(Today’s guest post is by  Professor Jorge L. Contreras of the University of Utah College of Law.  Professor Contreras is known for his excellent work on remedies, particularly in the SEP and FRAND context.  I’m especially excited about today’s guest post, as it dovetails nicely into a one-week course on Remedies in Patent Law at Iowa Law in early January, taught by another leading expert on remedies law, Prof. Tom Cotter of Minnesota. – Jason) 

On December 21, 2017, the U.S. District Court for the Central District of California released its long-awaited Memorandum of Findings of Fact and Conclusions of Law in TCL Communications v. Ericsson (SACV 14-341 JVS(DFMx) and CV 15-2370 JVS (DFMx)).  In a lengthy and carefully crafted decision, Judge James Selna sets forth some important new points regarding the calculation of fair, reasonable and non-discriminatory (FRAND) royalties for standards-essential patents (SEPs).  Among other things, the decision offers a strong endorsement of “top down” methodologies for the calculation of SEP royalties, and makes significant use of the non-discrimination (ND) prong of the FRAND commitment in arriving at a FRAND royalty rate.  Equally importantly, the case establishes that, for non-discrimination purposes, even low end vendors like TCL will be considered “similarly situated” to high end vendors like Apple, giving them the benefit of the rates that high end vendors can negotiate with SEP holders for far more expensive consumer products.

Background

The case involves the sale of cellular handsets by TCL, a Chinese firm reported to be the seventh largest global manufacturer of mobile phones.  Ericsson is one of the largest holders of patents essential to the implementation of the 2G, 3G and 4G wireless telecommunications standards published by the European Telecommunications Standards Institute (ETSI) (standards-essential patents or SEPs).  Under ETSI’s policies, ETSI participants are required to grant licenses under their SEPs to implementers of ETSI standards on terms that are fair, reasonable and non-discriminatory (FRAND).

In 2007, TCL obtained a 7-year license under Ericsson’s patents covering ETSI’s 2G standards.  In 2011, the parties began to negotiate a license under Ericsson’s 3G SEPs, and in 2013, these negotiations expanded to include Ericsson’s 4G SEPs. Over the next several years, the parties were unable to reach agreement on the terms of this license, and during the course of negotiations, Ericsson sued TCL for infringement of its SEPs in six non-U.S. jurisdictions.  In March 2014, prior to the expiration of TCL’s 2G license, TCL filed an action in the Central District of California seeking a judicial declaration that Ericsson breached its obligation to offer TCL a license on FRAND terms.  TCL agreed to abide by the court’s determination of FRAND terms for a worldwide license under Ericsson’s 2G, 3G and 4G SEPs (slip op. at p.9).  Partially based on this assurance, in June 2015 the court entered an “anti-suit injunction” against Ericsson, prohibiting it from pursuing further infringement litigation against TCL until the resolution of the FRAND issues (I discuss TCL’s anti-suit injunction here). The court ruled that the nature of TCL’s claims was equitable (p.8), making it suitable for judicial (rather than jury) determination, and a 10-day bench trial was held in early 2017.  The court’s decision was rendered in November 2017, and a public version was released in December 2017 in which certain competitive information was redacted.

FRAND Royalties

Numerous U.S. cases have made clear that a FRAND royalty must be “premised on the value of the patented feature, not any value added by the standard’s adoption of the patented technology … [so that] the royalty award is based on the incremental value that the patented invention adds to the product, not any value added by the standardization of that technology” (p. 108, quoting Ericsson v. D-Link, 773 F.3d at 1232-33 (Fed. Cir. 2014)).  Unlike the recent UK decision in Unwired Planet v. Huawei ([2017] EWHC 711 (Pat), 5 Apr. 2017] (which I discuss here and here), which held that there is but a single FRAND rate applicable to any given set of parties and SEPs ((¶804(4)), Judge Selna in TCL v. Ericsson holds that there is no single FRAND rate (p. 109).

Top-Down vs. Bottom-Up Royalty Calculations

There are two general schools of thought regarding the calculation of SEP royalties subject to FRAND commitments: bottom-up and top-down.  “Bottom-up” approaches attempt to assess the value of asserted SEPs in isolation, using comparable license agreements and other methodologies, but without significant reference to other patents covering the same standard (I critique bottom-up methodologies here and here).  In contrast, top-down approaches first determine the aggregate royalty that should be paid for all SEPs covering a particular standard, and then allocate an appropriate portion of the total to the asserted SEPs (I discuss top-down royalty calculations at length here, as do Norman Siebrasse and Tom Cotter in this recent chapter).

Top-down approaches were used by the UK court in in Unwired Planet and by the Japanese IP High Court in Apple Japan v. Samsung (2014) (both discussed here).  And in November 2017, the European Commission emphasized in its Communication on SEPs that “an individual SEP cannot be considered in isolation. Parties need to take into account a reasonable aggregate rate for the standard, assessing the overall added value of the technology” (p. 7). However, with the exception of In re. Innovatio IP Ventures (N.D. Ill. 2013), most U.S. courts making FRAND royalty determinations have used bottom-up methodologies heavily dependent on an analysis of comparable licenses (e.g., Microsoft v. Motorola (9^th Cir. 2014), Ericsson v. D-Link (Fed. Cir. 2014)).

In TCL v. Ericsson, Judge Selna largely adopts the top-down methodology proposed by TCL (see below). He notes that the “appeal of a top down approach is that it prevents royalty stacking”, which occurs when individual SEP holders each demand a royalty that, when combined, can be excessive (p.15).

However, the court also notes that top-down methods cannot assess whether the licensor complied with the non-discrimination prong of the FRAND commitment.  Accordingly, Judge Selna undertakes a separate non-discrimination analysis based principally on the review of comparable licenses (discussed below).  He then combines the top-down and comparables approaches to determine the appropriate FRAND royalty rate.

The Aggregate Rate

A top-down royalty calculation methodology has two steps:  determining the aggregate SEP royalty applicable to a standard, then allocating an appropriate portion of the total to the asserted SEPs.  As I have discussed before, the UK and Japanese courts that applied top-down methodologies in FRAND cases based their aggregate rates on public statements made by SEP holders and other market participants.  Judge Selna also adopts this approach, citing various public statements and press releases by Ericsson that support an aggregate royalty of 5% on the 2G and 3G standards and a rate between 6% and 10% on the 4G standard (pp.19-26). While the court acknowledges that this method “is not perfect” (p.25), one of its merits is its dependence on statements made by Ericsson itself to induce the market to adopt standards covered by its own SEPs (p.25) (for a discussion of “market reliance” on FRAND commitments, see this paper).

