Comments for FTC Hearings on Damages, Injunctions and the IP Marketplace

 

In May, 2009, the Federal Trade Commission held hearings on a number of patent issues.  Mr. Schlicher testified at those hearings and submitted written comments on patent damages, permanent injunctions, the Supreme Court’s decisions Medimmune v. Genentech and Quanta v. LG, the effects of uncertainty, and “transparency.”   This is parts of the summary to his comments.

I.  Summary

I offer these comments in response to the FTC’s notice.  I comment on the subjects of patent damages, injunctions, two Supreme Court decisions, the effects of uncertainty, and “transparency” though I am not sure what that means.

A.  Four General Comments

First, the notice asks whether patent owners are being systematically over-compensated or under-compensated.  The notice identifies the harm of over-compensation being that “supra-competitive prices for technology would unduly dampen future innovation and prices for products incorporating patented inventions would increase unjustifiably.”  I do not know what is a “supra-competitive price” for a patented invention or an “unjustifiably high” price for a patented product.  Patent law leaves the prices of patented products to the market.  High prices for patented products are a sign the patent system is working, not a cause for concern.

Second, the goal of patent law is not simply to provide patent owners with “compensation” for use of their inventions.  If “compensation” were the goal of patent law, the Patent Act would provide that anyone may use a patented invention subject only to liability for damages providing compensation.  Patent law does not do that.  The Patent Act provides for injunctions to prevent violations of the rights and damages to compensate for infringement.  The reason for exclusive rights and injunctions is to enable a patent owner to control who uses an invention and how they use it and to require anyone who wishes to use an invention to buy the right to do so at prices and on other terns the patent owner and the user agree are mutually advantageous.  One benefit of that approach is that the value of inventions is determined by market transactions, not by damages rules and judicial determinations of value.

Third, the notice asks about the extent to which available remedies influence agreements to settle patent infringement actions and license patents.  The patent system relies on agreements between patent owners and producers to determine how inventions are used and the value that patent owners capture from commercial uses of their inventions.  Those contracts and agreements are a far more efficient (that is better and less expensive) mechanism for the decision of patent issues and the transfer of revenue from producers to patent owners than administrative or judicial proceedings by the government.  For this reason, an important test, if not the most important test, for determining the desirability of a particular legal rule is the extent to which it facilitates or interferes with agreements between patent owners and producers involving use of patented inventions.  There are many legal rules that interfere unnecessarily with those agreements between patent owners and producers.  These are the rules the government should change.  The Patent Reform Acts ignore them.

Fourth, this is not the best time to legislate changes to patent law.  The costs of patents are always in the present.  The benefits of patent rights are in the future and for many important inventions in the far distant future.  Under normal economic conditions, there is a tendency for people in elective office to be more concerned about the effects of law in the next one to three years than in later years.  Under current conditions, that tendency is even more pronounced.  For that reason, the government should be approach legislative changes to patent law with great caution.

B.  Permanent Injunctions

The notice mentions the Supreme Court’s decision eBay and asks for comments about when it makes economic sense for a court to issue a permanent injunction and what a court should do when it does not make sense.  The eBay decision has caused confusion about how judges are to decide whether to issue permanent injunctions, particularly in cases involving a patent owner which exploits an invention entirely or in part by licensing.  For such a patent owner, the question is whether an injunction should issue even though it would be in the economic interests of the patent owner to license an infringer.  eBay did not decide that issue.  Subject to a special class of situations, the answer is that a patent owner’s willingness to license some infringer is not a reason to deny an injunction.  It is the reason to grant an injunction.

eBay means that the courts must apply the principles of equity in determining whether to grant an injunction and may not apply a general rule that an injunction should issue if the patent owner prevails.  Unfortunately, the Court simply referred the lower courts to the traditional four-part test for deciding whether to grant patent injunctions.  When applied to patent actions, the traditional test is too vague to provide sensible guidance.  The test focuses primarily on the effect of an injunction on how one existing invention will be used in the future and the effects of that use on the parties and the public.  Such a test is inappropriate to a well-functioning patent system.  The decision to grant or deny an injunction for one patent in one action will affect the future expected value of patents to other inventors of other inventions.  The traditional test may easily be read to make this vital consideration irrelevant.

The traditional test also assumes that future use of the invention will be controlled by the injunction.  However, a patent owner may waive its rights by granting a license, even after an injunction issues.  The patent system is designed so that decisions by patent owners and users of inventions govern by whom and how inventions are used and at what prices.  Only injunctions provide those people with the right incentives to make those decisions.  The patent system is not designed to require judges to make those decisions.  If judges think they have that responsibly (as the traditional test may be understood to imply), judges become the government regulators of patent licensing, selecting what companies use inventions, how inventions are used, and the prices for use.  The choice between private control of licensing and judicial control should be easy.  As Judge Easterbrook once observed,

“A private outcome of these negotiations [between a patent owner and an infringer after an injunction issues] - whether they end in a license at a particular royalty or in the exclusion of an infringer from the market - is much preferable to a judicial guesstimate about what a royalty should be.  The actual market beats judicial attempts to mimic the market every time, making injunctions the normal and preferred remedy.”

