The chart above might take a couple of minutes to digest, but the main point is that, while the median patent has remained relatively unchanged, patents are becoming more uniform in size.
The chart above shows the number of claims per patent for three groups of patents – those issued in 2007, 2012, and 2017 (through 10/16/17). For each group, the chart provides a size-strata based upon the number of claims found in each patent and a relative percentage. Thus, in 2007 about 27% of patents issued with 1 to 9 claims; in 2017 that figure has dropped to 20%. Similarly, in 2007 about 30% of patents issued with 21 or more claims; in 2017 that figure has dropped to 18%. That decrease at the extremes of claim count size has been replaced by a bulge in the middle – patents with 10 to 20 claims (especially those with exactly 20 claims).
This graphical representation is akin to a measure of variance or standard deviation. As expected, the standard deviation has also dropped during this time as shown by the chart below:
This second chart reports both the average claim count and the standard deviation for each year of utility patent issuance. Note that the average claim count has dropped slightly (especially since 2011) while the standard deviation has decreased substantially each year.
Causes: It is easy to postulate on the causes for the changes here. First up is USPTO Fees. Starting in 2004, the USPTO substantially increased the cost of filing patents with >20 claims or >3 independent claims. This led to a substantial decrease in “jumbo” patents with many patent claims. Today, it is quite a bit cheaper (w/r/t USPTO fees) to file two patent applications with 20-claims (3 independent) each rather than one application with 40-claims (6 independent).
In addition to USPTO fees, the market for patent prosecution has shifted significantly to patents-as-a-commodity with recognized usual costs. The fixed-cost approach means that a change from the usual 20-claim standard will involve a heightened level of review and approval — something many outside counsel avoid. Often, the fixed-cost agreement includes the particular number of claims to be drafted. This reasoning can also be combined with the continued increase in the proportion of foreign-originated patents – many foreign applicants enter the US system with a smaller claim-set (based upon high foreign fees for jumbo claims).
Finally, we should recognize that independent inventors are a smaller proportion of patent applicants than ever before – as is the proportion of applicants that file without the advice of a patent attorney. The very-small claim sets tend to come from these groups of independent inventors and those who do-it-yourself.
Some folks believe that restriction practice is also to blame here, but I have not thus-far found a significant increase in restriction requirements during this time. (Percentage of applications filed as divisionals is down during this time period).
Good or Bad?: Perhaps a question for a different post, but the standardization has some important benefits: reducing transaction costs between applicant and attorney and facilitating better management of patent examiners. The lower number of claims may also make it easier for third parties to understand the scope of what is claimed — however this is only true if the ‘extra’ claims are not simply being replaced by an ‘extra’ patent.
I’m presenting some data from the Compendium of Federal Circuit Decisionslater today at Chicago-Kent’s terrific conference on The Power of the PTAB. Below are a few of the graphs I’ll be showing, along with a bonus graph involving Federal Circuit review of inter partes review proceedings. All data is through October 17, 2017.
This first graph shows the distribution of dispute types in opinions and Rule 36 summary affirmances that arise from the USPTO. Unsurprisingly, most of the growth in these decisions comes from appeals in IPRs. More surprising to me was that a substantial chunk of the growth from 2014-16 came from appeals from inter partes reexamaminations. Those will drop off as the last reexaminations work through the system.
This second graph shows the affirmance rates in different dispute types for the period 2012-2016. Note the difference between affirmance rates in ex parte-type proceedings versus the rates in inter partes-type proceedings. For the CBMR and Interference rates, keep in mind that the number of these decisions is small for this time period (CBMR, n=15, Interference, n=25).
Here’s the bonus graph: the affirmance rates for inter partes reviews only. Although the affirmance rate was above 85% for 2015 (n=44), it was lower in 2016 (n=107), and so far for 2017 has been even lower (n=100 through Oct. 17, 2017).
This last graph shows the frequency of the Federal Circuit’s use of Rule 36 summary affirmances. The “post-Crouch” drop may vindicate Dennis’s February 1, 2017 post on Rule 36 affirmances. That bar reflects the frequency of Rule 36 affirmances in these appeals since that date.
Thanks to my research assistants, Louis Constantinou, Sarah Jack and Brett Winborn for their hard work on the Compendium. If you are a researcher on the Federal Circuit and would like access to the Compendium in its beta state, please drop me an email.