Allocation of Ericsson’s Proportional Share

Once an aggregate royalty rate for all SEPs covering a standard has been determined, the appropriate portion must be allocated to the SEPs asserted in the case.  In TCL v. Ericsson, this determination involved two contentious steps: determining the total number of SEPs covering each standard (the denominator), then determining Ericsson’s share of those SEPs (the numerator).  The percentage of SEPs held by the SEP holder is the quotient of the numerator divided by the denominator.

Essentiality. It is well known in the literature that many patents declared by their owners as “essential” to a particular standard are, upon closer inspection, not really essential at all (up to 80% in some cases).  This is the problem of “over-declaration”, and it occurs because there is no verification by any third party of the essentiality of patents declared by their owners to be SEPs. As a result, courts considering total royalties attributable to SEPs covering a standard must also consider how many patents are actually essential to the standard.

Optional Portions.   An initial question addressed by the court is whether patents covering optional portions of a standard should be considered “essential” to the standard.  After analyzing the specific language of the ETSI policy, the court concludes that patents covering optional portions of an ETSI standard should not count as SEPs (p.27).

Essentiality Sampling.  Instead of analyzing the essentiality of each patent declared essential to the 2G, 3G and 4G standards, TCL’s experts sampled one-third of the patents covering each standard for each of the fifteen largest patent holders.  Thus, of 7,106 declared patent families covering user equipment, TCL analyzed the essentiality of approximately 2,600 patent families.  After various forms of cross-checking, it determined that a total of 413 patent families were essential to the 2G standard, 1,076 to 3G and 1,673 to 4G (pp.28-29).  Interestingly, it appears that TCL’s experts charged approximately $100 per patent for this analysis (p.30), which is significantly lower than the $10,000 per patent that is generally acknowledged as the cost of essentiality analyses for patent pools (some figures are collected here). One of the reasons for the low cost of TCL’s analysis was that TCL’s experts reviewed only the claims of the examined patents, not the full specifications.  Given that a review of patent specifications could have resulted in additional patents being found non-essential (p.31), the court adjusts the totals downward to arrive at 365 SEPs covering 2G, 953 covering 3G and 1,481 covering 4G (p.32).

Ericsson’s Share.  To compute Ericsson’s share of SEPs covering the relevant standards (the numerator), the parties’ experts determined which of the SEPs already identified would be owned by Ericsson during the term of a 5-year (60-month) license (p. 37).  Under the holding of Brulotte v. Thys, 379 U.S. 29 (1964), which prohibits post-expiration patent royalties, the court eliminates from Ericsson’s total any patents that expired prior to the date of closing arguments (May 18, 2017) (p.36). Interestingly, the court did not require the elimination of expired SEPs from the total number of SEPs (the denominator).  It explained that “[b]ecause the total aggregate royalty represents the value of all expired and unexpired inventions in the standard, … removing an expired SEP from the denominator treats the invention as no longer having value.  The invention however still has value, that value has merely been transferred to the public domain.  To remove expired patents from the denominator (without decreasing the total aggregate royalty) would result in transferring the value from expired inventions to the remaining patents in the standard instead of the public.” (p.36).

Interestingly, while the parties agreed that Ericsson held 12 2G SEPs, they disagreed with respect to the number of 3G and 4G SEPs SEPs held by Ericsson (TCL finding 19.65 3G SEPs and 69.88 4G SEPs, and Ericsson finding 24.65 3G SEPs and 111.51 4G SEPs) (p. 37).  In any event, even using Ericsson’s estimate of approximately 150 SEPs, this is a relatively modest share of the 3,162 patent families essential to the 2G, 3G and 4G standards.

Relative Strength.  TCL argued that Ericsson’s proportionate share should be adjusted based on the relative importance of Ericsson’s SEPs compared to other SEPs covering the standards at issue (pp. 38-40) (this concept was introduced by Judge Robart in Microsoft v. Motorola, in which the court evaluated both the importance of the asserted patents to the standard and the importance of the standard to the overall product).  Though Judge Selna did not accept TCL’s methodology for gauging the importance of Ericsson’s SEPs, it did concede that “Ericsson’s patent portfolio is certainly not as strong or essential as it has claimed” (p. 43).

Geographical Variance

The court recognized that Ericsson’s patent strength was greatest in the U.S. and therefor determined that a discount rate should be applied to Ericsson’s FRAND royalty outside of the U.S.  It reasoned that “a global patent rate that does not account for differences in national patent strength provides the SEP owner a royalty based on features that are unpatented in many jurisdictions” (p. 44). For the sake of simplicity, the court divided the world into three regions: U.S., Europe and Rest of World (ROW) and established precise discounts for non-U.S. regions for each standard (e.g., for ROW, Ericsson’s 2G value share is 54.9% of the U.S. value)  (p. 45). This approach is significantly more fine-grained than that taken by the UK court in Unwired Planet, which divided the world into just two categories: Major Markets (U.S., Japan, Korea, India and several European countries) and all other countries, including China.  The FRAND rate for non-Major Market countries was simply 50% of the Major Market rate.

Violation of “Fair and Reasonable” Prong of FRAND

Even though the court does not accept each of TCL’s methodological steps in its top-down royalty analysis, the court finds, on the basis of those portions of the analysis that it accepts, that Ericsson’s offers to TCL are not “fair and reasonable” under its ETSI FRAND commitment.

Non-Discrimination

The court next analyzes whether Ericsson’s offers to TCL complied with the non-discrimination prong of its FRAND commitment.

Similarly Situated. As noted above, a FRAND license must be non-discriminatory.  This means that the licensor must not discriminate against similarly-situated licensees (p. 54). In TCL v. Ericsson, the court undertakes the most detailed analysis to-date to identify which firms are similarly situated with the potential licensee.  First, it concludes that the basis for comparison must be “all firms reasonably well-established in the world market” [for telecommunications products] (p. 56).  The court expressly excludes from this group “local kings” – firms that sell most of their products in a single country (e.g., India’s Karbonn and China’s Coolpad) (p. 59).  The firms that the court finds to be similarly situated to TCL are Apple, Samsung, Huawei, LG, HTC and ZTE (p. 58). Ericsson argued that Apple and Samsung are not similar to TCL given their greater market shares and brand recognition, but the court rejects that argument, reasoning that “the prohibition on discrimination would mean very little if the largest, most profitable firms could always be a category unto themselves simply because they were the largest and most profitable firms” (p. 61).

 The court found Ericsson’s licenses to Apple and Huawei to be suitable benchmarks for comparison to its offers to TCL (p. 91).  This conclusion is critical, because it establishes that low end vendors like TCL will be compared with high end vendors like Apple as to FRAND rates, giving low end vendors the benefit of favorable rate packages that high end vendors have been able to negotiate with respect to far more expensive products.

