Johnson v. McDonald

BARTLEY, Judge,

with whom LANCE, PIETSCH, and MOORMAN, Judges, join, dissenting:

Because the majority fails to apply the appropriate analysis for substantial justification under EAJA—whether, considering the totality of the circumstances, the Agency’s litigation and agency positions were reasonable at the time those positions were taken—I dissent. Although the majority cites the totality of the circumstances standard, in substance the majority focuses on one factor, the Federal Circuit’s merits determination in Johnson v. McDonald (Johnson II), 762 F.3d 1362 (Fed.Cir.2014), without considering all of the pertinent factors surrounding the government’s conduct. This approach represents the application of an erroneous legal test. As described below, because in this case the Secretary applied an incorrect but still reasonable interpretation of his own regulation in a close case of first impression, I would hold that the Secretary has met his burden of establishing substantial justification and deny EAJA fees.

I. APPLICABLE LAW

A. Principles of Substantial Justification Under EAJA

“[T]he substantial justification inquiry requires an analysis of the ‘totality of the circumstances’ surrounding the government’s adoption of a particular position.” Patrick v. Shinseki, 668 F.3d 1325, 1332 (Fed.Cir.2011) (quoting Smith v. Principi, 343 F.3d 1358, 1362 (Fed.Cir.2003)); see Roanoke River Basin Ass’n v. Hudson, 991 F.2d 132, 139 (4th Cir.1993); Secretary’s Response to EAJA Application at 10-11. “The standard to be applied ... for the issue of substantial justification is one of reasonableness....” Essex Electro Eng’rs, Inc. v. United States, 757 F.2d 247, 252 (Fed.Cir.1985); see Cline v. Shinseki, 26 Vet.App. 325, 326 (2013) (This “inquiry is designed to evaluate the reasonableness of ‘the position taken by the government on the issue on which the claimant prevailed.’”). “The totality of the circumstances, by its very description, does not exclude any valid issue from consideration.” White v. Nicholson, 412 F.3d 1314, 1317 (Fed.Cir.2005). “[A] position can be justified even though it is not correct, and ... it can be substantially (ie., for the most part) justified if a reasonable person could think it correct, that is, if it has a reasonable basis in law and fact.” Pierce v. Underwood, 487 U.S. 552, 566 n. 2, 108 S.Ct. 2541, 101 L.Ed.2d 490 (1988). “Put *146another way, substantially justified means there is a dispute over which reasonable minds could differ.” Norris v. S.E.C., 695 F.3d 1261, 1265 (Fed.Cir.2012). “Substantially justified” does not mean “justified to a high degree,” but rather means “ ‘justified in substance or in the main’—that is, justified to a degree that could satisfy a reasonable person.” Underwood, 487 U.S. at 565, 108 S.Ct. 2541.

Courts do not ask whether correct Government positions were substantially justified. See, e.g., Patrick, 668 F.3d at 1330 (substantial justification involves asking whether the Government “adopted a reasonable, albeit incorrect, interpretation of a particular statute or regulation”). Therefore, it is obvious that “[t]he Secretary’s failure to prevail on the merits of the case does not raise a presumption that his position was not substantially justified.” Bates v. Nicholson, 20 Vet.App. 185, 191 (2006); see also Pottgieser v. Kizer, 906 F.2d 1319, 1324 (9th Cir.1990) (“Failure to prevail does not raise a presumption that the government’s position was unreasonable.”); United States v. Yoffe, 775 F.2d 447, 450 (1st Cir.1985) (“Since fees are awarded only to a prevailing party, it follows that the fact that the government lost does not create a presumption that its position was not substantially justified.”). The Supreme Court has stated that “the fact that one court agreed or disagreed with the Government does not establish whether or not its position is substantially justified.” Underwood, 487 U.S. at 565, 108 S.Ct. 2541.

Moreover, no matter how strongly worded a court opinion invalidating a Government position on the merits may be, a subsequent court reviewing an EAJA application as to that position may not simply cite the merits opinion as proof that the position was not substantially justified, but must itself conduct an independent analysis to determine whether the Government’s position, at the time that position was taken, was “justified to a degree that could satisfy a reasonable person.” Id.; see Gonzales v. Free Speech Coalition, 408 F.3d 613, 620 (9th Cir.2005) (reversing District Court’s EAJA award that relied on the “clarity of the” Supreme Court’s merits determination because, “[t]o be sure, the Supreme Court soundly rejected the government’s arguments. But ... [b]y putting undue weight on the Supreme Court’s holding on the merits, the district court seemed to rely on hindsight, rather than an assessment of the reasonableness of the government’s position at the time of the litigation” (emphasis in original)); Taucher v. Brown-Hruska, 396 F.3d 1168, 1173 (D.C.Cir.2005) (“In considering substantial justification under EAJA, ... as in other areas[,] courts need to guard against being subtly influenced by the familiar shortcomings of hindsight judgment.”). Accordingly, unlike prevailing party status, which simply asks whether the Government lost, substantial justification asks a more nuanced question: even though the Government lost, at the time it made the decision under appeal, was it reasonable in the attempt? See Roanoke River Basin Ass’n, 991 F.2d at 139 (“The focus when determining whether a petitioner is a prevailing party is aimed at the degree of success obtained by the petitioner. Whether the government’s ‘position in the litigation’ is substantially justified, in contrast, focuses, not the on the government’s success or failure, but on the reasonableness of its position in bringing about or continuing the litigation.”).

