chevron-down Created with Sketch Beta.
May 01, 2018 Feature

The Cost of Doubling Up: An Economic Assessment of Duplication in PTAB Proceedings and Patent Infringement Litigation

By Anne S. Layne-Farrar

©2018. Published in Landslide, Vol. 10, No. 5, May/June 2018, by the American Bar Association. Reproduced with permission. All rights reserved. This information or any portion thereof may not be copied or disseminated in any form or by any means or stored in an electronic database or retrieval system without the express written consent of the American Bar Association or the copyright holder.

With the Supreme Court’s recent ruling that Patent Trial and Appeal Board (PTAB) proceedings are constitutional,1 now is a good time to consider whether those proceedings live up to their policy promise. Congress’s goal for inter partes review (IPR) was for the process to be a “cheaper, faster” alternative to district court litigation.


It is a happy occasion when legislation achieves its original aims—and indeed conducting careful cost-benefit analysis before enacting legislation works to that end. Looking at a history of over five years of IPR trials, however, it appears that Congress has fallen short of its goal. As explained here, PTAB proceedings have reached some of the sought-after benefits, such as establishing a faster and cheaper process for patent challenges as compared to federal court litigation. But the goal that IPRs substitute for litigation has not been achieved: A significant number of patents face a duplicative and costly regime today, wherein challengers can file serial IPR petitions and, whenever elements of those petitions fail to be instituted, they can be repurposed for district court. While the PTAB has made some much-needed corrections regarding serial IPRs, other reforms are still needed.

First, the Good News

The data indicate success in regard to IPR proceedings costing less than district court litigation. A 2015 American Intellectual Property Law Association (AIPLA) Economic Survey found that the median cost of an IPR through a PTAB hearing was $275,000 and through appeal was $350,000.2 In contrast, even with relatively low stakes ($1–$10 million at risk), patent infringement litigation typically costs nearly $1 million through the end of discovery and $2 million through final disposition.3 As the stakes increase, so do the costs of litigation. For $10–$25 million at risk, median costs through discovery are $1.9 million and over $3 million through final disposition.4 If more than $25 million is at risk, taking the case through discovery typically costs $3 million, with a median cost of $5 million to reach final disposition.5 IPR proceedings are therefore clearly cheaper than patent litigation.

PTAB proceedings are also faster than the typical district court case. The America Invents Act (AIA) requires that the PTAB issue a final written decision on the patentability of any challenged claim within one year of instituting a trial. Only if the PTAB establishes good cause can it extend the one-year period by up to another six months. Patent trials, on the other hand, can be lengthy affairs. It is not uncommon for a trial to begin two to three years after the initial complaint is filed.6 And it can take 13–20 months just to receive a Markman ruling.7

Now, the Bad News

Of course, cost and speed matter, but what about Congress’s main goal that PTAB proceedings substitute for district court litigation? Here, the record is far less favorable. Some argue that “rather than an absolute alternative to civil litigation, the AIA’s new proceedings increase the likelihood of duplicative concurrent inter partes administrative and judicial adjudication of a patent’s validity.”8 Duplicative patent challenges within and across IPRs and district courts are the focus of this article.

Why Not Stay?

PTAB proceedings have become a popular counterattack strategy for defendants in patent infringement litigation. According to 2017 statistics, 85 percent of patents challenged in an IPR also are litigated in the federal courts.9 Hence, the potential for duplication is high.

