Thursday, September 12, 2013

Day 2 of University of Florida Workshop on Standard Essential Patents and FRAND: Page on "Judging Monopolistic Pricing: F/RAND and Antitrust Injury"

Professor William Page also presented a paper on Microsoft v. Motorola.  Looks at the case through the lens of antitrust injury.  Judge Robart calculated the range of FRAND royalties for two standards.  In form, he conducted a hypothetical bilateral negotiation, but in reality he calculated the royalty in light of the economic consensus of what a FRAND rate should be.  Effectively regulating monopoly pricing.  Generally, in the antitrust context, U.S. courts don't do this; no barometer to determine if prices are too high, and don't want to undermine incentives to lawfully acquire monopoly power.  But courts do calculate monopolistic overcharges attributable to specific unlawful acts, guided by antitrust injury doctrine.  Project counterfactual but-for world unaffected by unlawful conduct.  Motorola was strictly a contract case, but the calculation of the FRAND rate is parallel to the calculation of an illegal overcharge.  Standard setting gives the technologies a degree of exclusivity, which increases as standard becomes more successful.  FRAND intended to avoid monopolistic exploitation of that standardization.  SEP owner entitled only to the increment in value the patent had in relation to the next-best technology; cannot engage in holdup.  Holdup reduces social welfare because it reduces social investments in technologies covered by standards and inhibits adoption of technologically superior standards.  Should also prevent royalty stacking.  In Motorola, Judge Robart's treatment of FRAND was consistent with the economic rationale for enforcing the FRAND commitment.  Microsoft urged the use of the ex ante incremental value.  Judge Robart agreed in principle, but in form he conducted a modified version of the Georgia-Pacific factors, but modified in such a way that it was window dressing; obscures his actual calculation of the FRAND royalty because it took into the need to avoid holdup and stacking and facilitate widespread adoption of the standard.  The circularity is that it assumes he's already established a legal rule that these considerations need to be taken into account.  Judge Robart then focused on certain patent pools that he viewed as comparables.  Considered the quantity and quality of Motorola's patents and concluded they weren't very important; there were good alternatives ex ante.   Considered value of Motorola's patents in light of pooled patents.

So why is this like antitrust injury?  The calculation of FRAND royalties doesn't look like the calculation of contract damages and doesn't serve the same purpose.  The injury is to the class that was overcharged as a result of the ex post facto opportunism.  The point of the FRAND commitment is to deter this holdup; sounds a lot like the goal of antitrust injury doctrine.   Courts do try to calculate antitrust damages consisting of monopoly overcharge above what would have been the case but for the violation.  Not a perfect analogy, but it identifies a degree of monopoly power attributable to the unlawful conduct.  The ex ante standard reflects a before-and-after model of antitrust damages.  Judge Robart did something similar, using the patent pool.  Patent pool as reflecting the basic goals of the standard and trying to exclude the holdup and Cournot complements issues.

Failure of SSO to constrain monopoly exploitation of the standard could be an antitrust violation?

Standardization aggravates the Cournot complements problem by creating complementariness.  The additional output restriction attributable to that is analogous to antitrust injury.

My questions:  First, a technical point.  You said that "The calculation of FRAND royalties doesn't look like the calculation of contract damages and doesn't serve the same purpose."  Just to be clear, Judge Robart wasn't assessing contract damages.  Microsoft's damages were the losses it suffered as a result of having to move some of its supply centers, etc. as a result of Motorola's failure to bargain to the FRAND royalty the judge had calculated.  The jury assessed those damages last week. 

Second, you say that Judge Robart didn't really apply a hypothetical negotiation framework, because he took into account the need to avoid holdup and royalty stacking and the desire for widespread adoption, and thus imposed something of a normative overlay on the Georgia-Pacific framework.  I'm not sure I agree.  Ex ante, before the standard is chosen, the user would not agree to the holdup rate or to a rate that would threaten to cause royalty stacking.  So it is appropriate for the court to take that consideration into account because it is a factor that real-world parties would have considered. 

And while ex ante incremental value is the economically correct touchstone, we always have to trade off some degree of accuracy for expediency.  Maybe, given the unlikelihood that the patent incentive is very robust in this space and the fact that widespead adoption leads to substantial aggregate royalties, we can afford to trade off some accuracy for expediency here. 

Page:  There is a sort of rhetorical value in not adopting the ex ante incremental approach  because it would make it more difficult to use the pool as a comparable.

Haw:  Does this mean that patent pools are a safe harbor?

Page:  Contreras paper suggests that SSOs should encourage the formation of patent pools.  But what Robart did doesn't strike me as anything like what would go in a negotiation.

Blair:  If you join a patent pool, what determines the royalty you would earn?

Page:  In these pools, it has nothing to do with the value of your patent, but rather on the number of patents you own in the pool.

Yoo:  But it doesn't have to be that way.  The reason the SSOs don't form patent pools is to allow private bargaining.  Pools avoid Cournot complements problem by setting the price, but can still be a mess because high-value patents may not join.

Haw:  It's an adverse selection problem.

Blair:  With pools, we're just shifting the problem to a different location.

Yoo:  And it depends on the internal governance of the pool or SSO.  The Allied Tube problem.

Page:  A big part of it is that with pools, the licensors often are also licensees.  Robart highlighted this.  Microsoft valued access to the patents in the pool more than the royalties it derived from the pool.

Yoo:  This suggests we want people on both sides of the transaction.  Uses Comcast/NBC merger example.

Blair:  If there are multiple claimants, though, you still have to figure out allocation of surplus.

Page:  The other form of integration is ASCAP/BMI. 

Carlton:  The problem is that ASCAP and BMI are entities that divvy up the royalties.  The rates they get to set are set by some judge in New York.  A bizarre regulatory situation.

Yoo:  That's the liability rule side of it.

Carlton:  A relevant question is, if you think that's a solution, do you want a regulatory agency or a court to do it?

Yoo:  BMI is about substitutes, not complements.  Putting substitutes in a patent pool is a problem from an antitrust context.   Technically, patent pools are to solve vertical holdup.  BMI serves a different holdup problem.

Blair:   Look at the economies of scale with BMI.  Blanket licenses economize on transaction costs.

Carlton:  But now they can count actual uses on radio.  Music publishers are beginning to withdraw from ASCAP and BMI because they can do the monitoring themselves.  Only purpose of ASCAP and BMI is to provide an economy of scale for being a collection agency for composers.

Yoo:  Monitoring issues are technologically contingent, but holdup isn't.  



No comments:

Post a Comment