- Sitting by Designation: The CAFC has continued to offer district court judges from across the country the opportunity to sit on the appellate bench for a day. On June 3, 2008, Judge Ward from the Eastern District of Texas filled out a panel with Judges Newman and Gajarsa. They will be deciding three patent cases: Hyatt v. Dudas (waiver of arguments before the BPAI); Atlanta Pharma v. Teva (appeal from denial of preliminary injunction to stop teva from selling a generic version of the ulcer med Protonix); and D Beam v. Roller Derby.
- New USPTO Fees: They are going up, but you can put in your two cents by July 3. [New Fees][Contact Walter.Schlueter@uspto.gov. Include RIN number RIN 0651–AC21 in the subject line.]
- FTC Commissioner Rosch recently spoke about patents and antitrust. He believes that the FTC Act, the Clayton Act and the Sherman Act all create “viable enforcement tools” against companies that create “patent walls” — especially when they acquire those patents from third parties. The FTC is currently going after N-Data in a case where the patent covers an ethernet standard. [N-Data][Zura’s Comments] Three important cases on this issue:
- US v. Singer Mfg (Antitrust violation to charge competitors with patent infringement after U.S. company obtained patent from Swiss company)
- Kobe v. Dempsey Pump (obtaining and using “every important patent” in the field in order to exclude competition, together with other anticompetitive activity, constitutes an antitrust violation)
- Case of Xerox Corp (settlement consent decree where FTC challenged Xerox’s purchase of additional plain paper copier patents).
- Pre-Purchase Review: There is some room to debate here on the question of organic-patenting (patents on inventions via the company’s own R&D) versus acquisitional-patenting (buying up patents invented by others). At the firm level, acquisitional growth is typically reviewed much more harshly for potential antitrust violations than is organic growth. Is the FTC/DOJ headed toward a system of pre-purchase review of major patent acquisitions in the same way that the agencies conduct pre-merger reviews?
Re the note about FTC Commissioner Rosch recent speaking about illegal “patent walls” especially when they acquire those patents from third parties. [The FTC is currently going after N-Data in a case where the patent covers an Ethernet standard.] “Three important cases” are cited on this issue: US v. Singer Mfg, Kobe v. Dempsey Pump and Case of Xerox Corp (settlement consent decree).
An uncontested settlement consent decree is something only the FTC and DOJ have the chutzpah to cite as if were legal authority. [It is not a decision by any court or judge.] If a law firm tried to cite one of their own complaints as legal authority they would be bounced down the courthouse steps. {Many consent judgments are entered into simply to avoid legal expenses, adverse publicity, jury prejudice in having Uncle Sam as the plaintiff, etc.} This is amply demonstrated by the fact that when the same FTC “patent thicket” allegations against Xerox were repeated in a civil suit by SCM in a REAL case they were rejected, by the Appellate Court! (2d Cir. 1981).
Leo,
Reason is a libertarian magazine. “Free Markets and Free Minds.” Socially, I am 100% there with them. But they also swallow all that deregulation stuff, which I partially disagree with. Regulated “free markets” work.
Trust, sure the buyer would like to buy an exclusive license, but would the seller want to sell? I can sell my patent to 20 competitors for $100 each. What is it worth to one buyer in order to exclude the competition and extract higher rents? Anything over $2,000 is a monopoly premium. Of course it is worth more, but how much more? The seller may never be confident that he is getting full value; the buyer may be unwilling to pay a premium for an untested product.
Instead, sellers and buyers may hedge their bets through nonexclusive licenses. The price and risk will be much lower. (There may also be an anti-monopoly premium to prevent accumulation of market power, but we are getting much too complicated).
My basic argument is that there is no difference in obtaining market power via patent purchase or issuance. And to be a contrarian, I suggested that market forces in a purchase would provide at least some check on market power while there is no check at all on a patent issuance.
Anti, you are correct that the anti-anticompetitive market forces would only work with a super patent. A super patent is, of course, silly. Generally no one knows whether a patent will be a super patent. One exception is for a standard set by an SSO which is why these are more highly scrutinized (SSOs are interesting because they represent anti-Coasian a priori collective decision making)
Our concern then is a “patent wall” where we admit that each patent may not be valuable but the aggregate allows for the same type of supracompetitive or monopolistic rate setting as a super patent. For a patent wall, we must accept that at some point, a single patent added to a portfolio will tip the scale from competitive to anticompetitive market power. This single patent has a high marginal value. But since we do not know which patent purchase will do this, we must monitor every patent purchase, including purchases of admittedly nonvaluable patents.