The following is an excerpt from PTO Director Michelle Lee’s keynote address today, at George Washington University Law School. Read the full remarks here.- DC
How do we continue to incentivize … incredible innovations? First, we must ensure that the USPTO issues the highest quality patents as quickly and efficiently as possible. This means, at a minimum, continuing to bring down our backlog and pendency so that good ideas can be patented quickly. Which is why I am proud to report that we’ve reduced our backlog of unexamined patent applications by almost 30 percent from its peak in January 2009, despite a 32 percent increase in applications over this time period. Our so-called first action pendencies – that measure the time from filing an application to the time of obtaining an initial decision from the examiner— are down by 43 percent—from 28 months in 2011 to 16 months today. Our total pendencies – the time from filing an application to the time of obtaining a final office action, such as an allowance—are down by 26 percent—from 35 months in 2010 to 26 months today. Our backlog of Ex Parte Appeals – that’s not AIA proceedings, but our internal appeal process to the Board after the examiner issues her decision – has also gone down. We have reduced our inventory of ex parte appeals by 45 percent from a high of about 26,000 in 2012 to about 15,000 today. And, we have reduced the average pendency for appeals by 30 percent from 28 months in 2016 to 19 months today. In short, our backlog and pendencies are now lower than they’ve been in more than a decade, and they will continue to go down. Simply stated, this means more inventors are obtaining more patents more quickly. . . .
We constantly welcome—in fact, we solicit—feedback and input. And we have shown that we are willing to refine and improve as many times as needed. For example, we’ve provided applicants with more access to examiner interviews more than doubling the number of interview hours in just eight years. We’ve had more RFC’s, Proposed Rules, and public roundtables than ever before – including on such topics as our Enhanced Patent Quality Initiative, our 101 guidance, and our Patent Trial and Appeal Board proceedings. We’ve also brought a broader range of services to support American innovators where and when needed, including through four regional offices across the country, and through over a dozen IP attaches across the globe. For those of you who may not know, our attaches in China, the EU, India, Brazil, and other locations help U.S. innovators obtain and enforce IP rights outside the US, and also help advocate for improved IP laws outside the U.S. . . .
Examining patents quickly and efficiently is only part of the job. The public relies on us to issue quality patents. Today, we have about a dozen programs underway that we believe will meaningfully improve patent quality. These include programs for making sure we’re getting the most relevant prior art before our examiners as early as possible by making prior art cited in our PTAB proceedings available to examiners handling related applications, and transitioning all our examiners from the decades old, antiquated U.S. Patent Classification System to the updated, increasingly global Cooperative Patent Classification System. Developing best practices such as for enhancing the clarity of the record. And developing new and better ways to measure our progress. For example, with our Master Review Form, we are now processing twice as many reviews, and capturing data not only about the correctness of our actions, but also about the clarity of our actions. This helps us identify trends, and helps us pinpoint very specific areas for training. Finally, and, importantly, for the first time in 40 years, we are doing a comprehensive review of the amount of time our examiners need to do their work in light of the many recent changes within the patent system. We are already seeing measurable and statistically significant results in our patent quality efforts even in the short amount of time since we began the Enhanced Patent Quality Initiative. And you can look forward to more improvements in the future. In sum, the USPTO is operating well, and is issuing more, higher quality patents than ever. …
What work lies ahead? Our top priority today is to make sure the Patent Trial and Appeal Board’s AIA proceedings are as effective and as fair as possible–within our Congressional mandate. Now is the right time to examine and make any needed reforms to these proceedings, because we have five years of data and experience to guide us. …
Let’s take a look at where we are with the AIA trials today. We have shared regular updates on the number of cases and the results of those cases, but we want to break down the numbers even further so that everyone has the facts, as there have been many numbers cited – some of which are accurate, some of which are not.
This waterfall slide is our attempt to present even more data in a more accessible and transparent format. According to waterfall slide data, which shows the status of every petition filed since beginning of AIA, of those petitions that have reached some sort of final disposition (i.e., are not pending), about one third of all petitions are denied institution and do not go to trial; about one third of all petitions settle either before or after the decision to institute; and only about one third of all petitions ever reach final written decision. It is only at that point of reaching final written decision is it relevant to talk about the patentability or unpatentability of the claims of a patent. And, at that point, it is true that 66 percent of petitions will find all instituted claims of the patent unpatentable, 17 percent of petitions will find all instituted claims of the patent patentable, and an additional 17 percent will have mixed results (some instituted claims will be found to be patentable and some will be found to be unpatentable). So statements that “all” or 95 percent or even 80 percent of patents are found unpatentable are not supported by the data and do not account for prior disposition by settlement or by denial of institution.
Drilling down further on the institution rate, here is some helpful data from the years the AIA trials have been in effect. As you can see, the institution rate in our proceedings has been steadily declining from a high of 87 percent in the beginning, to about two thirds today. These are the numbers, and these are the facts. Now, our stakeholders have understandably asked how the rates of unpatentability in AIA proceedings, reflect on the quality of all of the other patents we’ve issued. . . . [T]here are almost 2.8 million patents in force. Incredible. Of those 2.8 million about 4,000 patents have been challenged in AIA trials. This represents less than 0.2 percent of the patents currently in force. …
As I have said many times in the past, the USPTO is constantly looking for ways to make the AIA trials as effective and as fair as possible within our Congressional mandate. And we welcome your suggestions and ideas, the more specific, the better. To this end, we’ve set up a mailbox where you can submit your ideas at PTABProceduralReformInitiative@uspto.gov. This is an early opportunity to provide input. Your feedback here will help form the basis for our recommendations to the administration on next steps.
In fact, we have already heard from a number of you regarding a number of issues. As discussed:
Multiple petitions, and how best to curb abusive practices.
Whether to provide other opportunities to amend claims later in the proceedings when patent owners may have a better sense of where their claims stand,
Whether to reevaluate our claim construction standards in view of new data and experiences, and
How to best incorporate the findings and legal conclusions of prior proceedings at the USPTO, including examination and other post-grant proceedings, as well as, importantly, proceedings conducted in the federal courts.
What considerations should go into a decision to institute, for example, how can we better take into account considerations such as the interests of justice, economic factors, and hardship.
As to decisions to institute, whether there should be additional review of institution and termination decisions—and by whom.
When and how to permit parties to join proceedings.
When and under what circumstances might it be appropriate to extend the length of the proceedings beyond 12 months and what impact would extensions have on litigations before, not only the Board, but district courts. . . .
[W]e are [also] streamlining the way the Board identifies and designates opinions as precedential, with the goal of designating more opinions precedential. Why? Because doing so 1). Improves the consistency across panels of the Board, and 2). Provides greater notice to the public to help you make better informed decisions on how to manage your Board proceedings. . . .
Fees: Finally, we are also proposing increases to our AIA trial fees. . . . We believe, as many of you do, that these proceedings should be self-funded. I commend our team for making thoughtful initial cost projections and for recommending revisions to ensure that these proceedings are fully self-funded. There is much to do on AIA proceedings, patent quality, and in many other areas, and I promise you that we are working diligently, with a sense of purpose, every day.
In closing, we all want what’s best for this country’s inventors, innovators and entrepreneurs. Because it is through their efforts that our country remains at the forefront – economically and globally. Simply stated—and I think President Trump and Secretary Ross would agree—we want the United States to out-innovate and out-perform our global competitors, producing new jobs and new technologies and new industries that benefit all Americans, including individual inventors and those in our small businesses and startups. Let’s not forget that the modest businesses and unknown inventors of today may become the largest and most well-known drivers of our economy tomorrow. To make that happen, Congress, the courts, the administration (especially the USPTO), and even all of you here today, must all work together to make sure the system envisioned by our Founding Fathers lives up to its purpose to promote innovation. You have the commitment of the entire USPTO team that we will play our part to ensure that we continue to incentivize the kind of innovations that make our lives better and safer And that put men and women—like Mr. Iver Anderson with his lead-free solder—in the National Inventors Hall of Fame. Our future depends upon it. Thank you.
Once initiated, CBMs are identical to post-grant reviews (PGR) – allowing for patents to be challenged on any patentability grounds. As implemented, this includes 101 and 112 challenges in addition to the more traditional obviousness and novelty grounds. PGRs, however, are limited to only AIA-patents and must be filed within a 9-month window from issuance. Those caveats have severely limited the number of PGR petitions filed thus far. For CBMs, the AIA-patent restriction and 9-month window are both eliminated. However, the statute creates a subject-matter limitation that restricts CBMs to only non-technological financial-services business method patents.
Another feature of the CBM program is that it is “transitional” – i.e., it sunsets in 2020 and no petitions will be accepted after that date.