While either or both parties in litigation can request a stay of the court trial pending an IPR ruling, such stays are not always granted. In a study of all district court cases in which a stay motion was filed between September 16, 2011 (the start of PTAB proceedings) and June 15, 2016, a total of 752 decisions had stay requests, with 529, or 70 percent, of those requests granted.10 Thus, in 223 cases (30 percent), the parties had to contend with simultaneous district court litigation and an IPR. Breaking the granted stay figure into stipulated and contested requests, the former comprised 45 percent, or around 238 of the 529 cases. Among this group, 97 percent were granted. Among the contested requests, however, the stay grant rate was only 58 percent. Moreover, it appears that stay grant rates are falling over time, as earlier reports had the rate from 2011–2014 at around 80 percent, 10 percentage points higher than the 2011–2016 70 percent figure.11

The AIA lays out specific guidance for staying litigation when a covered business method (CBM) review is underway, but the courts have applied similar logic in deciding stays for IPRs as well, albeit with considerable discretion.12 The four factors listed in the AIA are:

  1. Whether the stay will simplify issues in the litigation;
  2. Whether discovery is complete and a trial date has been set;
  3. Whether the stay will unduly prejudice the nonmoving party; and
  4. Whether the stay, or its denial, will reduce the burden of litigation for the parties or the court.

The third factor likely explains the far lower grant rate for contested stay requests. Because a stay delays infringement and damages determination, patent holders may be prejudiced. Prejudice assessments also can depend on the business model of the patent holder: “[C]ourts typically have found that any harm to nonpracticing entities can be addressed through monetary damages and prejudgment interest. These lines of reasoning have led some to suggest that stays against nonpracticing entities should be automatic.”13 On the other hand, even when the patent holder is a nonpracticing entity, the circumstances may warrant denying a stay, such as when prolonged infringement is expected to reduce the value of the patents for licensing or when the patents will expire soon.14

Courts also consider the amount of time a party waits to file an IPR petition and request a stay. “[S]ome patentees contend that the delay may cause harm to the patentee’s business, provide a tactical advantage to the defendant hoping to learn about a patentee’s litigation positions before entering the IPR, or be unfairly prejudicial if the parties have expended significant resources and are close to trial.”15

Timing (and Efficiency) Is Everything

The second factor the AIA looks at is the stage of litigation. Being either too early or too late in requesting a stay can lower the odds the request is granted. If the request to stay district court litigation comes before the PTAB has made a decision to institute the IPR, the district court judge is far less likely to stay the lawsuit.16 Institution is not automatic: Across all technology categories, the PTAB institution rate is currently around 64 percent.17 Thus, judges are reluctant to stay a trial before it is clear the IPR will move forward. On the other hand, if the district court case is already far along, then staying the trial is unlikely to result in any cost savings or efficiencies. And the main point of staying litigation is efficiency: “[T]he AIA furthers simplification by removing certain issues from litigation once the PTAB renders a final decision.”18

Finally, courts are more likely to grant a stay for litigation if the overlap between the IPR and the district court trial is relatively high. There is little efficiency or cost savings in staying a district court trial on wholly separate patent claims; thus, “the party desiring the stay will typically need to challenge at the PTAB all patent claims that are likely to remain in the district court action.”19 When all of the claims asserted in the suit also are challenged in an IPR, the IPR may resolve many of the patent issues, obviating the need for litigation,20 or the IPR at least may simplify litigation considerably or pave the way for settlement between the two parties. In contrast, for an IPR with fewer patent claims challenged than in litigation, or for litigation that involves nonpatent disputes (e.g., trade secret misappropriation or breach of contract allegations), the PTAB’s findings will not resolve the bulk of the issues being litigated, so there is likely little to be gained by staying the court case pending the IPR decision.

An Uneven Application

Despite the national reach of the criteria guiding the grant of a stay, district courts have applied these factors unevenly. Certain districts—including the Northern District of Illinois, the Northern District of California, the Southern District of Ohio, and the District of Massachusetts—have strong records of granting contested stay requests when an IPR is pending, ranging from 61–82 percent.21 Other districts, however, have dramatically lower rates. The Eastern District of Tennessee, the Eastern District of Texas, the District of New Jersey, and the Southern District of Indiana have granted contested stay requests at rates from 0–30 percent, respectively.22 In a post–TC Heartland23 world, such differences across the districts are likely to matter more.