How will we monitor? A privately held hedge fund is sucking up a lot of patents and not recording any assignments. No one knows what is in their portfolio. The FTC will need some authority to require the public disclosure of each purchase. Note that (almost) all patents are public, so Commissioner Rosch’s concern about hiding a patent in one’s pocket is nonsensical.
Let’s say we know a firm’s “patent wall,” then what? Can we determine whether simply ownership of this portfolio is anticompetitive and require divestiture? No of course not. We have to wait until they enforce the patents to determine whether they have abused or merely exploited their property.
Then what? We will need to determine whether the rates they charge are supracompetitive enough to say they have violated an antitrust law. We may look, as Commissioner Rosch suggested, to determine if they are using only part of the patent wall to cover a commercial embodiment and not practicing the rest to erect an anticompetitive barrier to entry.
So let’s see what this means. An investor made a bet early on that these individually nonvaluable patents would be valuable in the aggregate (say like acorns). He sets a price that the market will bear, which we determine later is too high. He further determined which patented product will most likely be successful and which part of the portfolio should not be practiced to maximize his profit, but will look bad for not practicing every claim in every patent. We determine later that this was not a legitimate business decision and therefore violates FTC section 5. This is way too much hindsight.
I respectfully submit that government anticompetitive review of patent acquisition is silly.
“I’m sticking with the view that acquiring a patent from the Patent Office and acquiring one (or more) from another are distinctly different transactions, from an antitrust sense”
Leopold, my earlier point was related to this. I have no doubt that this is how the law may have developed (I say “may” because I have not researched the issue). It just doesn’t make sense to me (and apparently to some others), because it can lead to ridiculous results by only looking at the “means” (which may have little to do with the ultimate harm) rather than the “ends” (which is the ultimate harm).
To me, “anti” “trust” means “against” “limiting competition”. It seems ridiculous, to me, to allow 100% market control because I patented it myself, but to preclude a much lower percentage of market control (equating to much less harm) because I purchased the patent or patents.
“asymmetries,” of course. I’m not sure what “assymetries” might be. I’m surprised the spam filter didn’t catch it…
“Ignoring the fact that actual pollution damage is generally hard to quantify, indirect effects due to pollution will affect a much broader area – e.g., carbon emissions into the atmosphere.”
Lionel, Coase was certainly aware of these problems, which I think can be considered as types of transaction costs. The cost of information, assymetries in information, collective action problems, free-rider problems must all be considered.
This is precisely why I think Coase’s insights are useful, however. For a real-world situation, recognizing precisely why rights might NOT be efficiently distributed is the first step to figuring out how (or whether) to fix it.
P.S. I don’t know anything about Reason magazine – I just stumbled across this article when looking for a summary.
“I think the determination of abuse is the antitrust analysis, so you’ll get accused of an antitrust violation if you are abusing it. This suggests you think both (issuance and purchase) should be analyzed for antitrust issues. Is this correct?”
No. Looking for antitrust issues at the issuance of a patent doesn’t really make any sense to me. By virtue of the Patent Act, we’ve already decided that certain limited monopolies are OK. Thus, we decided ahead of time that 100% control of a narrowly defined market (defined by the claims) is OK.
The abuse I was referring to was patent misuse (to the extent that anything remains of that doctrine). In my view, merely seeking the legitimate issuance of a patent can’t be abuse, in an antitrust sense. How you choose to exploit it might be.
I’m sticking with the view that acquiring a patent from the Patent Office and acquiring one (or more) from another are distinctly different transactions, from an antitrust sense. The latter can be anticompetitive, and presumably actionable under the antitrust laws. The former (in my opinion) is certainly anticompetitive, but it’s part of a bargain that Congress has already established under the Patent Act.
I might be quibbling over semantics. I appreciate the discussion.
CD and LB,
“If I buy a super patent that gives me complete market dominance, I will surely pay a steep monopoly premium.
There is no monopoly premium if the super patent issues from the PTO.”