The law should be that an injunction is the preferred remedy against patent infringement.  There are five reasons.  I mentioned one, to enable inventions to be used in ways and at prices that reflect private decisions and market transactions.  The others are that injunctions provide patent owners with control over use of inventions (so that inventions have greater value than they otherwise would), reduce the resources wasted due to infringement, reduce litigation costs, and limit the effect of deficiencies in current damages law.

While each reason is important, the reason listed last could be the most significant.  If injunctions are denied, the value of a patent depends entirely on whether the law of damages permits patent owners to capture amounts equal to the economic value of their inventions.  If injunctions are denied, the ability and incentives of patent owners and invention users to do business in patent rights depends entirely on the law of damages.  The law of damages does not consistently and predictably result in awards equal to the economic value of inventions.  Hence, denying injunctions undermines the desired effects of patent rights by increasing the importance of damages.  If damages fail, the system fails.

This is one example.  It is likely that the most frequently used measure of damages is the hypothetical negotiation approach.  Damages are measured by what a patent owner and an infringer would likely have agreed on in a hypothetical negotiation for a license.  It is implicit in this approach that the infringer will not be able to use the invention without first reaching agreement with the patent owner due to injunctive relief.  Since the infringer in the hypothetical negotiation may not use the invention without an agreement, it will pay up to the economic value of the invention for a license.  However, if the law is that injunctions will not issue in some category of situations, the hypothetical negotiation measure of damages is impossible to apply.  In that category of situations, since the infringer may use the invention without an agreement, it will pay for a license up to the amount of the damages award.

What is the amount of the damages award?  It is the amount likely to be agreed on in a hypothetical negotiation.  However, the hypothetical negotiation cannot produce an agreed amount unless the amount of damages is determined in some other way.  Under this state of affairs, damages may not be determined without knowing the result of a hypothetical negotiation and the result of a hypothetical negotiation may not be determined without knowing damages.  Under this state of affairs, patent owners and producers will have no idea what to do.  The amount of damages will be almost completely random.  As a result, in the categories of situations where injunctions are denied, licensing will become virtually impossible.  There will be no way to agree on a price because there will be no way to determine the amount of likely damages if there is no agreement.  Over time, patents in those situations could easily come to have only a small fraction of their previous value because the remedies for patent infringement will have almost completely failed.

The difficult issue is whether an injunction should be denied where it is clear that it is in the patent owner’s interest to grant the infringer a license and in the infringer’s interest to operate under a license and there is good reason to believe an injunction would distort the amount of the royalty in an undesirable way.  The potential distortion is that an injunction would permit a patent owner to negotiate a higher royalty due to an infringer’s product-specific investments and this would adversely impact the incentives of patent owners and producers in the future.  I do not believe there is an easy answer to this problem.  At present, I suggest the following rule.

An injunction should be granted unless: (1) it is without doubt in the patent owner’s business interest to grant the infringer a license and in the infringer’s interest to accept a license at a payment rate equal the economic value of the invention when used by the infringer and there is no way the patent owner could make more money if it or someone else used the invention and the infringer did not; (2) the infringer made large investments because it was necessary to do so to produce any product and not merely to produce the patented product, those investments are so large that they must be recovered (with an adequate return on investment) from sales revenue over a significant period of time, and those investments are large relative to the value of the patented invention; (3) it is clear that the payments made to the patent owner during the period no injunction is in place will be equal to the full economic value of the invention at the time of sales (and not based on the damages award rate); (4) there is no evidence or reason to believe the infringer made the investments to prevent an injunction; and (5) the period of denial is no longer than absolutely necessary.  The same general approach could apply to the related problem of products subject to large demand side scale economies.

C.  Damages

Damages are an issue due to the patent reform bills in the Congress.  In my view, the proposed legislation on damages in the Patent Reform Act is undesirable.  The Patent Reform Act proposal is based on the view that damage awards are too high, when based on compensation not less than a reasonable royalty.  While there are occasionally cases in which it seems plain that the damage awards are larger than the economic value of some invention, it is not possible say that is happening in a significant percentage of cases.  That said, much of the law on patent damages obscures the effort to match damage awards to the economic values of inventions.

