Perryman PTAB Study for Unified Patents Leaves Out Half of the Story

Publisher’s Note: We did not receive the Unified Patents response until shortly before scheduled publication, which accounts for the answers to questions posed to Unified being appended to this article. At 8:10pm ET I updated the article to provided hyperlinks where appropriate to take the reader to the relevant response from Unified. 

___________

In late June, Unified Patents published the findings of an economic report conducted by consulting firm The Perryman Group on the supposed impacts of validity trials conducted under the America Invents Act (AIA) at the Patent Trial and Appeal Board (PTAB) on the U.S. economy. Unsurprisingly, the report is very bullish on the effects of the AIA, and Perryman pegs the positive impact of the AIA and the PTAB at $2.95 billion in terms of increased gross domestic product, $1.41 billion in increased personal income and 13,500 of additional job-years of employment.

Careful observers of the U.S. patent system, however, will note that the data only does a good job of examining one side of the PTAB’s impact. To its own detriment, the Perryman study does an arguably poor job at accounting for any losses to the U.S. economy through the invalidation of patent rights, which leaves inventors without an ability to generate licensing revenues or attract investment to grow their own companies, both of which are legitimate ends of the patent system. (See Unified Response.) Instead, Perryman conceives of the U.S. patent system as a zero-sum game wherein one patent owner’s loss is another business owner’s gain—a conceit which may work well to advance certain policy points, but which does an inadequate job of depicting reality.

The Other Half

The economic study, published as part of Unified Patent’s Patent Quality Initiative, estimates reductions to district court litigation as a result of AIA trials conducted at the PTAB, finding that the 10,085 cases handled by the PTAB between 2014 and 2019 led to a reduction of $262,200 per case, or $2.644 billion total, in litigation costs. These cost savings were then fed through an input-output model of the United States, obtained from the Bureau of Economic Analysis, to arrive at the study’s findings on gross domestic product, personal income and job-year increases. The study then splits these figures up by industry.

While a surface level look at the data compiled by Perryman would give a reader no reason to question Unified Patents’ thesis that the implementation of AIA trials at the PTAB has been a net benefit to the U.S. economy, the study pays zero heed to U.S. economic losses stemming from the invalidation of patent rights. Several statements within the report itself underscore this myopic presentation of facts. First, in the section discussing total economic benefits, the study says that “these estimated direct savings represent a net gain in efficiency (reduction in cost with no corresponding loss of output).” How can PTAB trials represent a reduction in litigation cost that has absolutely zero corresponding loss of output affecting the U.S. economy? The study doesn’t even mention loss of licensing revenue for patent owners who have lost all ability to exercise a property right against infringers. That has an obvious impact on the ability of patent owners losing those rights to grow their business, which in turn produces reductions on personal income and job-years that should have been accounted for by Perryman if the study was to produce a more accurate depiction of the AIA’s impact on the U.S. economy. To do otherwise would require utter ignorance of other economic models showing the impact of patent rights on increasing venture capital investment and creating millions of jobs across the United States.

Perryman Group’s misapprehension of this side of the story is magnified by another statement made one paragraph later, where the study acknowledges that “[a]ny economic stimulus, whether positive or negative, generates multiplier effects throughout the economy.” If this is true, the fact that the Perryman study doesn’t even attempt to put a gloss on the aforementioned negative economic stimuli is very telling. The billion-dollar figures touted by Unified Patents in net litigation savings and U.S. GDP growth are easily wiped out by even just a few cases at the PTAB, where patent assets worth billions of dollars have been invalidated.

Business Model Biases

From there, the study’s inherent bias against patent owners fighting to vindicate their patent rights becomes very easy to see. For example, a box on page 2 of the report states that “PTAB trials typically cost less to litigate and conclude within 12 months from institution, whereas district court litigation may take several years.” It is abundantly clear that this only represents the viewpoint of PTAB petitioners looking to invalidate patent claims. Patent owners overwhelmingly see the PTAB as an added expense on top of the already costly process of vindicating patent rights in district court. (See Unified Response.) Further, the idea that an AIA trial at the PTAB reaches a favorable result within 12 months of institution is another viewpoint that reflects only the interests of petitioners. For patent owners, whose legal costs at the PTAB begin to rack up at the petition phase, well before the PTAB decides to institute, the 12-month period between institution decision and final decision is often followed by many months, or perhaps years, of rehearing requests at the PTAB and appeals to the Federal Circuit. If, and only if, the patent rights survive this gauntlet, can the patent owner return to any district court litigation that may have been stayed pending the PTAB trial, which the Perryman Group study, to its mild credit, acknowledges “may take several years.”

