Guest Post by Ted Sichelman, Professor of Law and Director of the Technology Entrepreneurship and Intellectual Property Clinic and Center for Intellectual Property Law & Markets at the University of San Diego School of Law, and Shawn Miller, Lecturer in Law and Teaching Fellow in Law, Science & Technology at Stanford Law School
As Patently-O has described in detail (e.g., here and here), patent litigation rates have been in flux over the last several years. First, in 2010, rates nominally went up because of false marking suits. Then, in 2011, following passage of the America Invents Act (AIA), patent litigation rates appeared to increase (and the media made much ado over this). Yet, the only thing that changed immediately after the AIA was how we counted patent litigation numbers because of new joinder rules. Indeed, the AIA’s joinder rules actually lowered total defendant counts a bit. Next came inter partes review (IPR) and covered business method (CBM) review, which appear to have substantially decreased litigation rates in 2013 and 2014. Then came Alice Corp. v. CLS Bank and its progeny, which some have claimed decreased litigation rates throughout 2014. And now rates seem to be back up in 2015—why, nobody is sure.
To further understand the drivers of patent litigation, Alan Marco (Chief Economist at the USPTO) and the two of us recently conducted the first rigorous study of the effects of the economy as a whole (the “macroeconomy”) on patent litigation rates over the period 1970-2009 (we stopped at 2009 to avoid the counting mess I just described). The study will be published this month in the Journal of Empirical Legal Studies (a draft is available here).
As background, we examined changes in the rates of total litigation, litigation per issued patent, and litigation per in-force patent. To do so, we used a new USPTO dataset that provides the most accurate numbers on total in-force patents by year gathered to date. Figure 1 below shows the number of in-force patents between 1971 and 2009 (the vertical bars represent periods of recession).
Figure 1. U.S. In-Force Patents (1971-2009) [from USPTO data].
Figure 2. U.S. Patent Litigation Per In-Force Patent (1971-2009)
(Total Patent Litigation in Gold (with annual filings on left y-axis) and Litigation Per In-Force Patent in Blue (with annual values on right y-axis)).
We then performed numerous time-series regressions against important macroeconomic variables like GDP, R & D spending, interest rates, and related factors, controlling for in-force patents and other key variables. Our major findings (over the last 20-25 years) are as follows:
- In economic upswings (i.e., increasing productivity, low interest rates, low economy-wide financial risk), patent litigation rates generally fall.
- In recessions, whether rates rise or fall depends on the availability of credit.
- When credit is freely available in a downturn, overall patent litigation rates generally rise.
- However, when credit is scarce (like in the most recent recession), overall patent litigation rates generally fall.
We explain the first finding on a “substitution” theory—namely, when selling products and services is highly profitable, companies and investors are less concerned about earning revenue from patent litigation. We speculate that our second finding is driven by the increasing reliance by plaintiffs on external funding. Indeed, the availability of credit has only played a significant role in patent litigation rates since roughly the mid-1990s, when licensing-driven litigation began to rise—first by practicing entities such as IBM and Texas Instruments and, later, by non-practicing entities (NPEs).
Over the last five years, the mean U.S. real GDP growth rate has been about 2-3% per year. Ignoring other factors, and assuming our model applies on a going forward basis since 2009, this increase in productivity has very roughly amounted to a 6-9% decrease in annual patent litigation counts over the same period. In general, it is likely that a sizeable portion of the decrease in patent litigation rates over the last several years has been not merely due to the rise of IPRs and CBMs, and the issuance of Alice Corp. v. CLS Bank, but also to the economy as a whole. So the next time the economy worsens, if credit remains relatively available, all other factors equal, it is likely that patent litigation rates will rise significantly—not by a huge amount, but enough to notice. Whether the current increase in patent litigation is somehow related to a declining macroeconomy, only time will tell.
Please note that the views expressed herein solely express our personal views and do not express the views of the U.S. Patent & Trademark Office.
Old saying,
“When times are good people will pay for anything: when times are hard they will fight over anything”.
Has anyone compared filing rates with litigation rates to see if they are negatively correlated? Just asking – no idea of what the answer is.
It is September 16th. Where is the Executive’s response to Ron Katznelson’s “Tr011” propaganda challenge?
Bueller…
.
Bueller…
Just a question: Are there a sufficient number of major general economic dips and rises over the study period for a highly accurate prediction of future correlation with patent suit filings?
It depends on whether the professor feels there are or not. All that matters is what the law professor feels should be true. Please stop trying to discuss reality. It is offensive to question the feelings of a law professor.
Maybe we need another guest post along the lines of “Wah, stop picking on us.”
The last time a guest professor did that, he rather selectively did not address some rather key points in the discussion of patent eligibility of software.
It gets embarrassing when people only want to monologue instead of dialogue.
So, to sum up the blog post:
1) Patents and litigation have not grown when normalized for population growth and the rise of technology since 1972;
2) The patent problem is a fabricated problem by Google and the like;
3) The professor’s blog post was not peer reviewed and any ethical violations that the professor commits will not be addressed by his university.
4) A professor with an agenda has just published a paper that illustrates his agenda is backwards.
Re: “..decreased litigation rates throughout 2014. And now rates seem to be back up in 2015—why, nobody is sure.”
The most likely explanation I have heard offered is a rush to get suits filed before the much stricter requirements for filing new patent suits in the new patent legislation pending in Congress could get enacted. [Just as we saw a temporary increase in filings to get cases filed before the AIA anti-joinder statute went into effect.]
Has any other logical explanation been offered?
Not to criticize this study, but there have been so many conflicting statistical reports and explanations re alleged increases or decreases in patent litigation that it is difficult for many of us to fully accept any of them. With many unaccounted for factors. Some studies did not even accurately take into account the AIA non-joinder statute – counting suits rather than defendants.
It is also predicable, but too soon to really tell, the full impact of the apparent recent great increase in attorney fee sanction awards, Alice 101 and ambiguity defenses, 102(f) expansion, stricter control over damages evidence and awards, etc.
Even looking back in time for long term trends, the switch of almost all patent suits from bench trials to jury trials, and the switch of all patent suit appeals and authority to the Fed. Cir. might well have had more effect on decisions to file patent suits than general economic factors? Even as to economic incentives, the book “Rembrandt’s in the Attic” several years ago, and PR about IBM making a billion dollars a year in licensing, both had effects unrelated to the economy in general.
David Stein: If you have even a halfway plausible argument, patents can be attacked, with a decent chance of success, just by rolling the dice on the judicial process. If you have the cash, you can just keep arguing and appealing, repeatedly, until you get the result that you want. If you can outspend your adversary to exhaust their budget the time you get your desired verdict, then you win! This outcome does not actually depend on the strength of your argument vs. theirs, but on the size of your litigation budget.
Shorter David Stein: “I liked it better when all you needed was a junk patent and a bulldog attorney and you could make it millions and millions of dollars just threatening whoever.”
Seriously, David: welcome to the US legal system. The changes that are taking place in the patent system are taking place because the system went off the rails. See Figure 1 above if you have any questions.
Why is this so hard for people?
Except that thing we call reality MM. You know, the real world that is not part of the Google bucks fueled psychotic nonsense that you push.
Let’s look at reality: In 1972 there were 209 million people in the US and now there 330 million people.
Furthermore, technology is a far larger percentage of the US economy now then it was in 1972. I’d wager that if the number of patents were normalized for population and percentage of technology of GDP that the number of patents in force is much smaller now then it was in 1972.
Reality. Now, professor who published this paper why didn’t you mention this? We are getting into the non-peer reviewed nonsense of law professors where no ethics violation in regards to a journal article is too much to warrant notice from a dean.
I suppose an ethical person would also point out that patents –if working properly–would tend to increase the number of litigations and the number of patents in force. I.e., the incentive to invest works so more invest, and thus more patents and more litigation. Etc…. This is not a bad thing inherently.
But, you know, little facts like this don’t matter in the shadow of Google bucks.
NW: I’d wager that if the number of patents were normalized for population growth
“Population growth” has absolutely nothing to do with the patent bubble and reasonable people who can do math will see that immediately from Figure 1.
technology is a far larger percentage of the US economy now then it was in 1972
I’ll grant you that litigation over endless reams of “computer-related” “innovations” that should never have been the subject of patent grants is a “far larger percentage” of the US economy now than in 1972. I’ll also remind you that conducting business by information exchange “on the Internet” or via some other system of connected computers hasn’t qualified as “technology” for a long, long time. Of course, if you want to include all the business conducted over the phone, radio, TV, newspaper and through the use of ball point pens and paper in 1972 as “technology” then go right ahead.
Of course this brings up another ethics violation. Asserting that information processing machines are easy to build or are merely implementing some existing method.
Of course no scientist that are paid by Google would support this and making this assertion should be enough to impeach a judge (Stevens, Posner, etc.) or to remove a professor from his/her tenured position.
But that is reality and we live in the Google bucks world of Citizens United. Why don’t you just write on a piece of paper what you do MM and ask a boy clerk to implement it on a computer? Then you could sit around all day and collect your pay without doing anything?