Competitive Harm.  Ericsson argued that in order for an instance of discrimination to violate Ericsson’s FRAND commitment, it must have the effect of “impairing the development of standards” (p. 91).  A similar systemic approach was taken in Unwired Planet, in which the UK court held that a violation of FRAND would not arise unless discriminatory treatment of licensees would “distort competition” (¶501). Judge Selna in TCL v. Ericsson takes a different view, holding instead that  discrimination in violation of a FRAND commitment can be found so long as an individual firm is harmed.  He expressly rejects the application of an antitrust-based standard, which requires harm to competition rather than harm to a competitor, to the analysis of a FRAND commitment (p. 91).

Comparison to Ericsson’s Offers. Though the options offered by Ericsson were complex and involved both lump sum payments and royalty floors within certain ranges (making them difficult to compare to other licenses), the court estimated that under one option, Ericsson’s offer to TCL translated to a running royalty on handsets of approximately 1% for 2G, 3G and 4G, and under another option 0.8% – 1.0% for 2G, 1.2% for 3G and 1.5% for 4G with a $2.00 per unit floor and a $4.50 per unit cap (p. 90).  The royalty floor proposed by Ericsson was apparently intended to address TCL’s low selling price for handsets, so that Ericsson would receive an assured royalty stream no matter how cheaply TCL priced its handsets.  Slightly different royalty schedules were proposed for external modems (p.90).

Discrimination.  Based on this analysis, the court holds that Ericsson’s offers to TCL “are radically divergent from the rates which Ericsson agreed to accept from licensees similarly situated to TCL” and that Ericsson’s offers to TCL were therefore discriminatory and noncompliant with its FRAND obligations (p. 94). In particular, the court holds that Ericsson’s proposed “floor” on royalties payable by TCL was discriminatory (p. 113).  This being said, the court also finds that Ericsson negotiated in good faith and that its conduct during the negotiations did not violate its FRAND obligations (p. 3).

Having concluded that Ericsson’s offers to TCL were not FRAND, the court proceeds to determine a FRAND rate for TCL’s desired license. It does so using a combination of the top-down rates derived above, as well as the comparable licenses reviewed in its non-discrimination analysis.  Below is a table containing the court’s final determination of FRAND rates for the different standards and geographic regions at issue (p. 104):

Royalty Base and SSPPU?  It is notable that the court’s decision in TCL v. Ericsson does not discuss the often contentious issue of the appropriate royalty “base” for TCL’s products – the figure against which the percentage royalty is applied.  As explained in cases such as Ericsson v. D-Link, parties often disagree whether the SEP holder’s royalty should be applied against a component (e.g., a chip) embodying the standardized technology or against an end user product such as a smart phone.  If the percentage royalty rate is not adjusted, the choice of the royalty base could result in radically different payments to the SEP holder. This concern has led to debates over the appropriateness of using constructs such as the “smallest salable patent practicing unit” (SSPPU) as the royalty base.  I understand that this debate was largely avoided in this case because TCL conceded that the royalty would be charged against the selling price of its handset units.

Holding and Conclusions

On the basis of these findings, the court prescribes that the parties enter into a 5-year license agreement reflecting the FRAND rates described above (p. 115).  In addition, TCL must pay Ericsson approximately $16.5 million for past unlicensed sales.

While the outcome of this case will likely make it easier for firms such as TCL to compete in the U.S. and other major markets, it also establishes several important guideposts for future FRAND license negotiations. First, the case establishes that, for non-discrimination purposes, even low end vendors like TCL will be considered “similarly situated” to high end vendors like Apple, giving them the benefit of the rates that high end vendors negotiate with SEP holders for much more expensive products.  Equally importantly, it highlights the growing predominance of top-down royalty calculation methodologies for FRAND licenses.

By Jason Rantanen

Here’s my addition to Dennis’s holiday patent post, based off materials I used for my Patent Law exam this year.   The technology involves recent interest of mine: making clear ice cubes.  Thanks to the cocktail renaissance of the 2000’s, there’s been a new interest among home bartenders in upping their game.   And what better way to show off than with a crystal-clear sphere of ice?

A day before to the exam, students were given a packet of documents that included a copy of a 2009 weblog post by Camper English, reprinted with permission.  English’s blog post talks about using “directional freezing” (described as freezing water from the top-down) in order to produce clear ice.  Students were asked to assume the following:

Camper English was the first person to discover that clear ice could be produced in a home freezer by freezing the ice in a directional manner.  English published these findings on a weblog on December 28, 2009, a copy of which you were provided in Appendix A.  English immediately filed a patent application that contained the following claim.

I claim:

1. A method of producing ice comprising freezing water in a directional manner in a home freezer.

Analyze the patentability of the claim under current patent eligible subject matter law.

***

The remainder of this series of questions revolved around an actual patent for producing clear ice spheres, Patent No 9,784,492, and a hypothetical infringement suit.  I found the concept embodied in the ‘492 patent quite clever: create an insulated two-part mold with a small hole in the bottom leading to a reservoir.  This setup produces clear, shaped ice. The ‘492 patent is assigned to Wintersmith (commercial embodiment).  Claim 1 reads:

1. A device for producing a piece of ice when the device is placed in a freezing environment, wherein the piece of ice has a shape, the device comprising:

Fig. 1 of Patent No. 9,784,492

Eric Sutton is a senior patent counsel at Oracle and an adjunct professor at Chicago-Kent College of Law. Any views or opinions expressed by him in this article are solely his own and do not necessarily represent those of Oracle Corporation, its subsidiaries or affiliates.

As known to most patent practitioners,[i] a parent provisional does not lock in a priority date for a child non-provisional unless the parent provisional provides a written description and enablement of an invention that is claimed in the child non-provisional. This assessment is conducted on a claim-by-claim basis[ii] to see if the parent provisional adequately supports and enables the precise subject matter of each child non-provisional claim for which priority is desired. If none of the claims in the child non-provisional are fully supported and enabled by the parent provisional, then the priority date for all of the claims is the later filing date of the non-provisional. In this scenario, the provisional has no clear value.

Under our first-inventor-to-file system, references that are published or patent applications that are filed between an applicant’s parent provisional filing and the applicant’s child non-provisional filing open the door for prior art to the claims of the child non-provisional filing, if those claims are not fully supported and enabled by the parent provisional.

Patent applicants, especially for software-related technologies, often file provisionals just before product releases, presentations at conferences, or publications of white papers or marketing materials. In-house patent portfolio managers frequently reassure business leaders that the technologies are protected by the provisionals, and outside patent counsel often similarly reassure in-house patent portfolio managers. However, due to time and budget pressures that often impact provisionals, relying on a provisional for priority may be significantly riskier than relying on a non-provisional.

Consider the hypothetical example below:

In the example, suppose your company files many provisionals and, on average, files many non-provisionals — each about ten months after its corresponding provisional. You file the provisionals ahead of product releases, and your competitor follows up with very similar product releases within a few months. The competitor’s product releases and several publications about the underlying technology occur during the ten-month span between provisional and non-provisional filings. Any aspects of these publications that were not completely derived from your product would serve as potentially invalidating prior art to your non-provisional filings unless your provisional filings satisfy the written description and enablement requirements with respect to your non-provisional claims. This scenario could happen over and over again, and negative effects from this scenario could even be baked into your provisional and non-provisional filing strategies in a fast-paced product development environment.

In a big data analysis of provisional application quality, focused on comparing provisional application content to non-provisional claims using Innography®, LexisNexis® PatentAdvisor®, and self-developed software, I found that many patent attorneys seemingly do rely rather extensively on their provisionals.[iii] On average, applicants rely extensively on their provisionals by allowing an average of ten months of risk for intervening prior art by filing the first-filed child non-provisional ten months after the corresponding provisional. Whether or not that intervening prior art actually shows up is beyond their control, particularly if they have already let the cat out of the bag by sharing their technology with others.

How can we know, with high probability for a large set of provisionals, whether those provisionals provide support for their child non-provisional claims? While there are many approaches available, some are more or less time-consuming, more or less subjective, more or less scalable, and more or less reproducible. I chose to analyze a large chunk of application pairs[iv] to see whether word stems present in the non-provisional claims[v] were also present in the corresponding parent provisionals.[vi] I then further excluded word stems that often appear in patent claims to produce even more interesting results that focus in on the less common word stems that theoretically would be more troubling if missing from the provisionals. The results are objective for what they show, i.e., that provisionals are, on average, missing several word stems from the broadest independent claims of the non-provisionals.

The study does not conclude missing word stems or even missing uncommon word stems necessarily mean that there is no support or enablement in the provisional. On the contrary, the conclusion of this study depends on your personal experience and judgment. Suppose you know nothing about a large set of provisionals in your portfolio other than how many unique word stems and, more specifically, how many uncommon unique word stems from the broadest child non-provisional claims were missing from the provisionals.

My guess is that most people would be very comfortable if all of the word stems or uncommon word stems were present in the provisional (0% miss rate) even though it is still possible they have not supported or enabled the claims in particular scenarios. On the other hand, I think most people would be very uncomfortable if all of the word stems or uncommon word stems were absent from the provisional (100% miss rate) even though it is still possible that they have supported or enabled the claims. Different people will have different comfort levels between 0% and 100% depending on their own experiences and understanding, but at higher percentages one would likely begin to bet against priority rather than for it.

The study evaluated 1905 pairs of parent provisionals and non-provisionals, filed between March 2012 and July 2013.[vii] In the sample, there were 16,739 unique word stems in the first claims of the 1905 non-provisionals. For the parent provisional / child non-provisional pairs, an average of 7.0 unique word stems from claim 1 were missing from the parent provisionals out of an average of 49.8 unique word stems per claim, for a 13.3% unique word-stem miss rate. As mentioned above, applicants waited an average of 10 months between filing the first-filed non-provisional from the provisionals, which means applicants relied on these provisionals by taking on unnecessary risk, despite the significant average miss rates.

The chart below shows percentages of cases at different levels of word stem miss rates.

As shown, about 19.3% of cases were missing 25% or more of the unique word stems, a point that is likely beyond the comfort level of most patent practitioners. That said, 21.5% of cases had a 0% miss rate, indicating that a large group of patent practitioners is avoiding the written description and enablement risk that comes with varied language.

After a careful analysis of stems at different thresholds, it turned out that the top 3% of stems (516 stems) occurred 30 or more times in the sample and included stems that, for the most part, would likely not be considered relevant to a written description analysis, whether or not a synonym was present in the provisional. The choice of a cutoff level was subjective, but the cutoff itself was objectively applied in the sense that all terms above the cutoff were removed from the study and all terms below the cutoff remained in the study. The rationale used for the subjective cutoff level was to pick a level that was just beyond any level including the most common terms occurring in computer-related claims. Although the sample of cases was not limited to computer-related cases, computer-related terms excluded as being part of the top 3% were, for example:

comput, store, storage, devic, service, server, system, machin, network, send, receiv, transmit, communic, provid, transit, instruct, engine, modul, configur, process.

Using the list above as a proxy for familiar common terms was helpful in picking a cutoff point even though a significant portion of the sample did not involve computer-related innovation. Stems beyond the top 3% boundary remained in the study. These less common stems occurred 29 or fewer times in the sample of 1905 claims.

Should the miss rates go up or down after removing the common stems? My hypothesis was that miss rates would go down because terminology such as “compris” (excluded as within the top 3%) might often be added by attorneys down the road as claims are solidified. At a minimum, more cases would have a 0% miss rate when fewer stems are searched. Others predicted that the miss rates would go up for other buckets because common terminology is usually present in the provisional, at least as boilerplate. Therefore, removing the common terminology would reveal the true incompatibility between provisionals and non-provisional claims.

I was not willing to tolerate the same miss rates as I did in the analysis for all stems. Because we are focusing on only the stems that occur relatively infrequently (in 29 or fewer out of 1905 claims), it is significantly more likely (though not certain) that these stems would reflect content that could be relevant to written description or enablement issues.

Others were right to think the results would get worse. Removing the top 3% of stems reduced the average number of unique stems per claim by 27 stems (45.8%) to 22.8 stems. Out of those 22.8 uncommon unique stems remaining, on average, provisionals were missing 5.3 for an average miss rate of 17.7%, up by 4.4%. Personally, I was shocked at this result and felt uncomfortable on behalf of the patent bar.

The chart below shows percentages of cases at different levels of uncommon word stem miss rates.

As shown, about 26.7% of cases were missing 25% or more of the unique word stems. The 26.7% uncommon stems mark is up from the 19.3% mark for all stems, reflecting a delta of 7.4%. If having 25% or more uncommon stems missing is beyond the comfort level of most patent practitioners, then practitioners should be uncomfortable with at least 26.7% of the provisionals being filed. I was equally as shocked to see that 9.8% (up by 5.5%) of cases were missing 50% or more of the uncommon stems. The largest cluster was 24.1% (up by 2.6%) having a miss rate of 0%, although there were a few miss rates even at 100%.

Due to the skewed distribution with miss rates falling at both ends of the spectrum, statistical metrics such as standard deviation did not provide much insight into the dataset beyond the charts shown. Nevertheless:

• a miss rate of 21% (30% of the sample) was greater than one standard deviation away from the peak of 0%,
• a miss rate of 34% (19% of the sample) was 1.645 times the standard deviation away from the peak (considered significantly different than the peak of 0% for a right-tailed distribution, p = .05, if the peak miss rate of 0% is considered to be the target). Note this analysis incorrectly treats the 100% miss rates as deviating from the target, when at least these cases clearly have a different target. This unnecessarily inflates the miss rate required to qualify as significantly different.
• a miss rate of 42% (14% of the sample) was greater than two standard deviations away from the peak.

