encourage inventors to file invented
      technical information. This disclosure gives public access to information so that
      other people can use such information to further innovate. Second, whilst patent
      holders have exclusive right, other people cannot effectively use such patented
      technology, and this might have negative impact on incentive of other people to
      further innovate. If we think about design of legal framework from viewpoint of economist as Mackie Mason does, we have to include these two dimensions
      into economic model or formula. I thought this point should have mentioned
      in his piece, but do you agree?

      3. The-fact Refusal to deal

      Some of us talk about how difficult to give advice on this issue as a lawyer. In this connection, although this might be basic, but I am wondering the extent to which those in possession of essential facilities can demand extremely high price instead of merely refusing it. Unless it is clearly obvious, it might be a great way to lessen legal risk of being sued under section 2. Am I right?

      4. Google’s data base will be an essential facility?

      This is speculative, and I am writing this rather for fun.
      I am wondering whether huge data base of Google might be regarded as essential
      facility in the future. Let’s say that Bing is out of the market, and Google becomes dominant. Then, new company comes up and asks Google to let it use some data base so that it can operate search business. Can Google always refuse such request? I think it is not entirely implausible.

      Takashige Yamada

      16 June 2010 at 11:51 am

      • First, how worried are we about affecting the dominant firm’s incentive to innovate at the primary level? The way I see it, there are strong reasons that this shouldn’t concern courts (although I am sure there is a good counter-example out there that somebody could provide). Thinking about the Microsoft EU case and namely the server-side aspect of it, I tend not to be concerned. First, as the court points out, this doesn’t dampen overall incentive to innovate the primary product (PC OS), just perhaps the interface with the secondary product (server OS). Even then, the disincentive to innovate only occurs because the monopolist is thinking about two markets at once. Otherwise, it would have incentive to innovate its interface capabilities so that servers easily work with it so that it becomes even more dominant in the primary market (or at least maintains dominance). It is only not doing so because it wanted to capture the benefits of that innovation at the server level (or, at both levels, depending on your views regarding law of one monopoly price). However, if the dominant firm got monopolies at both levels, that would also seem to reduce its incentives to innovate, meaning that either way you could argue it won’t be innovating more. Moreover, how much leeway do firms have to make decisions about primary products based not on the most competitive choice for that product but on the optimal strategy for secondary products? They own both, so I don’t see why they can’t do what they want, but if we’re worried about consumer welfare, then that might be a concern.

        On a related note, how do we even feel about this primary/secondary distinction? On the one hand, it seems both intuitive and helpful. My thoughts on the last example are clearly shaped by my feeling that those were different products. But there are surely plenty of cases where it is very hard to tell whether there are two levels or one. To what extent is a very raw input a separate product (especially if it can’t be used by itself). For example, what if the chemical in Commercial Solvents (I have no idea if this is true), was useless on its own but only good as part of the mixture? What if it was only good as part of the final product and also was not an input into any other potential product? What if that were true and someone invented a new product that needed that as an input (does it suddenly become a new product and change the rules regarding what the monopolist’s refusal to deal in the case of a competitor who wants to make the old, original product)? On the one hand, this concept seems great, and although I haven’t figured out exactly where I stand, I think a lot of how I feel about the case might depend on the relationship between primary and secondary products. On the other hand, I’m not sure the divide is as clean as it seems, so maybe there is a real problem with relying on it.


        16 June 2010 at 11:52 am

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: