Pursuant to the Request for Comments dated October 29, 2018, the R Street Institute respectfully submits the following comments regarding the U.S. Patent and Trademark Office’s proposed new motion to amend process. These comments make three points on the USPTO’s proposal.
First, in any post-grant proceeding, the Patent Trial and Appeal Board should perform an independent search and examination of any proposed claim amendments prior to granting a motion to amend and regardless of the presence of an opposing petitioner. An independent PTAB search fulfills the USPTO’s obligations to examine claims before issuing them, and there are multiple reasons to believe that petitioners to post-grant proceedings will lack sufficient incentives in all cases to fully oppose motions to amend.
Second, the USPTO should collect further statistics on the rates of motions to amend before adopting any new amendment procedure. The recent change to claim construction practice before the PTAB may in fact reduce the number of motions to amend filed, to the point where no further action on amendment procedure ends up being necessary.
Third, the USPTO should assess the costs of its proposed new amendment procedure. The procedure is likely to impose substantial costs on attorneys and the Office itself, costs that may very well not be warranted if, as hypothesized above, amendments actually become unnecessary in the mine run of post-grant proceedings.
I. The USPTO has a duty to examine any claim—even one in an amendment presented during a post-grant proceeding—before that claim issues in a United States patent.
The Request for Comments would permit newly presented claims to issue in view of an amendment during a post-grant proceeding, both when a post-grant petitioner remains in the proceeding and when the petitioner drops out but the PTAB declines to request examination. As explained below, this is improper; examination should be performed prior to the grant of any amended claims, for at least three reasons. First, it would allow unexamined claims to issue in patents, which is contrary to both law and policy. Second, it would mistakenly rely on petitioners to challenge amendments. The PTAB cannot rely on petitioners to dispute amendments because petitioners are not fully adverse parties and thus will often lack incentives to mount a full opposition to amended claims. Third, in many cases patent examiners are simply better than third parties at finding prior art to patent claims. Allowing claims to issue without examiner review, therefore, introduces a serious risk of erroneous patent grants.
A. The issuance of unexamined claims is contrary both to good patent policy and to the text of the Patent Act.
If a newly presented claim appears in an amendment during a post-grant proceeding, the PTAB has the duty to examine it in the same way that an examiner must search and examine any newly presented claim of an unissued patent application. This duty flows from the text of the Patent Act, the administrative role of the USPTO, and basic principles of patent policy.
Under 35 U.S.C. § 131, for any new patent application presented, the Director “shall cause an examination to be made.” It would circumvent this requirement to issue an unexamined claim after grant, when the claim would have been examined prior to grant. This is why post-grant procedures for substantively amending patents impose examination requirements. Ex parte reexamination, for example, is to be “conducted according to the procedures established for initial examination.” § 305. A reissue application is “examined in the same manner as a non-reissue, non-provisional application.” 37 C.F.R. § 1.176. Insofar as the Act directs the PTAB to determine “the patentability of…any new claim added” in all post-grant proceedings, it also requires examination in those proceedings. 35 U.S.C. §§ 318(a), 328(a).
In view of these statutes, failure to examine claims would be a failure of the USPTO to discharge its administrative duty. Courts have noted that patent examiners are those “whose duty it is to issue only valid patents.” PowerOasis, Inc. v. T-Mobile USA, Inc., 522 F.3d 1299, 1304 (Fed. Cir. 2008) (quoting Am. Hoist & Derrick Co. v. Sowa & Sons, Inc., 725 F.2d 1350, 1359 (Fed. Cir. 1984)). “[I]t must be remembered that the primary responsibility for sifting out unpatentable material lies in the Patent Office.” Graham v. John Deere Co. of Kan. City, 383 U.S. 1, 18 (1966). Indeed, the USPTO itself has stated that the role of its examiners is to “serve as advocate/protector of public interest with respect to intellectual property.” Were the PTAB to allow a newly presented claim to issue as a United States patent without an independent examination of patentability, the USPTO would be derelict in its duties to issue only valid patents and to sift out unpatentable material.
The task of examining claims before issuing patents is no mere bureaucratic red tape; indeed, examination is at the core of the fundamental policy concern of protecting the public’s freedom from the undue burden of incorrect patents. As the Supreme Court has said repeatedly, the public “has a paramount interest in seeing that patent monopolies are kept to within their legitimate scope” because erroneous patents tax or even forestall productive businesses and activities. Medtronic, Inc. v. Mirowski Family Ventures, LLC, 134 S. Ct. 843, 851 (2014) (quoting Precision Instrument Mfg. Co. v. Auto. Maint. Mach. Co., 324 U.S. 806, 816 (1945)) (internal quotations and alterations omitted). Issuing patents comprising unexamined claims, even in the course of a post-grant proceeding, would potentially fail to keep patents within their legitimate scope, thereby failing to promote future innovation and failing to protect individual liberties from the intrusion of a wrongful patent.
The PTAB may deny a motion to amend without conducting a search first. If the prior art of record is sufficient to render the amended claim unpatentable, then there is no need to locate more prior art. A search is required only before an amended claim is to be allowed.
B. Post-grant petitioners may lack incentives to contest the patentability of amended claims.
It is incorrect to suppose, as the Request for Comments does, that the petitioner to a post-grant proceeding will mount a sufficient opposition that will flesh out all possible issues with a proposed amendment during the proceeding. Petitioners will lack sufficient incentives to oppose amendments fully for at least two reasons.
First, it is possible that the amendment renders the petitioner’s activities noninfringing. In that case, the petitioner would have very little incentive to oppose the amendment, even if the amended claims would still be unpatentable and detrimental to other businesses or individuals. Indeed, a patent owner and petitioner could theoretically, as preliminary steps to a settlement, agree to claim amendments that would satisfy both parties’ interests. In those cases, the petitioner may still continue to participate in the post-grant proceeding to ensure that the original patent claims are indeed cancelled but permit the amendment to proceed to issuance without substantial opposition or examination.
Second, even if the petitioner would still infringe the claims of the proposed amendment, the petitioner need not oppose the amendment fully; it may be sufficient for the petitioner to oppose only the portion of claim scope that it would infringe. The petitioner, then, may not fill the record with references to prove the invalidity of the full scope of the claim.
Consider, for example, a proposed amendment adding the limitation that a relevant component is made of metal. If the petitioner’s allegedly infringing product is made of brass, then the petitioner may focus on prior art references describing the component being made of brass, without searching for references describing the same component being made of steel or aluminum.
Under current practice, the petitioner’s incentive misalignment makes no difference: The Board will ultimately reject the amendment so long as the petitioner’s identified prior art renders the amended claims unpatentable. But under the practice proposed in the Request for Comments, the patent owner has at least one opportunity to revise the proposed amendment in view of the prior art the petitioner identifies. The petitioner’s lack of interest in searching certain fields of the prior art—in the example above, steel or aluminum components—might then drive the patent owner to propose amendments in those unsearched fields. Absent its own independent prior art searches, the Board will be left with insufficient information about the patentability of the amended claims.
C. Experience with third-party prior art submissions proves that the USPTO cannot rely on outside sources to examine patent claims.
The USPTO’s experiences with third-party prior art submissions further demonstrate that the USPTO cannot rely on parties to an inter partes review to conduct sufficient prior art searches. As those experiences reveal, third parties often fail to find the most relevant prior art.
The first of these experiences is the Peer-to-Patent pilot program, which has operated since 2007. According to this program, certain patent applications are laid open to members of the public, who are then invited to submit prior art of relevance to the examiners reviewing those applications. The two-year anniversary report of the program found that for 66 applications within the pilot, the patent examiner relied on art identified by the public in only 18 out of 66 cases, and in 10 of those cases, the examiner found the same art independently. In other words, at least in the view of the relevant examiner, the public did not exceed the capabilities of the USPTO 88% of the time.
More recently, the America Invents Act introduced third-party submission practice, a structured procedure by which outside parties may submit prior art citations during the examination of applications. USPTO studies in 2013 and 2014 both found that approximately 86–87% of the time, examiners did not rely on art presented in these submissions.
This is certainly not intended to derogate the role of the public in identifying relevant prior art and bolstering the strength of patent examination; for the 12% or so of cases where the USPTO relied on public submissions, third-party art submissions demonstrably improved examination results. But the numbers do show that the USPTO has a critical role to play in searching prior art and examining claims for patentability. The Office should not ignore that critical role in the context of post-grant proceedings.
D. The USPTO must balance the need for examination against the statutory time constraints of post-grant proceedings.
Although the USPTO has a duty to examine claims before issuance, it also has a statutory duty to complete its post-grant proceedings within strict timeframes. See 35 U.S.C. § 316(a)(11). Neither obligation should yield to the other, so the Office must identify a way to fulfill both simultaneously.
Generally speaking, it is difficult to see why these objectives would conflict. Examiners are expected to complete examination of an entire application in under 31 hours. In the post-grant proceeding context, an examiner would perform a simpler supplemental search, which ought to require far less time than that required to examine an entire application. Accordingly, an internal examination step would likely not impose a serious time cost.