Last week, I hosted a quick anonymous survey on the transitional Covered Business Method Review program — asking whether the CBM program should be allowed to sunset or somehow extended. 240 Patently-O readers responded with results shown in the chart below. About 44% of responses favored ending of the program outright — allowing it to sunset. About 29% favored extending the program as-is, with the narrow financial-services scope. The remaining favored extension and expansion: 17% would expand the scope to include all information processing patents, and the remaining 10% would extend the program to include all patents. This final option would essentially mean ending the 9-month window for PGR filing.
The survey also offered (but did not require) an explanation of the answer. A variety of themes emerge from that explanation. The following are a few examples.
For patent challengers, the key response is that “it works” as a mechanism for cancelling patents, and could be extended to other technology areas.
CBM is a big success addressing one of the most abused categories of patents. Extend it to the very worst and most abused patents by including all of information processing and it can help clean up the system and make it stronger.
Business methods are not the only abstract processes being patented by the Office Patent. A majority of all information processing methods (even those outside of the Business arts) suffer from encompassing non-statutory abstract processes without reciting subject matter that amounts to anything significantly more than said abstract processes.
The historic problem associated with poor business method examination quality has now been fixed.
It was intended to handle a temporary problem in a specific area. State Street caused a flood of applications in an area that was new to the USPTO. Now skills and databases have developed and the stats show that there is no particular need for either expanding or extending CBM. Permanently singling out a particular subject matter for extra scrutiny could cause other countries to do the same in other areas.
If the goal was to clean up shoddy and overly broad patents and applications, then most all of the necessary work should be done by then. There are existing mechanisms in place that should be forcing quality such that this becomes redundant and therefore unnecessary.
It was a political sop to begin with and should be allowed to expire per the legislation and the underlying political agreement. It’s argument was to take care of “low-hanging fruit”, patents of old vintage, issued when the Office’s resources in this area were low. 8 years is more than enough time to pick that fruit.
CBM petitions are declining because most of the patents intended for consideration have already been undone.
Patents need to be strengthened, not weakened.
The patent systems is already nearly dead. Make patent owners in all areas feel the pain of having their patent rights trampled over by a kangaroo administrative court.
Broad restrictions on patentability are harming U.S. competitiveness in the areas of its greatest strength. China and the EU are poised to eat our lunch, and we are serving it up to them.
A terrible idea from the outset.
It (CBM) deprives some of the best technological innovators the chance to protect their valuable property. Abandon CBM, and instead seek recourse to the traditional approaches (102, 103 and 112) to rid the patent landscape of those patents that don’t rise to the level of technological innovation.
The PTAB process is either corrupt or incompetent.
It has been abused by petitioners and PTAB has taken it too far.
Go back to district court litigation. The present scheme is a disaster.
The USPTO is turning into a mini-court system. That is not its competency. It needs to focus on technology, granting patents to those inventions that meet the basic statutory criteria, and leave the legal hair-splitting to courts.
This is a corrupt Review that benefits a specific class of infringers and is detrimental to the development of new technology.
The approach should be ended because it violates the constitutional rights of patent owners.
AIA has overstepped its boundaries on constitutional grounds as patents are private rights.
All patent owners are entitled to due process, and that includes the right of access to a court of law before their patents are summarily cancelled by a political, the end-justifies-the means, so-called court.
Of course, there are other responses as well (perhaps more below in the comments).
The bottom line here, as you might expect, is that there is not yet any consensus on whether to extend the CBM program. My own general framework begins with the recognition that CBM does no longer adds much value post Alice/Mayo and with district court eligibility determinations being done on the pleadings. However, I would like to see the empirical evidence. The point of creating legislation that sunsets is that it effectively places the burden of proof on anyone wanting to continue the program. That work has not yet been done.
The Covered Business Method Review program is a transitional program that sunsets in 2020. These AIA trails have been extremely effective at knocking-out patents that qualify for review. The question of the day is whether Congress should extend and possibly expand the program beyond the 2020 deadline and beyond the non-technological financial services limitations.
I just read an interesting short article from Prof Gregory Mandel (Temple) titled What is IP For? Experiments in Lay and Expert Perceptions [available here]. Part of the papers includes survey results from both lay individuals and IP professionals who answered the question: What is the Purpose of Intellectual Property Rights? The vast majority of IP Professionals indicated the incentive justification for IP rights: The potential for strong rights sets out a carrot that encourages innovation. Law individuals, on the other hand, more often indicated that IP rights are designed to prevent or sanction plagiarism. Mandel also noted a severe lack-of-knowledge of US adults regarding intellectual property rights — with their study participants only barely outperforming random answers on a basic ten question IP law quiz.