U.S. ISSUES

It turns out that, in the U.S., the patent office does not even check for provisional application priority unless there is intervening prior art. That said, priority issues do come up in courts where litigants are equipped with discovery and additional resources.[viii] Based on the results of this study, it seems that there is some low hanging fruit where the patent office could object to a priority claim that is extremely unlikely to be accurate and, after patent office review, turns out to be inaccurate regardless of whether the patent office has found intervening prior art.

INTERNATIONAL ISSUES

Although I am not qualified to opine on the laws of specific other countries, I do know that many other countries generally have a stricter view on specification support for the purposes of preserving priority. According to Heather McCann, a European Patent Attorney at EIP: “Having a mismatch in the literal terminology of the U.S. provisional and a child PCT application or a child European direct national filing could significantly reduce the possibility of obtaining patent protection and successfully defending a validity challenge, post-grant, if the contents of the U.S. provisional application were published (even by the applicant) prior to the child PCT or child direct national filing.”[ix]

Unfortunately for U.S. applicants, unless they are keen on priority issues, non-U.S. patent offices might be able to reject cases from U.S. applicants more frequently based on a difference between common U.S. practice and foreign written description and enablement laws.

INEQUITABLE CONDUCT

Because inequitable conduct involves intent, which is fact-specific, this article does not delve any deeper into this potential issue. That said, it would be interesting to hear an argument and supporting explanation against such a finding if the applicant incorrectly claimed priority even though 100% of the uncommon word stems were missing from the provisional.

GENERAL CONCLUSIONS

The study is an objective report of terminology disagreement between provisionals and non-provisionals. Individual attorneys will vary on how much terminology disagreement they find comfortable in light of their understanding of 35 U.S.C. 112(a).

With 27% of the provisionals missing a quarter or more of the broadest non-provisional claim’s uncommon unique word stems, there is much room for the patent bar to either improve provisional application quality or avoid inaccurate priority claims. Perhaps this could also be accomplished by prioritizing the conversion to a non-provisional or filing incremental provisionals as more details become available, rather than waiting 10 months for the conversion.

[i] For more information on the written description and enablement requirements, particularly as they apply to software-related innovation, see Sutton, E. Software Patents: A Practical Perspective (ver. 2.2, Sept. 2017), available on Amazon at http://a.co/9xcPNr9 (a software patents textbook in support of a law course offered at Chicago-Kent College of Law).

[ii] Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center v. Eli Lilly and Company, 849 F.3d 1049 (Fed. Cir. 2017) (inadequate written description in provisional); Phigenix, Inc. v. Genentech Inc., 238 F.Supp.3d 1177 (N.D. Cal. 2017) (inadequate written description in provisional); See Storer v. Clark, 860 F.3d 1340 (Fed. Cir. 2017) (provisional not enabling); Dynamic Drinkware, LLC v. National Graphics, Inc., 800 F.3d 1375 (Fed. Cir. 2015) (“priority claims are not examined by the PTO as a matter of course, and consequently are not entitled to a presumption of adequate written description support in the provisional application,” and “[a] provisional application’s effectiveness as prior art depends on its written description support for the claims of the issued patent . . . .”).

[iii] N=1905, down from an original sample of 2341 after excluding: (a) unparseable provisionals, (b) clearly incomplete claims (for example, a method claim ending at the colon), and (c) later-filed provisionals that paired with the same first-filed non-provisional; provisionals filed Mar. 2012 to July 2013 across all companies, spanning the AIA transition; raw provisional pdfs available from PAIR, provided by Innography® in Jan. 2017; filing dates provided by Innography® directly and through use of their excellent patent portfolio analytics tool; sample spread over TC 1600, 1700, 2100, 2400, 2600, 2800, 3600, and 3700.

[iv] N=1905, same sample as above, comparing claim 1 of the non-provisional to the entire contents of the provisional application pdf, drawings included, based on Official Character Recognition (OCR) performed in batch by Adobe® Acrobat® 9 Standard, version 9.0.0. Provisionals were paired up with their first-filed non-provisionals based on data provided by Innography®. The claims of the non-provisionals were downloaded from Innography®.

[v] The first claims of the first-filed non-provisionals were stemmed using a popular word stemming library to result in unique word stems.

[vi] The text of the corresponding provisionals, as a result of the Adobe® Acrobat® 9 Standard OCR (see above), was searched for any text that would match the unique word stems. Spacing issues and formatting issues were corrected before performing the comparison.

[vii] The provisionals involved were filed between March 2012 and July 2013, bridging the AIA transition. If attorneys changed practice from pre-AIA to post-AIA, those changes were not sufficiently detectable in this sample. Although I personally believe that attorneys have not changed practice post-AIA; to fully test that hypothesis, I would need a sample significantly earlier than the AIA and a separate sample significantly later than the AIA. These samples simply were not available to me when I started the study.

[viii] Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center v. Eli Lilly and Company, 849 F.3d 1049 (Fed. Cir. 2017) (inadequate written description in provisional); Phigenix, Inc. v. Genentech Inc., 238 F.Supp.3d 1177 (N.D. Cal. 2017) (inadequate written description in provisional); See Storer v. Clark, 860 F.3d 1340 (Fed. Cir. 2017) (provisional not enabling); Dynamic Drinkware, LLC v. National Graphics, Inc., 800 F.3d 1375 (Fed. Cir. 2015) (“priority claims are not examined by the PTO as a matter of course, and consequently are not entitled to a presumption of adequate written description support in the provisional application,” and “[a] provisional application’s effectiveness as prior art depends on its written description support for the claims of the issued patent . . . .”).

[ix] See, for example, European decision T 1983/14 “Purchase of mobile top-up/JANUARY PATENTS LTD.” (Nov. 23, 2015) (“The case law of the Boards of Appeal requires that the subject matter be directly and unambiguously derivable by the skilled person from the disclosure as a whole. The expressions ‘invoiced together’ and ‘charged for together’ . . . disclosed no more than that transactions could take place on the same machine[, not] multiple products being invoiced as one transaction.”); European decision T 2061/13  (June 20, 2017) (“subject-matter of claim 1 is . . . not directly and unambiguously derivable by the skilled person . . . such that . . . claim 1 does not have a right to priority from P1.”).

By Jason Rantanen

…is still claim construction in my book.  This is true even when the claim construction focuses less on the specific meaning of individual words or phrases and more on construing the invention as a whole.

***

One of the most enjoyable things about teaching a course like Patent Law, which I’m doing again this fall, is that you get to re-read all the old classics and work through them from first principles with students who are not (yet) locked in the dogma of accepted narratives.  In other words, to see cases such as Mayo, Myriad, and Alice through fresh eyes, or to explore the evolution of post-Markman claim construction from Markman to Cybor to Phillips to Teva.  Today we worked through some of the examples that the PTO has published–for those who aren’t aware, there’s a host of them on the PTO’s website.