To the extent that the time for examining an amended claim would force a post-grant proceeding beyond the statutory time frame, the PTAB may deny the amendment as presenting an unreasonable number of substitute claims. A motion to amend that forces the USPTO to violate the statutory time limit is certainly unreasonable. See § 316(d)(1)(B). To the extent that the proposed amendments would in fact render otherwise defective claims patentable, the patent owner may initiate a reissue proceeding or supplemental examination to correct the patent. See §§ 251, 257; John E. Kind, Post Invalidation Reissue: An Underused Tool in the Patent Arsenal?, 57 IDEA 1, 19–22 (2016) (providing reasons why reissue should be available after AIA post-grant proceedings). Accordingly, denial of the amendment to meet the statutory time limit does no significant harm to the patent owner and ensures that the USPTO remains compliant with the law.
II. Insofar as amendment policy is closely tied to claim construction, the USPTO should not implement any new pilot program or rule change until its recent claim construction rule change has had sufficient time to percolate.
The potentially heavy costs to both parties and the USPTO warrant careful consideration of whether any change in amendment procedure is actually necessary. Indeed, there is strong reason to believe that patent owners’ need for amendments will largely dry up as a result of the recent change in the claim construction standard used before the PTAB.
Applying the construction standard of Phillips v. AWH Corp., 415 F.3d 1303 (Fed. Cir. 2005), in post-grant proceedings, as the USPTO decided it would do in November, obviates the need for substantial amendment practice. The district courts apply Phillips because they cannot amend patents. See In re Yamamoto, 740 F.2d 1569, 1572 (Fed. Cir. 1984). To the commenter’s knowledge, no patent stakeholder has proposed giving district courts a power to amend. As the USPTO itself has noted, amendments will generally be rare not only because of intervening rights, but also because Phillips largely offers adjudicators a degree of flexibility to interpret claims—particularly in ways that preserve the validity of the claims. Accordingly, multiple stakeholders and the federal courts themselves have recognized that availability of amendments is inversely tied to the application of Phillips, where one option at least partially substitutes for another.
Having decided to apply Phillips to post-grant proceedings, the natural expectation is that the need for amendments will decrease. The USPTO should observe the magnitude of that decrease before determining whether to introduce a new, untested, and likely costly motion-to-amend procedure. If there is a significant decrease in amendment practice, then the costs of the new procedure may not outweigh any possible benefits.
Accordingly, the USPTO should collect sufficient data on amendment practice following the change to the claim construction rule, before changing amendment procedure. Given the high degree of variability in quarterly amendment motion statistics, the Office would likely need to wait several quarters before it can statistically distinguish effects attributable to the claim construction rule-change from ordinary variance in amendment filing rates. The many statisticians within the Office are well-positioned to determine how much data would be optimal.
Should the USPTO alter motion-to-amend procedures too early, any decrease in amendment practice resulting from applying Phillips may be counteracted or even outweighed by an increase in amendment practice driven by incentives to protract litigation. As noted below, if the USPTO adopts the proposed amendment procedure, at least some patent owners will move to amend more often than they would otherwise in order to drive up costs for their opponents. Should that occur, it may become impossible to disentangle the effects of the claim construction change on amendment motion rates, frustrating any cost–benefit analysis.
The proposed pilot program may be a way of testing the relative rates of amendment filings between different procedural rules. But the already high variability in amendment practice would mean that the pilot program would have to be exceptionally large and operate for an exceptionally long time before the resulting data is statistically significant. If the new amendment procedure turns out not to be cost-effective, then the pilot program will have imposed unnecessary and potentially large costs on both parties and the USPTO without generating a benefit.
III. The USPTO must assess the costs that this proceeding will impose on all parties and the agency itself—all of which are likely to be unnecessarily high.
The Request for Comments makes no suggestion that the USPTO has estimated the predicted costs that its proposed procedural change would impose. The USPTO must estimate those costs because they are likely to be high for all parties involved, including the agency itself.
According to the proposed amendment procedure, the PTAB must write at least one new advisory opinion on the likely patentability of the amended claims, and the petitioner and patent owner must file up to four additional briefs in total. Furthermore, assuming that the USPTO agrees with the need to search prior to issuing amended claims for the reasons laid out above, the agency would incur the costs of examiner search time.
The costs imposed on the USPTO are likely to be considerable. Patentability analysis of amended claims is a substantial task, requiring interpretation of the amended claims, review of the specification for 35 U.S.C. § 112 compliance, possible review of the chain of priority to identify support in preceding applications, and consideration of prior art of record. The administrative patent judges would also need to confer with one another to reach a decision. Depending on how much time this requires, additional judges may need to be hired. If a search and examination is conducted, it would require resources from the Central Reexamination Unit, possibly including the hiring or elevation of more examiners to that unit. All of these costs must be assessed in determining whether to implement the proposed amendment procedure.
Attorney costs for both parties are also likely to be substantial. Modestly assuming a billing rate of $750 per hour for attorney time and 10 hours to prepare a paper for filing, the additional record under the USPTO’s proposed procedure will impose costs of $30,000 across both parties. Compared to the $275,000 in attorney fees that a post-grant proceeding is estimated to cost, the new amendment procedure would amount to an 11% increase in attorney fees. Given a rate of 100 amendments per year, the procedural change would introduce an annual cost of at least $3 million in attorney fees.