I disagree with this discussion of the premium because it’s a risk/benefit analysis. The way I see it, the patent seller is only trading the risk to the patent buyer.
“Market forces prevent accumulation of anticompetitive market power precisely because the owner would generally rather grant many non-exclusive licenses than a single license.”
I also disagree with this sentence. Generally, licensees want to be exclusive licensees to ensure that they can profit from the license. Otherwise, they end up having to compete with the other non-exclusive licensees on the basis of price (assuming the licenses are the same to all licensees). That’s why the owner can get better license rates for an excludsive license. In other words, market forces encourage consolidation of patent power.
I am confused by your analysis of patent issuance vs patent purchase. I still don’t see the difference between the two. Either both should be analyzed for antitrust issues, or neither should be analyzed for antitrust issues.
“But you’re not going to be accused of an antitrust violation for exploiting a single government-granted monopoly, as long as it’s not abused, regardless of whether the patent is home-grown or purchased. This is not arbitrary, it’s common sense. And the law.”
I think the determination of abuse is the antitrust analysis, so you’ll get accused of an antitrust violation if you are abusing it. This suggests you think both (issuance and purchase) should be analyzed for antitrust issues. Is this correct?
BTW, rereading what I wrote, I did nor meant to imply anyone, especially Coase, was saying ignore transaction costs. I meant to say even in a world where transaction costs were zero, there were still problems with the examples used.
Generally, I agree that Coase’s ideas does supply a useful way of framing a decision on allocating rights and responsibilities.
Leo,
I like Coase’s article and logic and this is one of the more balanced economic articles that I have read Reason magazine, but even if we accept that we can ignore transaction costs, there are still some substantial problems that Coase’s ideas do not overcome (I was surprised to see Reason admit this as a possibility). Specifically, with the examples provided. Ignoring the fact that actual pollution damage is generally hard to quantify, indirect effects due to pollution will affect a much broader area – e.g., carbon emissions into the atmosphere. The idea that a polluted town and a factory could come to an economic agreement is about as unrealistic a model as any in economics.
Leo,
Thanks for the link. I like this guy Ronald Coase, my kind of inventor.
BTW, I had trouble with the link until I tried it without the period on the end as so:
link to daviddfriedman.com
I usually put links on separate lines cause if I don’t they don’t always link up.
anti, I think you’re right.
Here’s a good summary of Coase’s contributions, for which he won the Nobel Prize: link to daviddfriedman.com. A nice feature of Coase’s work is that it’s not unnecessarily obscured by academic jargon.
Coase’s theorem technically doesn’t directly apply to very many real-world situations, since it assumes zero transaction costs, but (in my view) it can provide tremendous insights into many problems. It’s especially relevant to issues around property rights.
I have never formally studied economics, and am not familiar with Coase’s Theorem, but it seems that the anti-anticompetitive market force discussed above only applies to the so-called “super patents”. The “patent wall”, on the other hand, suggests a monopoly formed from aggregating a large number of patents to cover multiple segments of the market. Presumably, each patent ‘piece’ is not, on its own, particularly valuable, and therefore would not see the dramatic (or, as dramatic of a) “monopoly power” markup as discussed above.
“Doesn’t every business want to crush their competitors? Don’t we want them to have that intent?”
Yes, and yes, and we also always want them to fail. It’s the spirit that keeps things going, but you never want the objective that is driving the spirit to go away. You always want competition, and you always want to dangle the carrot of smashing it in front of it’s competitors, but never let it actually get smashed completely. This isn’t something I came up with, I learned that in school somewhere.
Amongst learned men I would think this to be self-evident. Also, LB laid out a good analysis.
VA on the other hand is still but hurt. Tell me VA, did god have to come up with a new human genome to make skin as thin as yours?
Yes Mr. Bloom, exactly right. We will, of course, ignore transaction costs.
Market forces prevent accumulation of anticompetitive market power precisely because the owner would generally rather grant many non-exclusive licenses than a single license. Otherwise he would have to guess what the monopoly premium would be and find a buyer willing to pay.
Many licensees = competitive market
Under this view, one can only achieve anticompetitive market power by growing your own patent. Therefore, government antitrust oversight is only required for patent issuance not patent purchase
“The owner could have licensed the patent instead to everyone (and perhaps made even more money).”