Doubling Up Appears Inevitable

While stay request timing issues (too early or too late) are likely to be resolved as lawyers become more accustomed to court preferences and the IPR process, the extent (or lack) of overlap between an IPR petition and a district court lawsuit suggests that litigation stays will never reach 100 percent. Even with full overlap on patent claims, differences in the procedures themselves suggest that IPRs will never be able to replace federal court patent litigation. For example, “[i]n IPR, only printed publications can be used to attack patents. In district court, challengers can also attack with the other categories of prior art: prior public use or sale, public knowledge, derivation or incorrect inventorship, and invention by another.”24 As a result, real-world complications limit the extent to which the PTAB can replace litigation, even for those patents with instituted claims.

The Cost of Doubling Up

For the 30 percent of nonstayed court cases with IPR proceedings, the implications of doubling up can be significant. Even for those claims that overlap across the district court trial and the IPR, given the typical stakes, the parties cannot risk any reduction in efforts for either litigation or the IPR. For patent holders, that means incurring expenses for outside counsel, infringement experts, invalidity rebuttal experts, and a damages expert for the district court trial—as if no IPR were concurrently underway.

This is especially true because the PTAB follows a different and incompatible legal standard for patent review as compared to federal court. In addition to the prior art rules already noted above, validity standards differ as well. For instance, patents enjoy a statutory presumption of validity in district court, but in PTAB proceedings, petitioners must clear only the much lower bar of “preponderance of the evidence” that the claims are more likely than not unpatentable.25

Moreover, claims construction is handled differently between the two proceedings. Federal courts rely on the Phillips decision,26 whereas the PTAB applies the “broadest reasonable construction.”27 The PTAB’s rubric makes it easier to find invalidating prior art as compared to the litigation standard. Thus, an IPR and a district court case looking at the exact same patent claims and the exact same proposed invalidating prior art nevertheless can arrive at very different outcomes.

Duplicative PTAB Proceedings

The 30 percent no-stay statistic itself is misleading in that it glosses over the risks of repeat actions within the PTAB. Specifically, the 30 percent includes patents that have multiple IPRs filed against them. The U.S. Patent and Trademark Office (USPTO) reports PTAB statistics on a petitions-filed basis.28 So, for example, suppose that four overlapping petitions are filed with the PTAB on claims within patent number 123,456, but the PTAB decides to institute only one of these petitions. The USPTO would then report a 25 percent institution rate, as one out of four petitions was instituted. That is an accurate statistic, of course, but also a misleading one in that it ignores the institution rate by patent, which in this simple example is 100 percent. For patent holders, the latter statistic is arguably more informative of the odds they face in having to respond to an IPR petition instituted against one of their patent holdings.29

Serial Petitioners

Multiple IPR filings against the same patents are not simply a theoretical problem but one seen in practice. According to a special report by the USPTO, around 15 percent of patents with an IPR face two or more petitions.30 Sixteen percent of multiple petitions are filed after the PTAB decides to institute the first IPR.31 And around 5 percent of multiple petitions face multiple rounds of institution, where second, third, and higher petitions are instituted after the first-round decision has issued.32

Academic research slicing the data differently finds equally troubling statistics for serial petitions. For example, a 2017 study found that out of 5,173 petitions filed with the PTAB (including both IPRs and CBMs), 2,555, or 49 percent, reflect “serial petitions,” meaning a second (or higher) petition filed by the same petitioner.33 On a per-patent basis, out of 3,460 patents with an IPR challenge filed, 842 (24 percent) were “serially petitioned patents.”34 Among the patents with three or more IPR challenges, the serial petitions involved an overlap in claims, an overlap in the prior art asserted, or both.35

Academic review of serial PTAB petitions involving the same prior art also reveals a troubling inconsistency: “Out of 294 patents, 63 patents (21.43 percent) had 1 or more prior art references admitted after the Board had already denied this very same prior art” in an earlier petition.36 In other words, it appears the PTAB has not been applying a stable, reliable approach to accepting prior art as grounds for institution, which could encourage more serial petitioning.

The PTAB has recognized that “[a]llowing similar, serial challenges to the same patent, by the same petitioner, risks harassment of patent owners and frustration of Congress’s intent” in enacting the AIA.37 Such multiple proceedings may reflect harassment by a single party38 or coordinated/collusive efforts by a group of patent challengers.39 The AIA tools for dealing with multiple filings include petition joinders, consolidations, terminations, or stays, but until recently these tools have not been employed in any meaningful way.40

There are, however, signs of positive change. On October 18, 2017, the PTAB made its decision in General Plastic Industrial Co., Ltd. v. Canon Kabushiki Kaisha precedential. This move renders the seven factors for determining whether to reject follow-on petitions—first laid out in the PTAB’s May 2016 Nvidia v. Samsung decision—binding for future IPRs. Those factors include consideration of the PTAB’s finite resources and time constraints, whether the same petitioner previously challenged the same claims, and whether the petitioner should have known earlier about the evidence used.41 In General Plastic, the PTAB denied institution for five follow-on petitions covering the same two patents that General Plastic had challenged nine months earlier. The PTAB explained that “the absence of any restrictions on follow-on petitions would allow petitioners the opportunity to strategically stage their prior art and arguments in multiple petitions, using our decisions as a roadmap, until a ground is found that results in the grant of review.”

Duplicative Federal Court Proceedings

Multiple IPRs are not the only duplication worth worrying about though. We also see multiple lawsuits in the federal courts challenging patent validity. A key element of the repeat litigation risk derives from the rules on estoppel relating to IPR challenges. Section 315(e) in title 35 of the U.S. Code states:

(1) Proceedings before the Office.—The petitioner in an inter partes review . . . may not request or maintain a proceeding before the Office with respect to that claim on any ground that the petitioner raised or reasonably could have raised during that inter partes review.

(2) Civil actions and other proceedings.—The petitioner . . . may not assert . . . that the claim is invalid on any ground that the petitioner raised or reasonably could have raised during that inter partes review.

The Federal Circuit weighed in on estoppel in its 2016 Shaw v. Automated Creel decision.42 The court determined that an IPR proceeds in two phases: First, the PTAB evaluates the petition to determine whether or not to institute an IPR; second, the PTAB conducts the IPR itself and makes a final decision based on the evidence.43 With this determination, the court then interpreted the statute language “during the inter partes review” very narrowly, concluding that only instituted claims can be considered as “raised during review.”

Since Shaw, district courts have been divided in their treatment of estoppel.44 Some judges have followed Shaw even when it appears they disagree. For example, Judge Robinson worried in her Intellectual Ventures opinion that allowing the defendant to assert at district trial “references that were never presented to the PTAB at all (despite their public nature) confounds the very purpose of” the IPR.45 Nevertheless, Judge Robinson could not “divine a reasoned way around the Federal Circuit’s interpretation in Shaw” and refused to apply estoppel.46 After complaining that “Shaw’s narrow view of § 315(e) estoppel undermines the purported efficiency of IPR,” Judge Peterson in Douglas Dynamics, LLC v. Meyer Products LLC still refused to stay all noninstituted claims.47

Second Bite of the Apple

Others, however, have skirted Shaw. In Biscotti Inc. v. Microsoft Corp., Judge Payne explained that Shaw should be read “to exempt an IPR petitioner from § 315(e)’s estoppel provision only if the PTAB precludes the petitioner from raising a ground during the IPR proceeding for purely procedural reasons, such as redundancy.”48 And in Precision Fabrics Group, Inc. v. TieTex International, Ltd., Judge Schroeder applied estoppel arguing that “the PTAB’s refusal to institute an inter partes review is indicative of the weakness of TieTex’s claim of invalidity.”49

Contrary to Judge Robinson, Judge Jordan found in Parallel Networks Licensing LLC v. IBM Corp. that Shaw applied only to the “raised during” and not the “could have raised” element of estoppel.50 He argued that “[a]llowing IBM to raise arguments here that it elected not to raise during the IPR would give it a second bite at the apple and allow it to reap the benefits of the IPR without the downside of meaningful estoppel.”51 Aiming “to prevent that unfair result,” he “conclude[d] that IBM is estopped from asserting prior art references and combinations that it reasonably could have raised before the PTAB.”52