The bias is clear. On the last page of the report, an about box description of Perryman touts the fact that the firm has worked with the 12 largest tech companies in the world; more than one-half of the Fortune 100; two-thirds of the Global 25; and the nine largest firms in the U.S. These are exactly the type of market dominant players that do not need a working patent system to succeed, and arguably thrive better when they can infringe patent rights with impunity. The fact that the Perryman study would willingly ignore certain inconvenient truths about the PTAB’s negative economic impacts is thus completely understandable.

Everyone is entitled to their own policy positions, just as our readers are entitled to a thorough conversation dissecting the various data points supporting those policies. It’s probably no secret that a study funded by a business with a profit model predicated on invalidating patent rights in a particular executive agency tribunal would be resoundingly positive on the economic impacts of that tribunal. However, pointing out the biases inherent in such a study is time well-spent for anyone who actually cares about how secure patent rights can better produce a stronger U.S. economy than the race to invalidation that the Perryman study suggests.

Here is what some other stakeholders had to say about the study:

Jim Carmichael, Carmichael IP and a former administrative patent judge: “The PTAB does not reduce the cost of litigation for my patent owner clients. For them, the PTAB is a crushing burden that adds to their costs. The delay in vindicating their rights can be a death knell for small businesses when the fruits of their labors have been stolen by efficient infringers. Knowing this, potential innovators are discouraged and the useful arts are diminished, not promoted. This diminishment has real ripple effects not accounted for in this Perryman report. The report uses unreliable assumptions that are employed to get to the result they want. The report seems to assume that all patents are bad. That’s the only way that it makes sense to say that it saves money compared to district court litigation. I don’t think that approach works for patents that protect true innovations and survive the PTAB.”

Dr. Kristina M. L. Acri, née Lybecker, Associate Professor of Economics, Department of Economics & Business, Colorado College: “In general, I find the analysis of the Perryman economic study to be quite ‘thin’ and the methodological appendix to be woefully inadequate. Overall, I feel as though there isn’t enough information provided about how the estimates were reached to be able to place much confidence in them or the methodology used. The industry categories used are far too broad to accurately describe the U.S. economy. The fact that 50% of the GDP impact is in manufacturing reveals very little. I imagine that the impact on manufacturing in pharma is very different than in autos or furniture. I would have liked to see this broken out at a finer level. Further, my understanding is that the study evaluated patent litigation costs and trademark litigation costs. I would expect those costs to differ substantially across industries and sectors. How did they create the average cost across industries? No justification is provided. (See Unified Response.)

“While only 10% of the cases reach the trial stage, I would again expect this proportion to differ by industry. (See Unified Response) It is unclear whether this was considered or accounted for. I appreciate the use of constant 2019 dollars and the use of historical data to estimate trends. Those assumptions strike me as very reasonable and workable.”

Paul Morinville, Founder, US Inventor: “Biased research was used to crash the patent system, thus sending the majority of early stage investment in technology to China. Now, biased research is used to maintain its destruction. The PTAB invalidates or neuters 84% of the patents it reviews. One side of the USPTO creates a property right. When the patent holder finally develops market value for the patent, the other side of the USPTO, the PTAB, invalidates the patent, effectively transferring that market value to huge multinational corporations. No inventor can trust the U.S. government when they see such blatant corruption.”

The Response

In an email conversation with Unified’s Chief IP Counsel Jonathan Stroud, Unified’s Co-Founder and COO Shawn Ambwani, and Ray Perryman, President and CEO of the Perryman Group and author of the PTAB study, IPWatchdog posed a series of questions on the issues raised in this article, which elicited the following responses, shown in red and reproduced here in their entirety:

Q: The Perryman study does not seem to account for lost value to the U.S. economy caused by the invalidation of patent rights. Does every single patent invalidated at the PTAB represent a net gain for the U.S. economy?

A:  There is no “gain” to the U.S. economy from invalid patents. Rather, to the extent that holders seek compensation for invalid patents and parties incur transaction costs for patents that were wrongly issued, it is a net loss.