Much of the law governing compensation not less than a reasonable royalty is almost useless in arriving at an award in a particular case that approximates the true economic value of an invention that a patent infringer captured or the patent owner lost as a consequence of the infringement.  I describe those problems and the answers later.  The proposal in the Patent Reform Act would not improve those legal standards.  Those proposed changes would in some ways lead to awards that are too high and in other ways to awards that are too low.  In all cases, the changes would lead to enormous uncertainty that would take years and probably decades to resolve.

Most importantly, the damage proposal in that version of the Patent Reform Act would prevent the adoption of a standard for determining damages in that way that would result in awards approximating the economic value of some invention.  The proper way to determine that value is known and courts could easily correct the problem.  With some refinements, reasonable royalty damages should be the difference between the net profits the infringer earned from sales of the infringing product and net profits it could have earned using the next best non-infringing substitute available to it during the period of infringement.  The economic value of the invention is this difference in profits.  This approach to reasonable royalty damages solves the problem of patents to changes in existing products.  This approach is entirely consistent with the origins and purpose of reasonable royalty damages.  This approach is one feature of the existing law on reasonable royalty damages and fails to lead to sensible damages awards in all cases because other features of the law obscure its significance.  This approach has been applied by the courts in many cases and all that is needed is for the courts to require its use in all cases in place of the present amorphous approach.  Congress should not act until the courts have shown an unwillingness to do so.

D.  Supreme Court - Medimmune v. Genentech and Quanta v. LG

The Supreme Court has rendered two decisions in recent years that directly affect the value of patents.  The ultimate effects of the first of those decisions, Medimmune v. Genentech, are unclear because the reach of that decision is unclear.

The Medimmune decision may simply mean that a patent licensee may commence a declaratory judgment action seeking judgment that the patent is invalid where a patent owner grants a license that by its terms calls for royalty payments on products that are covered by a “valid claim” of the licensed patent, this contract provision means that validity is a condition to the payment obligation, the licensee has informed the patent owner that it is protesting payment of royalties because it believes a claim is invalid, and the licensee has a reason to believe that, if it ceased royalty payments, the patent owner would terminate the license and sue for infringement.

If Medimmune is confined to that particular situation, the value of many patents previously licensed under this definition of royalty bearing products will decline.  There is no particular reason for the federal government to intervene on their behalf at this point, since they elected to use that royalty provision.  If Medimmune is confined to that situation, patent owners may avoid the problem in the future by licensing under different terms.

Without prejudging the matter, there is a significant risk that Medimmune will be found, incorrectly, to apply to all licenses.  If that happens or even if patent owners believe it will happen and other contract terms to prevent or deter validity litigation fail, patents will be licensed only in situations where a patent owner has no option other than licensing or a patent is highly likely to be found valid by a court.  If that happens, the number of licensing transactions will decline, the revenue that patent owners earn under licenses will decline, and costs of litigation between patent owners, licensees, and infringers will increase, reducing the ultimate returns to inventing.

The Supreme Court decision’s is simply the latest in a rather sad saga that began in 1969.  I believe legislation should be introduced to deal with the legal rules that constrain the ability of patent owners and their licensees to agree to and enforce terms regarding royalty payments and the litigation regarding the validity of patents that would reduce uncertainty about whether licensees will pay those royalties and reduce the litigation costs that the parties must bear under licenses.

The Supreme Court’s other decision is Quanta v. LG.  The Supreme Court did not affirm the exhaustion doctrine.  The Court changed the exhaustion doctrine and the relationship between the exhaustion doctrine and implied license doctrine.  The Supreme Court did not find that exhaustion applied where a patent license purported to limit the rights transferred to purchasers of a patented product.  The Supreme Court found that exhaustion applied in that case because the provision of the license that said something about the rights of a purchaser also said that the provision did not operate to limit the operation of the exhaustion doctrine.  The Court reached its decision on exhaustion because the agreement expressly said that the parties did not intend to limit the exhaustion doctrine.

One novel and unfortunate feature of this decision is that the Court found that exhaustion applied to the sale of certain unpatented products and therefore effectively eliminated the implied licensed doctrine.  Another novel and unfortunate feature of this decision is the Court’s pronouncement that all post-sale restrictions were invalid.  If the Court’s decision is deemed to apply to situations other than that in the case before it and the Court’s statement that all post-sale restrictions are invalid is taken literally, patent owners will capture a much smaller part of the value of their inventions, there will be less licensing, and there will be less use of patented inventions as methods of selling and licensing that previously increased use of those inventions are no longer used.

E.  Uncertainty

Uncertainty regarding the validity and scope of patents has a significant effect on patent transactions.  Unnecessary uncertainty of the substantive law of patents undoubtedly reduces the value of patented inventions and increases the cost of products and services not introduced due to this uncertainty.  To the extent the substantive law could be change in ways that would reduce uncertainty without diminishing proper incentives to make inventions, the substantive law should be changed.