Oh reality…
So, NW talks about technology, and in response, you talk about patents. Do you agree that technology has exploded or not since 1972? Five years ago, I knew very few people who had a smartphone; now, I look very hard to find anyone who doesn’t. Computers were new when I was in college and I loved it when I got a 40MB hard drive. Now, I have 6 computers at home, one of which stores 11 TB of data. I have three tablets, two smart phones, and two programmable robots. I watch TV using my HT computer. I can control both my garage doors and my thermostats from anywhere. I control a lot of my lights using a device attached to my network. Start including all the other “smart” devices I have, such as a smart car, TV recording devices, microprocessor-controlled heating and cooling systems, many LED lights, etc., and you’re talking a lot of technology. And I’m a late adopter.
Again I ask: Do you agree that technology has exploded or not since 1972? And if there has been a technology explosion, what effect of that technology explosion is there on patents? Could there not be a correlation? (One that has nothing to do with “junk” patents?)
PatentBob, Sr., circa 1972: “Do you agree that technology has exploded or not since 1932? Five years ago, I knew very few people who had a color TV; now, I look very hard to find anyone who doesn’t.”
Deep compelling stuff.
if there has been a technology explosion, what effect of that technology explosion is there on patents? Could there not be a correlation?
If in response to the awesomeness of the telephone the Supreme Court decided that (1) everytime a telephone was hooked up to “new” parties turned that telephone into a “new” machine and (2) everytime “new” information was communicated over a telephone line that represented “new” technology, then we would have seen a nearly identical correlation to the “correlation” you are referring to.
This isn’t rocket science. You must either be born yesterday or ignorant (willfully or otherwise) to not appreciate exactly what happened in Figure 1 and why. There’s nothing complicated about it.
These arguments you make that all the information processing is easy is just ridiculous. All of the empirical evidence supports the opposite view. But reality means nothing compared to a Google buck.
…because saying “deep compelling stuff” is an automatic pass as to making comments that are deep and compelling…
…or something.
the “explosion” of more patents has brought a LOWERING rate of suit.
Yes, the data clearly shows that more junky patents is the best way to reduce the rate of “lawsuits per patent in force”! Someone run and tell the PTO, quick!
your premise does not stand up to the facts
LOLOLOLOL
Whatever that’s supposed to mean.
The facts pretty much speak for themselves now. It’s not 2000 anymore. It’s 2015. Shall we review the significant events over the past 20 years and bring everyone up to date?
Eventually someone will write a fascinating book about the entire debacle. The best part will be documenting the words and deeds of those folks who stuck their heads in the sand and did nothing else except screech “more patents is more progress!” for more than quarter century while the system imploded.
You really don’t grasp the meaning of BOTH graphs, do you?
This is not rocket science – H, this is something a fourth grader could grasp.
Malcolm, on the other hand, just stumbled along.
He funny.
The data clearly shows the “bubble” of “OH NOES” patent enforcement RATE clearly peaked BEFORE the rise of what you curse as (somehow) non-technology.
Given that lawsuits rarely occur immediately upon issue, that places that “OH NOES” bubble squarely in the realm of a FLAT period of patent grants.
Further, according to you, the slope should be in the opposite direction – instead of declining. Oops.
But hey, why let anything like facts or critical thinking get in the way of your baby rants, eh?
Dramatic increases in patent litigation rates are just one of many issues arising from a broken patent system. Between Figure 1 and the re-direction of patent ownership into the hands of b0ttom-feeding NPEs who do nothing except push paper around there really isn’t that much more that anybody needs to know. It’s this latter class of folks, of course, who spend half their time asserting junk and the other half screeching on the Internet that we’re going to become a third world country if we cease catering to them.
But wait! Little Ronny Katznelson tell us that patent litigation rates are “similar” to those in 1920-30. So let’s just all move along. Nothing to see here!
LOL
Whether the current increase in patent litigation is somehow related to a declining macroeconomy, only time will tell.
Do the authors have any comment on the oft-asserted direct correlation between the number of patents-in-force and the state of the “macro-economy”?
“Dramatic increases in patent litigation rates are just one of many issues arising from a broken patent system.” – Why doesn’t this signal the opposite? – an increase in patent litigation could equally signal that the patent system is working as intended – serving as a legal tool to exclude others from using, making, selling, importing, and offering for sale within the U.S.
an increase in patent litigation could equally signal that the patent system is working as intended – serving as a legal tool to exclude others from using, making, selling, importing, and offering for sale within the U.S
Except that here on planet earth we’re talking about a dramatic increase in litigation that correlates directly with (1) the granting of reams more patents than ever before in the history of any patent system on earth, (2) patents covering subject matter that the system was not designed to protect and promote.