The overall cost is indeed likely to be higher, because there is good reason to believe that under the proposed new procedures, amendments would occur in post-grant proceedings at an increased rate. For one, a procedure that facilitates the grant of amendments will encourage motions for amendments. Furthermore, patent owners and their attorneys will likely exploit the obvious costs of this procedure, using amendment practice as a way of protracting litigation and thus dragging their opponents into settlement negotiations. The likely increase in filing of motions to amend will mean that the overall costs to petitioners, patent owners, and the USPTO are likely to be even higher than the baseline estimates above.
As explained in the previous section, amendment practice is set to decrease substantially in light of the USPTO’s claim construction change. Given the high potential costs of this new and untested amendment procedure and the likely possibility that the procedure would not be used but for its ability to drag out post-grant proceedings, the Office should carefully weigh the costs and benefits before adopting the procedure.
The commenter thanks the USPTO for the opportunity to submit these comments. The undersigned attorney would be happy to provide any additional information or answer any further questions about this matter as needed.
Counsel for the R Street Institute
1212 New York Ave. #900
Washington DC, 20005
 Sue A. Purvis, U.S. Patent & Trademark Office, Presentation: The Role of the Patent Examiner 9 (Apr. 8, 2013), https://www.uspto.gov/sites/default/files/about/offices/ous/04082013˙StonyBrookU.pdf.
 See Beth Simone Noveck, “Peer to Patent”: Collective Intelligence, Open Review, and Patent Reform, 20 Harv. J.L. & Tech. 123 (2006).
 See Naomi Allen et al., Ctr. for Patent Innovations, N.Y. Law Sch., Peer to Patent Second Anniversary Report 23 fig.5 (2009), http://www.peertopatent.org/wp-content/uploads/sites/2/2013/11/CPI˙P2P˙YearTwo˙lo.pdf.
 Presentation at the U.S. Patent and Trademark Office: America Invents Act Second Anniversary Forum 24 (Sept. 16, 2013), https://www.uspto.gov/sites/default/files/aia˙implementation/aia˙second˙anniversary˙forum˙slides20130916.pdf; Rory P. Pheiffer, Evaluating the Effectiveness of Third Party Preissuance Submissions, Nutter IP L. Bull. (Nov. 2, 2015), https://www.nutter.com/ip-law-bulletin/evaluating-the-effectiveness-of-third-party-preissuance-submissions.
 See U.S. Gov’t Accountability Office, GAO-16-490, Patent Office Should Define Quality, Reassess Incentives, and Improve Clarity 10 (June 2016). This is not to say that 31 hours is sufficient time for examination (it is not), but insofar as the Office allocates that much time for examination prior to grant, it would not make sense for the Office also to say it needs to allocate more time for post-grant examination.
 Changes to the Claim Construction Standard for Interpreting Claims in Trial Proceedings Before the Patent Trial and Appeal Board, 83 Fed. Reg. 51340 (U.S. Patent & Trademark Office Oct. 11, 2018) [hereinafter Changes to the Claim Construction Standard].
 See Changes to the Claim Construction Standard, supra note 6, at 51348, 51353.
 See, e.g., Comments of the American Bar Association at 2, Changes to the Claim Construction Standard, supra note 6, https://www.uspto.gov/sites/default/files/documents/comment-aba-ipl.pdf; Yamamoto, 740 F.2d at 1571–72.
 U.S. Patent & Trademark Office, Patent Trial and Appeal Board Motion to Amend Study 9–10 (Mar. 31, 2018) [hereinafter Motion to Amend Study], https://www.uspto.gov/sites/default/files/documents/PTAB%20MTA%20Study%20%28Installment%204%20-%20update%20through%2003-31-2018%29.pdf.
 Cf. Scott Graham, Skilled in the Art: Apple Goes OT in Clash of Titans. Plus Who’s Billing $1,750 an Hour?, Law.com (July 13, 2018), https://www.law.com/2018/07/13/apple-goes-ot-in-clash-of-titans-plus-whos-billing-1750-an-hour/.
 This is reasonably consistent with at least one estimate that budgeted $50,000 for PTAB motions practice. See Tom Engellenner, Presentation at the AIPLA Mid-Winter Institute Meeting: Comparison of Federal Court, ITC, and USPTO Proceedings in IP Disputes 21 (Jan. 2014), https://slideplayer.com/slide/1655615/.
 Am. Intellectual Prop. Law Ass’n, Report of the Economic Survey 38 (2015).
 This number is estimated from the USPTO’s data on the first half of 2018. See Motion to Amend Study, supra note 9, at 9–10.