As I’ve read through the patent eligible subject eligibility cases again, I’m struck by the extent to which the courts engage in what I can only describe as claim construction–but a very different type of claim construction than what’s contained in the countless opinions deciding the meaning of individual claim terms over the last two decades.  Those cases address the meaning of individual terms or phrases in a patent claim, often in a way that is dispositive of an infringement, novelty or nonobviousness issue.

The different type of claim construction that I’m talking about is what courts necessarily do when conducting a patent eligible subject matter analyses (and perhaps nonobviousness and enablement as well): construe the invention from the claims rather than interpret a particular word or phrase. None of the judicial opinions talk about what they’re doing as claim construction, but that seems to me to be exactly what it is.

The idea that courts describe patent claims in words other than those of the claims themselves during patent eligible subject matter inquiries is nothing new–to the contrary, it’s a frequent complaint about the Supreme Court’s patent eligible subject matter cases.  Usually, it’s referred to as determining what the claims are “directed to,” or, in the second part of the Mayo/Alice inquiry, the search for an “inventive concept.”

But if claim construction is understood as the translation  of the words of a patent claim into text that will have meaning for the person deciding a legal issue such as infringement or validity, then discussions of the patent claims in the context of patent eligible subject matter analyses would seem to be exactly that.  Alice offers an obvious example: there, the Supreme Court explained that “[o]n their face, the claims before us are drawn to the concept of intermediated settlement,” (an abstract idea, failing Mayo step 1), and that “viewed as a whole, petitioner’s method claims simply recite the concept of intermediated settlement as performed by a generic computer” (thus failing Mayo step 2).  Put another way, the Court took the words of the claim, gave them legally operative meaning, and then assessed whether that legally operative meaning against the criteria of abstract idea and inventive concept.  That’s claim construction.

I can see some pushback here from folks who think of claim construction solely as the process of determining the meaning of individual words or phrases in a claim–in other words, as  questions of accuracy about the meaning of words.  The sharp focus on the meaning of individual claim terms has dominated thinking about patent law for the last two decades, but I think it’s a mistake to view claims only as individual limitations.  That approach is useful for inquiries such as literal infringement or novelty, but it’s of less use when conducting other types of patentability or validity analyses.  I’m not suggesting that claim term interpretation is irrelevant to patent eligible subject matter or enablement inquiries–sometimes the meaning of individual terms matters quite a bit.  But what I see courts doing again and again in the patent eligible subject matter cases is to read the claims to arrive at a construction of the invention, rather than resolve disputes about the meaning of particular terms.

My thinking in this area is heavily influenced by Andres Sawicki’s forthcoming article The Central Claiming Renaissance.  Professor Sawicki argues that the Supreme Court’s re-invigoration of patent eligible subject matter has been accompanied by a rebirth of “central claiming”: the idea that “courts use both the specification and the claims to situate the inventor’s work in the context of the technological field to understand just what it is that the inventor contributed.”  Id. at 10.  Sawicki observes that the Mayo/Alice analysis necessarily requires a determination of the inventor’s contribution–and particularly, whether that is the type of thing that is patent eligible subject matter.  Consistent with the idea of central claiming, he suggests that in identifying the inventor’s contribution, the Supreme Court looked mostly to the the specification when making this determination in Bilski, Mayo, and Alice,  even as he acknowledges the important role the claims play in what the Court was doing.  Id. at 23-35.

While I mostly agree with Professor Sawicki’s analysis of the Supreme Court’s patent eligible subject matter cases,  I think the article gets over-invests in the concept of “central claiming.”  As Sawicki recognizes, even when conducting the Mayo/Alice analysis, courts are still construing the actual claims, not the written description, to identify the invention.   They draw on the written description, but so too does conventional claim term interpretation.  I’m also hesitant because what’s involved isn’t “claiming;” it’s the process of construing an existent claim.   That said, I don’t have any sticky words to describe this alternate approach to claim construction.  The best I’ve come up with are “whole claim construction,” “invention-focused construction,” or “inventive contribution construction” as contrasted with “claim term interpretation.”

***

Why does it matter whether we call what the court’s doing “claim construction?” After all, a rose by any other name still smells as sweet.

There are a few reasons.  First, the old adage that “the name of the game is the claim” is just as true when courts are construing the claims in a general sense as it is when courts are engaged in term-based constructions.  As patent practitioners know, how the court articulates the claims for Mayo steps 1 and 2 matters–and can even be dispositive, as the recent case of Visual Memory LLC v. NVIDIA Corp. shows.  Recognizing that the courts are engaged in type of claim construction has the benefit of focusing the parties’ arguments on an issue that may determine the outcome of the case.

A second benefit from recognizing that courts are engaged in claim construction in the patent eligibility inquiry is that the discussion can then turn to what constitutes an abstract idea.  Sure, the court’s construction matters, but so too does the question of what an abstract idea actually is.  Untangling these two issues–the construction of the claim and the meaning of “abstract idea”–may bring clarity to the second one.

A third reason why we should acknowledge that there’s a type of claim construction going on in patent eligible subject matter analysis is because the Federal Circuit’s opinions don’t.  Recognizing that the courts are engaged in a type of claim construction raises an important question: we have a complex methodology and approach to interpreting claim terms; what then is the methodology for this other type of claim construction that the courts are doing?  As the patent eligible subject matter jurisprudence develops, methodological variants are emerging that reflect sharp differences in the way that claims are being construed when conducting the patent eligible subject inquiry.  Some of these divisions present the same types of issues as the conventional claim term interpretation cases raise, such as “how much weight to give the specification?,” while others present new divisions, such as the question of “at what level of generality the claims should be read?”

As a first pass at this idea, here are a few different ways that “invention-focused construction” can be done:

• At one extreme is the ipsis verbis methodology: to repeat the words of the claim and nothing but.  Only those words, and perhaps interpretations of some key terms, suffice.  One might picture an attorney standing before a judge who repeatedly asks “what’s the invention?” while the attorney dutifully recites the words of the claim again and again, neither understanding why the other doesn’t get it.
• At the other extreme is a claim construction untethered from the claims themselves.  An example might be to construe a claim to a typewriter as being to a hot dog.  While this example is absurd, the point is to illustrate that there is a range from ipsis verbis to a construction of the invention that has no connection to the claims.
• Another dimension of the invention-focused construction involves the degree to which the construction hones in on the inventor’s contribution.  To borrow from the USPTO’s gunpowder example, the “inventive concept” of a claim to “an intimate finely-ground mixture of 75% potassium nitrate, 15% charcoal and 10% sulfur” isn’t the individual ingredients; those already exist in nature.  Rather, the “invention” is the combination of the ingredients, or the specific ratios, or that they are finely ground in the mixture.
• Alternately, the court may treat the claim as a whole as the “invention,” not caring about which part of it constitutes the inventor’s contribution to the art.  I struggle with applying this approach: how do you conduct an analysis of whether the invention is eligible subject matter when you aren’t focusing in on what is new?