Hmm… There’s a potential research paper there. Because a monopolist can capture all or most of the consumer surplus, it ought to be able to extract higher aggregate rents than a bunch of competing entities could. (I think.) Coase’s theorem would seem to suggest that these patent rights should end up in the hands of the highest value user – the potential monopolist. Right? (Of course, Coase ignores transaction costs.) So shouldn’t we be on the alert for antitrust issues with any significant patent purchase?
That’s it for me…
Again, CD, very interesting comments.
“If I buy a super patent that gives me complete market dominance, I will surely pay a steep monopoly premium. The owner could have licensed the patent instead to everyone (and perhaps made even more money). The market forces a would-be monopolist to pay.”
That sounds right. However, are the costs I incur to become a monopolist relevant to the antitrust analysis?
My previous argument was really about which transactions the FTC should review. If we’re concerned about the potential anti-competitive effects of merely issuing a patent, shouldn’t that be addressed by the patent statutes?
Teacup, I submit that intent is not relevant to antitrust. Doesn’t every business want to crush their competitors? Don’t we want them to have that intent? Isn’t the desire to have a new, improved product exactly the same as the desire to crush their opponents?
There are no nice guys in business.
In fact, antitrust red flags go off when businesses cooperate (we call it “collude” to make it sound more sinister).
Mr. Bloom, one (not me, mind you) would argue that it is precisely because a patent issuance is not a market-based exchange that such an acquisition begs for antitrust review.
If I buy a super patent that gives me complete market dominance, I will surely pay a steep monopoly premium. The owner could have licensed the patent instead to everyone (and perhaps made even more money). The market forces a would-be monopolist to pay.
There is no monopoly premium if the super patent issues from the PTO. I pay the same fees regardless of the market power of the patent.
…that antitrust LAW isn’t so simple. I’m a m0r0n…
“I think that you proved his point – 90% good, 85% bad = arbitrary. Also, we appreciate your policing the board for spelling errors. Your case citation does not seem to be in blue book form, though, so -1 for you.”
Joe Joe, if simple formulas work for you, then OK. Other commentators appear to recognize that antitrust isn’t so simple. Incidentally, I wasn’t the slightest bit concerned about any spelling errors. However, “antitrust” is an adjective, not a noun. Shouldn’t using words correctly matter in a legal discussion?
Interesting comments, CD.
You said, “If the FTC can assert control over asset acquisition through a purchase, why cannot it not also assert control over asset acquisition through the issuance of a patent? Both are public transactions that affect the market.”
It seems to me that these two transactions are quite different in nature. I’m not sure that applying for a patent based on your developed technology is a market transaction at all – sure, it affects trade, but it isn’t a market-based exchange. Rather, the government is granting a monopoly in return for disclosure of your technology. On the other hand, a purchase of issued patents is clearly a market transaction.
It makes sense to me that the FTC might have jurisdiction over the latter, but not the former.
If a merger could create a monopoly that would eliminate competition, then a patent could as well. It all depends on the sizes of the companies involved or the scope of the patent(s) involved. Creating hard and fast rules, like 60% of market share or 10 patents with 20 or more claims, is nonsensical. It has to be case-by-case. Intent should also be a factor, i.e., did you intend to get a new and improved product to market or did you just want to crush the other guy?
The violin that weeps for this kind of Megacorp is smaller than a neutrino.
Leo:
I think that you proved his point – 90% good, 85% bad = arbitrary. Also, we appreciate your policing the board for spelling errors. Your case citation does not seem to be in blue book form, though, so -1 for you.
JT
The intersect of intellectual property and antitrust law is tricky and interesting. However, at its simplest level, these laws demonstrate America’s collective understanding that some monopolies are good (or least necessary) and some are bad.
I agree with Professor Crouch that there is no real difference in market power if a firm organically obtains patents or if a firm purchases them outright. If the FTC can assert control over asset acquisition through a purchase, why cannot it not also assert control over asset acquisition through the issuance of a patent? Both are public transactions that affect the market. It may make even more sense for the government to review government-granted monopolies than to review private transactions.
The FTC would claim to be reviewing whether the firm had market power in either case, and therefore whether the acquisition of an asset (another company, a patent, etc.) would be anticompetitive. I don’t think it’s a good idea, but it would arguably fall within the antitrust purview of the FTC or DOJ.