Concluding Thoughts

Since the Supreme Court has kept IPR proceedings at the PTAB alive, it is important to reflect on the five years of history with these proceedings. Yes, IPRs are fast and inexpensive, but the current state of affairs is neither “efficient” nor a “simplification” of the previous patent challenge process, and certainly does not meet the congressional goal of having IPRs substitute for federal patent litigation. Serial IPR petitions and uncertain litigation estoppel rules are particular concerns, meaning continued reforms are needed.


1. Oil States Energy Servs., LLC v. Greene’s Energy Grp., LLC, No. 16-712 (U.S. Apr. 24, 2018). In SAS Inst. Inc. v. Matal, No. 16-969 (U.S. Apr. 24, 2018), the Court ruled that the PTAB must issue a final written decision for every claim challenged by the petitioner.

2. Am. Intellectual Prop. Law Ass’n, 2015 Report of the Economic Survey 43 (2015) [hereinafter AIPLA 2015 Report]; see also Saurabh Vishnubhakat et al., Strategic Decision Making in Dual PTAB and District Court Proceedings, 31 Berkeley Tech. L.J. 45, 60 (2016). Note, however, that in late 2017 the USPTO announced increased fees for IPRs. Matthew Bultman, USPTO Hikes Patent Fees, Price of AIA Reviews, Law360 (Nov. 14, 2017),

3. AIPLA 2015 Report, supra note 2, at 37.

4. Id.

5. Id.

6. Mark A. Lemley, Where to File Your Patent Case, 38 AIPLA Q.J. 1, 17–19 (2010).

7. James C. Yoon, Wilson Sonsini Goodrich & Rosati, IP Litigation in United States 35 (Aug. 4, 2016),

8. Sarah E. Craven & Michael J. Flibbert, Limits on Concurrent Administrative and District Court Patent Challenges under the AIA’s New Post-Grant Proceedings, IP Litigator, Sept./Oct. 2013.

9. David P. Ruschke & William V. Saindon, U.S. Patent & Trademark Office, Chat with the Chief: An Analysis of Multiple Petitions in AIA Trials 10 (Oct. 24, 2017) [hereinafter USPTO Chat with the Chief],; see also Vishnubhakat et al., supra note 2, at 72.

10. Morgan Lewis, 2017 PTAB Digest: The Latest Trends and Developments in Post-Grant Proceedings 26 (2017),

11. “Litigation proceeding in parallel with an instituted IPR is stayed about 82 percent of the time.” Brian J. Love & Shawn Ambwani, Inter Partes Review: An Early Look at the Numbers, 81 U. Chi. L. Rev. Dialogue 93, 94 (2014). The more recent Morgan Lewis study places that figure around 70 percent (see supra note 10).

12. Jeffrey Miller & James Lyons, Parallel Track Proceedings: Determining Whether to Stay Litigation, in Kaye Scholer, PTAB Monitor: Developments in Inter Partes Review Practice (2016),

13. Jason Lang & Blake Davis, The New Wrinkle for Patent Defendants Seeking Stays Pending PTAB, Recorder (Apr. 6, 2016).

14. Id.

15. Benjamin A. Saidman & Jason E. Stach, Maximizing the Likelihood of a Litigation Stay Pending Inter Partes Review, IP Litigator, Sept./Oct. 2016,

16. Jeffrey Kushan, 3 Years of Coordinating Litigation with PTAB Proceedings, Law360 (Sept. 14, 2015),

17. See Patent Trial & Appeal Bd., U.S. Patent & Trademark Office, Trial Statistics: IPR, PGR, CMB 7 (May 2017) [hereinafter PTAB Trial Statistics],

18. Kushan, supra note 16.

19. Id.

20. Miller & Lyons, supra note 12.

21. Morgan Lewis, supra note 10, at 28.

22. Id.

23. TC Heartland LLC v. Kraft Foods Grp. Brands LLC, 137 S. Ct. 1514 (2017).

24. Samson Vermont, IPR Statistics Revisited: Yep, It’s a Patent Killing Field, (Feb. 8, 2017),

25. Michael J. Flibbert & Maureen D. Queler, 5 Distinctions between IPRs and District Court Patent Litigation, Corp. Couns. (Dec. 16, 2015),

26. Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005).