Those junky patents, of course, are going to be with us for a long time, even while the courts and Congress are busy cleaning up the mess that was made (especially if the PTO continues to sidestep or ignore the changes on behalf of its eternally disgruntled “clients”).
It seems to me that litigation is the perfect forum for judging the question of whether: “patents covering subject matter that the system was not designed to protect and promote”.
litigation is the perfect forum
Even professional litigators take a long pause before letting those words slip out of their mouths.
But really the “perfect forum” for deciding whether, e.g., methods of thinking about data are the sorts of methods the patent system was designed to protect is the “forum” where reasonable ordinary people who are breathing oxygen and have both feet on the ground are looking at the uncontested facts about the history of the system and the “new” “innovations” the system has purported to protect at various stages. That decision-making isn’t going to be helped by inviting just two completely compromised, biased, profit-obsessed silver-tongued mouthpieces to the table to battle piecemeal over whether some junk is “merely awful” or “disgracefully awful” which is the manner in which your “perfect forum” has been addressing the issues (or pretending to) for nearly a half century now.
Has it been established that the litigated patents are in-fact “methods of thinking about data”?
Of course not – why muddy up a good rant with facts?
Except reality. You know as pointed out above. You know, reality. You know things like normalizing numbers things that if you don’t do in a science journal you lose your tenure and job.
Gee, have you heard of the dramatic increase in XXX from 1972 to 2015? Oh wait, if it is normalized then a decline.
Ethics…..reality…don’t mix with Google bucks.
Baby steps Night Writer – academia finally followed my advice and normed by outstanding patents to show the (gasp) declining rate.
[…] I didn’t want to cover up this interesting guest post https://patentlyo.com/patent/2015/09/previously-improving-litigation.html […]
The two graphs yield a wealth of information.
Most notably, how the “horrors” of litigation are VASTLY overplayed and quite evidence the fingerprints of propaganda and manipulation in order to reach a certain philosophical point.
Note that the rise in litigation rate per patent enforced occurred BEFORE the rise in number of patents.
Note that even at its peak, we are talking about less than TWO PERCENT.
The flea on the tail of the dog wags the dog.
Because “everyone” knows that enforcing patents is like the “worst thing ever.”
Or something.
(Interestingly, I see a note to Ron Katznelson – and I have to wonder if anyone has an update on the now W A Y overdue response from the Executive on the bogus “Tr011” propaganda piece – by the by, which augments the take-away of this thread)…
(and may I humbly point out that the rate of suits per patent in force from the 1997 point of “explosion” of patents is ACTUALLY dropping – something I was the very first to suggest)
In the best Elvis tones: Thank you, thank you very much.
he rate of suits per patent in force from the 1997 point of “explosion” of patents is ACTUALLY dropping
LOL
Deep, deep thinking at its finest.
Yes, all we need to do if we want to lower the rate of junk patent lawsuits “per patent in force” is …. grant ten times more junky patents! Get to work, PTO! D
Also was this peer reviewed? Or is this some law professor generating nonsense–AGAIN for their own personal agenda. And would there be consequences if the professor had not been entirely honest in his presentation no of the data? The answer is no. This is well documented in other areas of law.
So, the value of this is zero.
Also was this peer reviewed?
What difference would that make? After all, Winston Churchill told us that peer review was the worst system ever, except for all the others. Or maybe it was some other super serious person who said that.
The interesting thing about the recent uptick is that while NPE lawsuits are down (probably thanks to both decreased NPE activity in general, and the rise of invalidations at IPR), the big increase is: lawsuits between non-NPE, actual-product-producing companies.
To me, the answer is an obvious consequence of converting so much of patent law into smell tests. It goes like this:
(A) Many critical requirements of patent law (101 “abstractness,” 112(b), 112(f), 103 combinations, etc.) are now determined not by objective evidence, like the contents of references or technical analyses – but according to the whims of the reviewer.
(B) Deference of reviewers to lower-level decisions is practically nonexistent. The PTAB frequently overrides examiners’ subjective conclusions; district courts frequently override the PTAB’s conclusions; and the Federal Circuit overrides district courts’ conclusions so often that the Supreme Court has had to intervene.
(C) Here’s the result of (A) + (B): If you have even a halfway plausible argument, patents can be attacked, with a decent chance of success, just by rolling the dice on the judicial process. If you have the cash, you can just keep arguing and appealing, repeatedly, until you get the result that you want. If you can outspend your adversary to exhaust their budget the time you get your desired verdict, then you win! This outcome does not actually depend on the strength of your argument vs. theirs, but on the size of your litigation budget.