In my next post I’ll write more about how these different approaches recently manifested in the Visual Memory case that Dennis blogged about last month.

By Jason Rantanen

One of my current projects is to create a transparent and user-friendly database of information about Federal Circuit decisions–particularly, its patent law decisions–with a structure that’s useful for empirical researchers and other people who write about the court. If folks are going to throw around numbers about the court, my hope is that they have some grounding in a relatively clear and consistent methodology.

My goal in creating this database is to focus on information that’s of interest to researchers and commentators rather than to create another database for online legal research.  While Westlaw, Lexis, and the various free alternatives are great tools for finding key cases, there are lots of problems with using them for empirical research, not the least of which is the periodically changing nature of their interfaces–changes that make replication of older studies difficult. Also, since their focus is on helping practitioners do legal research, commercial databases tend not to collect and record data in a way that’s useful for empirical work.

I’ve used components of the dataset before, including in PatentlyO posts and an article on dissents at the Federal Circuit, and others have used it in their own projects (Crouch; Gugliuzza & Lemley).  Currently, coded data includes basic information about each decision arising from the district courts and PTO back to 2007, with written opinions going back to 2004.   Fields include information such as document type, precedential status (and Rule 36’s!), panel members, dissents and concurrences, and opinion authors.  I’m interested in crowdsourcing additional fields to code once I’ve published the dataset. Everything’s being done in-house at Iowa–even the R Shiny app that powers the user interface was coded as a project for a graduate statistics consulting class under the instruction of Rhonda DeCook, and my research assistant Louis Constantinou has been a huge help in getting this project finally done.

I’ll be rolling out the web-based portal in the near future, but in the meantime I’ve had requests for some of the data–particularly an update on the court’s use of Rule 36 summary affirmances.   Those charts are below.  The basic methodology employed in creating these charts is described in my earlier post on Federal Circuit appeals and decisions.

Finally, since the purpose in creating this database is to share it, students or other researchers who are interested in drawing from it for their own work are welcome to email me before it goes public.  Basically, it’s a starting point that’s flexible enough to allow future researchers to do their own coding without having to redo basic information about each decision.

Given the increase in appeals from the PTO over the past few years, this graph is not all that surprising–but it’s still quite dramatic.  As of mid-2017, the number of decisions in appeals arising from the PTO has exceeded the number of decisions from the district courts for the first time in the history of the Federal Circuit to my knowledge.

The following two charts are updated versions of figures from my post from last summer on types of Federal Circuit decisions.

By Jason Rantanen

I recently read Graham Moore’s The Last Days of Night, a tale about the patent war between Thomas Edison and George Westinghouse from the point of view of Paul Cravath, the young lawyer representing Westinghouse (and yes, that Cravath).  The book is historical fiction (which means that it consists of made up/condensed elements set within the broader landscape of historical evidence), but many of the themes within it ring true today.  Among them are divergent views of the role of patents in technological development.  Edison personifies the corporation-as-inventor, with his organized system for inventing.  Westinghouse embodies the practical aspects of invention–the need to take a technology and make it into a commercial reality.  And Nikola Tesla is the “flash of genius” inventor who invents for the love of inventing rather than for material rewards.

Here are a couple of memorable bits (and slight spoilers):

Thomas Edison was not, Paul thought, the first man to become rich by inventing something clever. Rather, he was the first man to build a factory for harnessing cleverness.  Eli Whitney and Alexander Graham Bell had each made his name by inventing one brilliant thing.  Edison had formed a laboratory that had invented many.  His genius was not in inventing; rather, it was in inventing a system of invention.  Dozens of researchers and engineers and developmental tinkerers labored beneath Edison in a carefully constructed hierarchical organization that he founded and oversaw.  (p. 172)

The “you can’t handle the truth” scene of Cravath’s deposition of Edison is a trifle over the top, but easily captures the tension between Edison and Westinghouse, inventor and improver:

“I hired the band; I booked the hall.  I advertised the show.  And you hate me because my name is on the poster.  Well, I say this: The light bulb is mine.  If the word ‘invention’ is to maintain even a semblance of rational sense, then it must be said that the light bulb was my idea.  It was my invention.  And it my patent.  Every bulb.  Every vacuum.  Every one of your piddling filaments.  And to the mute ingratitude with which you’ve repaid me, I will say only one last thing.”

Edison leaned back in his chair before he loosed his final words.

“You’re welcome.”

(p. 211)

In the author’s note at the end, Moore discusses historical sources and intentional fictionalizations for the sake of narrative.  And a few pages later, he thanks legal historians Adam Mossoff of George Mason and Christopher Beauchamp, now of Brooklyn Law School.  Professor Beauchamp will be giving a talk at the Iowa College of Law later this week about his actual-history book, Invented by Law: Alexander Graham Bell and the Patent that Changed America that I’m very much looking forward to.  If you’re looking for a work that aims for historical accuracy about patents and innovation in the late 1800’s, I recommend giving that one a read.

(I did not receive any compensation for writing this, even a copy of the book – which actually came from the library.)

By Jason Rantanen

A question from two economist friends, Nicholas Ziebarth and Michael Andrews, got me interested in the geographic distribution of patent practitioners in the U.S. and any correlations with issued utility patents and populations.  Using the January 8, 2017 list of patent practitioners from the USPTO, the PTO’s data on utility patents issued to inventors by state, and population estimations for 2016 (wikipedia), I put together the following figures.  They show what one might expect: patents, population and patent attorneys exhibit high degrees of correlation, although there is some interesting variation.  All the linear regressions are highly significant (p<0.001).  Some data caveats follow.

California is a clear outlier, as is probably D.C.  In the below charts I’ve removed California.  As you can see, the correlation coefficient goes down.  It’s not apparent due to the scaling of the graphs, but the slope of the line also goes down.  Removing D.C. (where there are lots of registered patent practitioners but few inventors) has the opposite effect when it comes to patents/patent practitioner.  To give another sense of the magnitude of the California effect, 19% (8124/43542) of this set of patent practitioners are California residents and 29% (40/196/140928) of this set of patents has a first named inventor who was a California resident.

(Iowa is marked in red.  We have a lower than average number of patent attorneys per population but a slightly higher than average number of patents with U.S. inventors per patent attorney.)