There were 3 big antitrust cases at SCOTUS last year that patent attorneys may wish to review. Antitrust considerations are a bit different now. For example, tying is no longer per se illegal and rule of reason applies. Also, market power is not presumed based solely on ownership of a patent.
Regarding the metrics of determining whether a firm has market power. Market power is a critical factor for determining whether conduct is anticompetitive. As you would imagine, a single big player can affect its own monopolistic market much more than a single small player could affect a more distributed competitive market. Courts will look to several indicators, such as the HHI index, price inelasticity, and so on. A large patent portfolio may also reasonably be considered a factor for determining market power as would a patent covering standards set by an SSO.
FYI, Rambus was reversed by the DC Court of Appeals (a very sophisticated antitrust court) not the Fed Cir. Rambus looked like bad guys to me although not antitrust-level bad guys. N-Data look like good guys to me. I am disappointed with the FTC’s decision on N-Data.
Finally, I agree that agencies (FTC, PTO, SEC, FDA, and on and on) are attempting to grab more power, but that is what agencies do.
“Based on the above comments, controlling 90% of the market is not anti-trust [sic], but controlling 85% is. Kind of makes it all seem arbitrary.”
No, based on your hypothetical, controlling 90% of the market is not an antitrust violation, but controlling 85% is.
Part of this, of course, is defining “the market.” A patent can give you control of 100% of a market defined as the product covered by the patent. But you’re not going to be accused of an antitrust violation for exploiting a single government-granted monopoly, as long as it’s not abused, regardless of whether the patent is home-grown or purchased. This is not arbitrary, it’s common sense. And the law.
“The monopoly afforded is only temporary.”
I don’t think this matters very much. The concern of antitrust is the exploitation of market power to eliminate competition. Twenty years is plenty of time to do that. If you do that effectively enough, then you won’t even need the patents for a while. Again, see Hartford-Empire.
I think you also have to remember that patents only secure the exclusive rights to the invention FOR A LIMITED TIME. They expire. The monopoly afforded is only temporary. That’s the major difference between a true monopoly and the patent based monopoly. The true monopoly is designed to prevent market entry indefnitely. With a patent monopoly, you know the EXACT market entry date. Look what happened with the Zyrtec patents. They expired on Dec 25, 2007, and on Dec 26, 2007, product shipped from competitors (I may be off by a couple of days, but you get the point about them being back to back.)
“”Does this mean that buying one patent of very broad scope could also be an antitrust violation?”
No.
“If each of the little monopolies is proper, why is the big monopoly (i.e. the sum of the little monopolies) improper?”
Because of the undesirable restraints on competition that the aggregation of the proper activities causes. This is no different than other antitrust concerns. The application of this to patents is also not new. See Hartford-Empire Co. v. United States, 323 U.S. 386 (1945).”
So, what if we compare buying a single patent, which results in control of 90% of the market (based on the above comment that is okay and not antitrust), with buying a portfolio of 200 patents, which results in control of 85% of the market.
Based on the above comments, controlling 90% of the market is not anti-trust, but controlling 85% is. Kind of makes it all seem arbitrary.
So “anti-capitalist” is the new knee-jerk smear now? Is that like “unpatriotic” or “communist”?
N-Data was a good move, frankly. RAMBUS would have been, if not for the CAFC. Stealth monoplies on **established standards** is absolutely against the public interest – and that’s not even getting into the ethical problems.
Those of you arguing the capitalism card, remember that patent rights are not natural free-market forces – they are unnatural, government-enforced monopolies. I don’t see any fundamental problem with using government policy to curb or correct other government policy. Some of you are talking about this issue as if patent monopolies are some sacred, god-given right, rather than a purely legal mechanism employed to promote progress.
“Does this mean that buying one patent of very broad scope could also be an antitrust violation?”
No.
“If each of the little monopolies is proper, why is the big monopoly (i.e. the sum of the little monopolies) improper?”
Because of the undesirable restraints on competition that the aggregation of the proper activities causes. This is no different than other antitrust concerns. The application of this to patents is also not new. See Hartford-Empire Co. v. United States, 323 U.S. 386 (1945).