The Patent Reform Act will not reduce overall uncertainty; it will increase uncertainty.  The principle claim of the patent reformers is that the Patent Reform Act will reduce uncertainty by awarding a patent to the first inventor to file a patent application rather than the first inventor to make an invention.  Even if the claim is correct and the change has no other effects (and it will), the Patent Reform Act creates new uncertainty by changing the definition of the activities that constitute prior art, changing the scope of prior art activities from those that occur in the United States to those that take place anywhere in the world, requiring that the nonobviusness of an invention to an ordinarily skilled person be judged as of the time a patent application is filed and not at the earlier time when it was actually made (when the problem may have been more difficult and the information and technological principles helpful in solving it less useful), broadening so-called prior user rights, and making other changes.

The Patent Reform Act also ignores the issues on which legislation could helpfully reduce uncertainty, such as redefining the public use and on sale doctrines so inventors trip over them with less frequency, eliminating means-plus-function claims, eliminating the doctrine of equivalents or adopting a more well-defined doctrine of equivalents, eliminating the misuse defense or at least making it co-extensive with the antitrust limits on exploiting patents, redefining the law the courts have developed after Lear and will develop after Medimmune so that patent owners and licensees may provide royalties that sensibly reflect the value of an invention in view of legal uncertainties and that eliminate the possibility of needless and expensive litigation, and making many other desirable changes.

However given whatever substantive law exists at any point in time, patent owners and potential licensees should be permitted to adjust the price and other terms on which they do business in the face of that uncertainty.  The price adjustment in any particular situation will depend on the owner’s and licensee’s views of the probability a court would decide a certain issue in a certain way and perhaps even more importantly on the owner’s and licensee’s aversion to risk or, if it exists, preference for risk.  Whatever the outcome in a particular situation, the law should encourage them to make those agreements by enforcing their terms.

The law should not force patent owners and potential licensees to resolve the uncertainty by seeking assistance from the Patent and Trademark Office or the federal courts.  Private agreements between patent owners and potential licensees will reduce the costs of uncertainty and facilitate the use of patented inventions in spite of that uncertainty far more efficiently and quickly than proceedings in which the Patent and Trademark Office or the federal courts.  For that reason, I do not expect the various proposals in the Patent Reform Act for repeated Patent and Trademark Office review of patentability to yield significant benefits and rather expect it primarily to yield increased costs and delay.  Our prior experience with various forms of reexamination should have taught us about the limited value of repeated administrative review.

f.  Transparency

The notice asks about “transparency” in the market for intellectual property rights.  Transparency could mean several things.  Since I am not sure what the FTC means by transparency, I am not sure I have useful comments.  Transactions between patent owners and their licensees did not contribute to the current economic problems facing the United States that have prompted talk about transparency.  Transactions in intellectual property rights are private contracts between patent owners and producers of goods and services.  These citizens make contracts regarding rights granted by the federal government.  However, except for the recipients of government research subsidies, the people owning those rights spent their time and their money to create the underlying inventions and will bear the costs of operating under these contracts.

There is no legitimate government interest in forcing those individuals and companies to disclose the existence or the terms of those agreements to others, including the federal government, except to the extent that those disclosures are necessary for purposes of the securities laws or legitimate inquiries in connection with antitrust, tax, or other similar matters.  The existence and terms of license agreements should not be disclosed to the public to enhance transparency or to permit the government or people who are not parties to particular transactions to learn of their terms.  People and companies have a variety of reasons for keeping this information secret, including the fact that the confidentiality of contract terms often enhances the ability of the parties to reach an agreement on terms under which each party benefits.

Some people believe that the market in which patent rights are bought and sold would work better if sellers and buyers were required to disclose the prices of their transactions.  Nothing would be gained by such a legal requirement.  The only people who need to know the price of a particular transaction are the people who made the transaction.  They are the only people affected by the price.  If they chose for any reason to keep the price confidential, that is entirely their business.  If the law overrides their desire to keep the price confidential, the law defeats whatever purpose was served by keeping this information secret.

A government requirement of public disclosure is not a neutral act.  A disclosure requirement would change the terms on which people are willing the deal and reduce the number of such contracts, because disclosure defeats the interests of a party or parties served by confidentiality.  In particular, if the government required public disclosure of the price terms of licenses, the prices at which patent owner and potential licensees do business will change, the number of licenses will decline, and the profitability of licensing will be reduced.

 

JOHN W. SCHLICHER

PATENTS, PATENT LITIGATION, PATENT DISPUTE RESOLUTION AND

SETTLEMENT, LICENSING, ANTITRUST, LAW AND ECONOMICS