Notes on the data:

• This analysis entirely ignores patents whose first named inventor is not resident in the 50 states and District of Columbia and patent practitioners who do not reside in those places.  When it comes to patents, this is big: since 2009, over half of issued patents have had a first-named inventor who was not a U.S. resident. It’s less of an effect with patent practitioners: of the 44,897 registered practitioners, 43,579 are U.S. residents.
• The patent attorney list is subject to concerns about whether it is overinclusive, as discussed in Zachary Kinnaid’s previous posts, and how up-to-date the addresses are.  This comparison also involves data from different years (1/1/2017 versus patents granted in 2015), although there probably isn’t much a change in terms of geographic distribution of patent attorneys over the time. In addition, many registered patent attorneys do not currently engage in patent prosecution activities.
• The data on patents-per-state is based on the residence of the first named inventor.   This means that the other inventors are ignored in this analysis.  It also means that assignments are not reflected.

My data file is linked if anyone wants to play around with it.  Regression outputs from Stata are here.  If you see any mistakes, I’d appreciate it if you’d let me know via email.  You are free to reuse this dataset or graphs, although I would appreciate an acknowledgement.

By Zachary Kinnaird, Patent Attorney with International IP Law Group, PLLC

Based on my analysis of the USPTO roster data:

• There were 44,295 registered patent practitioners in the U.S. as of April 2016.
• After 20 years on the USPTO roster, at least 15%  of practitioners originally registered between 1988 and 1996 are no longer registered as of April 2016.
• Going back from April 2016, every subsequent year before 1986 shows an additional 2% of patent practitioners are no longer registered.
• The USPTO either skipped or removed a large number of registration numbers originally to be assigned in 1964, 1966, and 2001 for currently unclear reasons.

Retirement trends based on removals from the USPTO roster

The number of practitioners removed from the USPTO database reveals a practitioner percentage removal trend that can be seen as a retirement estimate for patent practitioners. This trend shows that the longer a practitioner has had a registration number, the more likely they are to have retired, or otherwise been removed, from the roster.

The chart below shows the percentage of patent practitioners who still remain registered on the USPTO roster as a function of each practitioner’s year of registration. The further to the right, the more recently the practitioner earned their registration number.

*Gaps years 1964, 1966 removed as likely outliers as discussed more below

** Value for 2001 shown in red as data for 2001 came from the USPTO Official Gazette rather than from a calculation from the USPTO database. Data for 2001 from USPTO database calculations was an outlier as discussed more below

From the graph above, there appears to be an initial sudden dip at 1996 that suggests that about 15% of practitioners leave the practice after 20 years of being registered agents and attorneys. These numbers stay fairly level for another 10 years on the roster, or back until 1986 before another large dip in the percentage of practitioners still registered. The trend after this first 30 years on the roster is a nearly linearly removal at a rate of approximately 2% a year. This retirement pattern continues until 55 years after initial registration, or 1960 in the above chart.

At the time this data was collected, there were 44,295 individuals in the USPTO database, however there are likely far fewer than this actively filing or prosecuting patents with the USPTO. The percentages of practitioners still on the roster are very likely to be higher than in actual practicing attorneys and agents. Remaining on the roster is relatively costless for a practitioner due to the lack of dues and continuation education requirements. It is likely that many practitioners have effectively retired without yet being removed from the roster. Accordingly, the percentages shown on the above graph are a ceiling of the actual figures of practitioners who are actually still practicing.

Tracking removal rates of registered attorneys from the USPTO roster

The above retirement graph was generated from percentages of practitioners removed from the USPTO database. Finding the percentage of those removed can be a challenge, however, as the USPTO only maintains the current state of the patent bar in its public database and does not publicly provide a list of all practitioners ever registered at the USPTO.

Instead, the Office of Enrollment and Discipline periodically conducts a mandatory survey of active registered practitioners pursuant to 37 C.F.R. § 11.11. This survey, in part, determines if the practitioner is still active and wants to remain on the USPTO roster. As noted above, there is essentially no cost to remain on the roster as there are currently no annual dues or continuing education requirements. However, if a practitioner is no longer practicing and either indicates the end of their practice in the survey or does not respond to the survey, the USPTO removes that practitioner from the roster.

The below graph is constructed with the knowledge that practitioners can be removed through this survey system and by assuming that the USPTO assigns numbers in a sequential order. This visualization shows the number of registration numbers removed each year compared to the number of registration numbers initially assigned in that given year. For instance, while the USPTO roster may include 1,210 practitioners who first gained registration numbers in 2015, the first registration number assigned in 2015 (i.e. 73,303) compared to the last registration number assigned in 2015 (i.e. 74,530) suggests that 1,227 registration numbers were actually assigned in 2015 and that some of these 2015 practitioners have already been removed from the database. Carrying out this calculation for each year going back to 1950 yields the graph below.

As noted above, there are unusual red spikes in 1964, 1966, and 2001. The red spikes corresponding to 1964 and 1966 are far out of place and I believe they are likely outliers. The red spike in 2001 IS an outlier. This can be said with confidence based on comparison to number of practitioners found in the USPTO Gazette for the year of 2001. Potential practitioners in the PTO Gazette are listed provisionally to allow time for the public to report any additional information about the practitioners before they formally receive a registration number. 1,853 potential practitioners were listed for 2001 in the USTPO Official Gazette compared to the calculated value difference of 2,697 between the first registration number assigned in 2001 and the last. The corrected graph below reflects the modified value for 2001 based on the Official Gazette listings. Unfortunately, the listing of potential practitioners in the online resources of the Official USPTO Gazette currently only goes back as far as 1995 and can’t be used to find values for 1964 and 1966.

The outlier variation in 2001 is likely due to a conscious choice by the USPTO to jump to a higher registration number for assignment rather than adhere to a generally sequential and quantum assignment of registration numbers. This is likely true for 1966 and 1964 as well. For example, at the end of 1966 the last registration number assigned was 23,099 which is followed by a jump to 24,002 on the very next day numbers were assigned. These gaps suggest that there may have been a USPTO policy at the time to skip blocks of numbers and perhaps start at the next round number in the sequence (e.g. 24,000 in 1967) at the start of a next year.

Data Retrieval Methodology

The USPTO provides both a database for practitioner information as well as a zip file of this data in spreadsheet form that is updated daily. However, the spreadsheet provided by the USPTO does not include the dates of registration as an attorney or agent. Accordingly, the process of retrieving the registration date information from the USPTO online database was automated with web macros to yield the data that is analyzed in this post. As many practitioners have been registered as both patent agents and as patent attorneys, only the date an individual first received a registration number was used for this analysis. For access to the data set being used, or if any errors are noticed, please feel free to reach out by email.

This article is a continuation and extension of my previous analysis on the USPTO database graphing the USPTO attorney and agent database as well as my update to that story as of April 30, 2016.

Editor’s note: Comment moderation policy is in effect.  Excessive or uncivil posts will be deleted without further notice.