If you can buy or develope yourself all aspects of a product,either patents to, or if you got the money, the competitors themselves, I call it great business savvy. What are you people, anti-capitalists-this is whats it all about-the American way.Until the government stepsin, you’re a genius.
Interesting take, has he looked at agriculture lately?
“You will learn in law school…”
He hasn’t learned anything here. What makes you think he’s going to learn anything in law school?
–I don’t think people understand, you have a monopoly only over the patent. If you intentionally accumulate a lot of patents to exclude all competition in an entire field (or industry) then you have, in aggregate, violated the anti-trust laws. No single action did it.–
E6K, we understand. Our questions are more along these lines:
If each of the little monopolies is proper, why is the big monopoly (i.e. the sum of the little monopolies) improper?
How many of the little monopolies will it take before the big monopoly is improper? 2, 10, 100…?
Does this mean that buying one patent of very broad scope could also be an antitrust violation?
How can a patent seller know beforehand whether or not the buyer will be permitted to buy? How can a buyer know beforehand whether the govt will approve the purchase? Antitrust concerns are typically foreseeable (e.g. Google buying Yahoo) b/c there are certain metrics, like market percentage, to look at. Do such metrics exist for patents?
As DC asked, why should there be a difference between developing every important patent in the field internally and buying every important patent in the field?
You will learn in law school that planning, including avoiding expensive litigation against the US govt, is a large part of being an attorney. Settled expectations on how law will be interpreted and applied are a large part of planning.
Fee increases – why? To pay the Tafas litigation/appeal? To pay for new desks for examiners to put their feet on while we do all the work for them as we assemble ESDs? Or maybe it’s to cover examiners’ gasoline costs, which could be avoided if they were allowed to telecommute (as the TM examiners are allowed to do). Or to cover the $600 refund from the IRS (i.e. redirecting to the PTO the funds that were diverted for all those years).
I don’t think people understand, you have a monopoly only over the patent. If you intentionally accumulate a lot of patents to exclude all competition in an entire field (or industry) then you have, in aggregate, violated the anti-trust laws. No single action did it.
The fact that situations like this are even being investigated belie the good intent of rewarding anyone other than the inventor himself with the exclusive right. Under any circumstances. Ebay should curb this though.
The pre-merger review process has always had the capability to be triggered based on transfer of patent assets in excess of the threshold value. However, adding a pre-merger(purchase) scrutiny without evidence of anti-competitive behavior such as, as was pointed out, tying, seems like a slouching toward the nanny (socialist) state.
It appears that every agency in this administration is out for an ad hoc power grab. There are plenty of well established anti-trust theories out there. Shatterproof Glass (organic invention tying) , the radio patent pools (cartels to exclude new players) , etc, etc. The FTC has been whacked since that innovation report a couple of years ago (we need a new federal agency of invention). To the extent that the FTC is advancing new theories (buying a patent then asserting it?) is just plain weird, but i feel for the defendants that are going to have to litigate down the new theories. This is a republican administration? Good grief.
I’m not taking the FTC’s side here, but (in response to prior commenters) do recall that the statutes prohibit attempts to monopolize as well as agreements in restraint of trade. Certainly it is joint action that they are most concerned with.
As with Walker Process, etc., a patent doesn’t create antitrust immunity.
Any organic activity would not likely be viewed as an attempt to monopolize, because it really doesn’t fall into the type of activity that is viewed with antitrust suspicion.
So, it’s O.K. to enjoy the fruits of one’s creations…that is, unless you acquire too many fruits within the same “family” of fruit…in which case you magically become an “antitruster,” just ripe for a public flogging and smack down.
Whether or not they acquire some–or even all–these patents from others (presumably for just value) is irrelevant.
Reminds one of the massive federal bailouts of Chrysler, big banks and brokerages, etc. … “’cause (with of course heartfelt apologies to all you no less deserving smaller companies) ‘But they’re just too big to let fail!'”
Ridiculous indeed.
These FTC investigations are ridiculous. Patents are legal monopoly rights (I know us patent attorneys do not like to use the M word because the general public, and apparently the FTC, gets confused, but they are monopoly rights). If a company buys a bunch of legal patents absent any other monopolistic activities (such as tying), there should be no investigation.
Comments are closed.