27. Id. at 1316.

28. PTAB Trial Statistics, supra note 17; see also Peter Harter & Gene Quinn, How IPR Gang Tackling Distorts PTAB Statistics, IP Watchdog (Apr. 5, 2017),

29. In a special study on multiple petitions, the USPTO found that the by-patent institution rate is currently 70 percent. See USPTO Chat with the Chief, supra note 9, at 23.

30. Id. at 14.

31. Id. at 17–19.

32. Id. at 21.

33. Jay P. Kesan et al., Serial Petitioning at the PTAB: Joinder, Denial, Precedent and Finality, Presentation at the 7th Annual Patent Conference (PatCon7) 2 (Apr. 7, 2017).

34. Id. at 3.

35. Id. at 7.

36. Id. at 16.

37. Gen. Plastic Indus. Co., Ltd. v. Canon Kabushiki Kaisha, Nos. IPR2016-01357 et al., Paper No. 19, at 18 n.14 (P.T.A.B. Sept. 6, 2017).

38. Gene Quinn, PTAB Refuses to Institute Harassing IPR Challenges against Finjan, IP Watchdog (Feb. 1, 2017),

39. Vishnubhakat et al., supra note 2, at 48.

40. In over half of the petitions filed up through 2016, § 315 options were never even discussed in the PTAB proceedings. Kesan et al., supra note 33, at 26, 27, 31, 32.

41. NVIDIA Corp. v. Samsung Elecs. Co., No. IPR2016-00134, Paper No. 9 (P.T.A.B. May 4, 2016).

42. Shaw Indus. Grp. v. Automated Creel Sys., Inc., 817 F.3d 1293, 1296 (Fed. Cir. 2016).

43. Shearman & Sterling LLP, District Courts Split on Scope of IPR Estoppel, Mondaq (Mar. 22, 2017),

44. For more on this, see Jonathan Bowser, Under the Same Precedent, Different Estoppel Standards Have Developed in the PTAB and the District Courts, Buchanan PTAB Rep. (Feb. 27, 2017),

45. Intellectual Ventures I LLC v. Toshiba Corp., 221 F. Supp. 3d 534, 553–54 (D. Del. 2016).

46. Id.

47. No. 14-cv-886-jdp, 2017 WL 1382556 (W.D. Wisc. Apr. 18, 2017). For a similar ruling, see Verinata Health Inc. v. Ariosa Diagnostics, Inc., No. 3:12-cv-05501-SI (N.D. Cal. Mar. 8, 2017).

48. No. 2:13-CV-01015-JRG-RSP, 2017 U.S. Dist. LEXIS 144164, at *22 (E.D. Tex. May 11, 2017).

49. Nos. 1:13-cv-645, -650, 2016 WL 6839394 (M.D.N.C. Nov. 21, 2016).

50. No. 1:2013cv02072, 2017 U.S. Dist. LEXIS 28461 (D. Del. Feb. 22, 2017).

51. Id. at *30.

52. Id.

Anne S. Layne-Farrar

Anne S. Layne-Farrar is a vice president in CRA’s Chicago office and an adjunct professor at Northwestern University Pritzker School of Law. Dr. Layne-Farrar specializes in the economics of intellectual property and antitrust.


The author thanks Jay Kesan, David Long, David Schwartz, and Koren Wong-Ervin for helpful comments and suggestions. The views expressed in this article are the author’s alone and should not be attributed to any of her affiliations.