Q: The Perryman study says that the direct savings created by AIA trials at the PTAB have “no corresponding loss of output”. This does not seem to square with studies like the September 2016 report from the U.S. Department of Commerce regarding the $6 trillion (38.2%) of U.S. GDP contributed by IP-intensive industries. Is it possible that the Perryman study overlooked any possible loss of U.S. economic output resulting from the invalidation of patent rights, especially given the importance of secure IP rights to the businesses contributing to more than one-third of the U.S. economy?

A: Legitimate IP rights support innovation and are a major and critical driver of the U.S. economy and have been since its inception. However, patent rights that are not valid are not beneficial. As noted, they are basically harmful in that they divert resources to compensation for illegitimate IP claims.

Q: If there are any such oversights in lost economic output, would that impact the Perryman study’s numbers on increased U.S. GDP, increased personal income and additional job-years of employment? Technology licensing and attracting venture capital investment are two legitimate ends of the patent system and the invalidation of patent rights would seem to damage the economic prospects of patent owners, at least, as to these legitimate business activities.

A: There is no lost economic value in appropriately conveying or maintaining patent rights. Owners of legitimate technology should be compensated and are integral to value creation. Paying for illegitimate property rights distorts the system and creates inefficiency. 

Q: Is Unified Patents aware of any patent owners who view the PTAB as a cost-effective alternative to district court?

A: Many of the largest holders of patents and leading innovators have availed themselves of the system to avoid district court when they felt that were being asked to pay others for technology that they believe not to be worthy of patent protection.

Q: An insert box on pg. 2 of the Perryman study says that “PTAB trials typically cost less to litigate and conclude within 12 months from institution, whereas district court litigation may take several years.” Does the Perryman study account for any added costs incurred at the PTAB by both petitioner and patent owner parties prior to the institution decision (petition phase) or by any post-proceeding motions for reconsideration or appeals to the Federal Circuit? Further, does the Perryman study account for any PTAB trials where patent claims have been upheld as valid and district court proceedings have been able to move forward afterwards? It would seem that, in those situations, the PTAB wouldn’t represent a cost savings if the district court case proceeds after patent claims are upheld as valid, especially if there are additional costs due to Federal Circuit appeals following the PTAB’s final decision.

A: We used all available information to account for total PTAB costs and total litigation cost and used some conservative assumptions to understate the net benefits (as detailed in the methodology section of the report. I am not sure that it is relevant and I have not seen any data (other than anecdotal narratives); litigation following a patent being upheld in a PTAB trial should be less expensive (with a greater incentive to settle, less time spent on validity, and cheaper discovery costs). In any case, we sought to measure net savings on a conservative basis.

Q: How did the Perryman study arrive at its industry-specific numbers available through the pie charts on pg. 7 and the table on pg. 8?

A: These are derived from the impact analysis using the public U.S. Input-Output Model maintained by the U.S. Department of Commerce, as described in the Appendix.

Q: The Perryman study indicates that it used trademark litigation costs as part of its methodology to estimate the cost of patent matters (pg. 11, first paragraph). Was any effort undertaken to account for any differences in patent and trademark litigation costs by industry sector in order to derive more accurate numbers for the financial figures on the PTAB’s cumulative impact on pgs. 7 and 8? Or are we assuming that the average costs of all trademark or all patent litigation are essentially the same for each industry sector?

A: What you suggest is not necessary. The way we used the trademark data is standard methodology (frequently referred to as a benchmark or yardstick approach in the antitrust world and commonly used in many applications of statistical analysis across multiple disciplines). In essence, patterns in patent and trademark litigation followed very similar patterns (and statistically significant) until AIA and then differed markedly. Thus, by examining how trademark cases evolved after AIA (which was very consistent with history) and controlling for other potential factors (most notably the Alice decision), we can observe how patent litigation would have evolved without the AIA and compare it to what actually happened. Obviously, we cannot observe the counterfactual situation, and our approach is analytically sound. As described in the report, we also examined the patterns relative to overall litigation and found them to be consistent.

Q: The Perryman study states that only 10% of patent infringement cases in district court reach late discovery or trial phases. Was any effort undertaken to account for possible differences in this percentage of cases proceeding to these later stages which might exist between different industry sectors?

A: We are not aware of that information being available, and in any case, it would not have affected the overall results. We captured the total net savings and allocated it using widely used and reliable data regarding IP in the relevant sectors. The only possible difference to the overall results from using a different allocation would be in the relative multipliers for the various sectors and there is not much variation. Thus, any difference in the results would be both minor and random (equally likely to be slightly lower or higher).      

Image Source: Deposit Photos
Copyright: Alexis84
Image ID: 90245832