In other words: When patent law becomes so arbitrarily discretionary, the only way to settle any argument is to resort to litigation, to the exhaustion of further appeals and/or budget. Adamant and/or spiteful litigants can therefore insist on throwing their budgets into the process to contest and/or protect the last verdict. So we have cases like eBay v. MercExchange (six years of litigation!) and AMP v. Myriad (two trips to the Supreme Court), where the issues are so fractured and complicated and case-specific as to have become meaningless.
This was the predictable result of the Supreme Court incrementally turning patent law into a smell-test. The law has become scorched earth, and the resulting unpredictability hurts nearly everyone with skin in the game.
This horrid state of affairs that will persist until the Court comes to its senses and stops creating smell-test legal requirements – or until Congress takes this power away from it.
David, when any company can repeatedly litigate validity as you indicate, res judicata, a central feature of a judicial system, is effectively abrogated. As a consequence, even one company can continue to litigate validity, repeatedly, in a variety of fora, even after final judgment. All one has to do is keep it up until the patent expires, the patent owner says uncle, or the infringer wins.
This is not the problem of the Supreme Court, David. There is another problem you identified very well, and we all know what it is.
Your excessive allegiance is audacious Mr. Heller.
I completely agree.
The tests that the Supreme Court has created are the problem, Ned. They enable this merry-go-round to keep happening, where courts keep second-guessing each other.
All of this could have been handled differently: by enforcing 102/103 more stringently. If a patent is overbroad or uninventive or (English-language) obvious – then it ensnares a whole lot of prior art, right? Why not reject it on that basis, using evidence? Instead, the Court opted to create opinion-based shortcuts that provide a cheaper and quicker answer, but at the cost of exacerbating litigation – as these very metrics reveal.
David, IF we could have a completely, hindsight free test of “invention” — how much of a departure from the known must be the claimed subject matter be before it is worthy of a patent, that would be ideal.
I suggest that we move to the Euro formula, where there is a problem and a solution that must be basis for the patentable advance describe in the specification. Any old difference should not be enough if it was never identified as solving any problem in the spec.
How do we get there? Well Hotchkiss required some functional improvement. If everything was old, combining it in known ways to achieve expected results is not invention. Thus we have KSR.
This is good, but not sufficient.
Then we have Eibel Process — where the improvement must be in kind, rather than degree.
But improvements are provable, not so, with evidence? This is not simply the opinion of some judge or expert we are talking about. But real, provable improvements.
If we could go this route for most inventions, it would help a lot. We could respond to OAs with evidence of improvement, and some synergy in the combination that was not expected. Other that that, it’s all BS.
I personally am sick and tired of experts testifying on the ultimate issue. It all conclusory witch doctor magic dust and incantation to me.
On 101, we need the PTO to take the lead and offer a definition of “abstract” that meets all existing cases, and expected future. Right now, we are left in the dark — that is, unless one thinks that we have returned to Hotel Security, which is my view.
I also see others floundering in trying to preserve their business method patents: offering up things like “claim as a whole” and “preemption.” It is odd that they refuse to get it.
The problem with the improved price is that it is non statutory because it is abstract. Price is not something that one can weigh, play catch with, etc. Get real.
Circular hoohaw Ned.
Great post again David. I agree. I also think that what all this has done is shifted the burden to the applicant for the quality of the patent.
It is also hard to take anything produced by an academic with anything but a grain a salt. The academics have proven that they have agendas and that their little papers are produced to push their agenda without regard to ethics or the facts.
The presumption should be that an academic (in law) has fabricated their results. At least until the universities start to enforce ethical standards on law professors. It is well documented that universities do not enforce any ethical standards and law professors.
This horrid state of affairs that will persist until the Court comes to its senses and stops creating smell-test legal requirements – or until Congress takes this power away from it.
Gee, maybe you and your cohorts who believe in patenting logic and/or information “on a computer” can come up with a test that you can live with but that doesn’t appall everyone else.
Good luck. It’s not like we haven’t been asking the software patent lovers for this for like 10 years now.
“Gee, maybe you and your cohorts who believe in patenting logic and/or information “on a computer” can come up with a test that you can live with but that doesn’t appall everyone else.”
Because “cohorts” sounds much more maniacal than “folks” and the test ALREADY out there (the exceptions to the judicial doctrine of printed matter) does not get to the philosophical want of Malcolm’s per se elimination of the single largest area (and MOST accessible to the most people not of the one percenters) of today’s innovation.
Yeah, yeah, that’s the ticket.
/off sardonic bemusment
Your dataset is probably too small – 39 years. There’s a lot of patterns to be seen in hindsight, but I bet the system you are modeling is more complex than we realize.
Comments are closed.