Against this backdrop, this Court has identified several important factors to consider in the totality of the circumstances, “including merits, conduct, reasons given, and consistency with judicial precedent and VA policy with respect to such position, and action or failure to act, as re-*147fleeted in the record on appeal.” Stillwell v. Brown, 6 Vet.App. 291, 302 (1994); see also Cline, 26 Vet.App. at 327-32 (discussing and applying the Stillwell factors). The Court in Stillwell further recognized:

Two special circumstances may also have a bearing upon the reasonableness of the litigation position of ... VA, and of the action or inaction by ... VA at the administrative level. One is the evolution of VA benefits law since the creation of this Court that has often resulted in new, different, or more stringent requirements for adjudication. The second is that some cases before this Court are ones of first impression involving good[-]faith arguments of the government that are eventually rejected by the Court.

6 Vet.App. at 303. The Court also noted that the Board’s decision may issue at a time when “the statutory and regulatory framework presents a ‘confusing tapestry,’ ... in which the meaning is not easily discerned.” Id.; see Meinhold v. U.S. Dep’t of Def., 123 F.3d 1275, 1278 (9th Cir.1997) (“Consequently it is not fair to conclude that every violation of a regulation by an agency stamps its position as unreasonable. The government may avoid EAJA fees if it can prove that the regulation it violated was ambiguous, complex, or required exceptional analysis.”). To determine whether the Secretary is substantially justified, the Court must “consider the applicable Stillwell factors, as well as any other circumstances that may apply.” Cline, 26 Vet.App. at 327.

The Federal Circuit has held that “substantial justification is measured, not against the case law existing at the time the EAJA motion is decided, but rather, against the case law that was prevailing at the time the government adopted its position.” Bowey v. West, 218 F.3d 1373, 1377 (Fed.Cir.2000); see Owen v. United States, 861 F.2d 1273, 1275 (Fed.Cir.1988) (en banc) (finding that “the position of the government was substantially justified when it was taken, based on precedents then standing”); (Leamon) Johnson v. Principi, 17 Vet.App. 436, 442 (2004) (“As to VA’s application of the revised regulation in the instant case, the Court cannot find that the Secretary’s position at the administrative level was not substantially justified because ‘the Board clearly relied upon then-current law.’” (citation omitted)); see also Gonzales, 408 F.3d at 620 (EAJA requires an assessment of the “reasonableness of the government’s position at the time of the [Government action in question]” (emphasis in original)).

Although no single issue is dispositive, courts are more likely to find that the Government’s position was substantially justified where that position was invalidated in a case of first impression. See White, 412 F.3d at 1316-17 (affirming this Court’s finding of substantial justification where case invalidating regulation that Board had relied on in adjudicating EAJA applicant’s case was “a case of first impression and [there had been] no prior adverse reaction” to the regulation); Ozer v. Principi, 16 Vet.App. 475, 479 (2002) (finding the Secretary’s position substantially justified where, inter alia, the “case was one of first impression” and “there was no prior disapproval of or challenge to the [Secretary’s interpretation]”); Felton v. Brown, 7 Vet.App. 276, 283 (1994) (finding the Secretary’s position in a case of first impression substantially justified “[g]iven the statutory silence on the particular matter and the lack of a conflict with adverse precedent”); see also Cody v. Caterisano, 631 F.3d 136, 142 (4th Cir.2011) (“As the Government notes, litigating cases of first impression is generally justifiable.”).

*148This is not to suggest that the lack of previous challenge to an erroneous government position will always render the government immune from EAJA fees; rather, “whether a case is one of first impression is only one factor for the Court to consider.” Felton, 7 Vet.App. at 288. This factor furthers the commonsense principle that “[f]or purposes of the EAJA, the more clearly established are the governing norms, and the more clearly they dictate the result in favor of the private litigant, the less ‘justified’ it is for the government to pursue or persist in litigation.” Martinez v. Sec’y of Health & Human Servs., 815 F.2d 1381, 1388 (10th Cir.1987) (citation omitted). Thus, the cases construing EAJA recognize a balance between encouraging the vindication of rights by “creating a limited exception to the ‘American Rule’ against awarding attorneys fees to prevailing parties,” Underwood, 487 U.S. at 575, 108 S.Ct. 2541 (Brennan, J., concurring in part and concurring in the judgment), and not discouraging Government agencies from attempting to safeguard taxpayer money by choosing to advance positions they reasonably believe are correct at the time, see Roanoke River Basin Ass’n, 991 F.2d at 139 (“While the EAJA redresses governmental abuse, it was never intended to chill the government’s right to litigate or to subject the public fisc to added risk of loss when the government chooses to litigate reasonably substantiated positions, whether or not the position later turns out to be wrong.”).

B. The Secretary’s Arguments and the Totality of the Circumstances

The Secretary advances several reasons why his conduct was substantially justified at both the administrative and litigation phases. He argues that the Board followed what was at the time of its decision the current state of the law, relied upon VA’s longstanding interpretation of § 3.321(b)(1), relied on agency guidance set forth in the Secretary’s Manual M21-1MR, which instructs adjudicators to refer claims for extraschedular consideration if “the schedular evaluations are considered to be inadequate for an individual disability,” and did not have the benefit of the Federal Circuit’s Johnson II decision. Secretary’s Response to EAJA Application at 6-8. The Secretary also argues that, at the litigation phase, the Secretary elucidated these factors and consistently asserted that the regulation had long been applied to disabilities singly, that the regulation was ambiguous, and that the Court should defer to the Secretary’s interpretation of the regulation. Id. at 6-9. Moreover, he argues that the parties submitted seven substantive pleadings before this Court addressing the parties’ differing interpretations, that after en banc review in this Court we upheld his interpretation, and that the Federal Circuit’s deciding of the case on plain language rather than using other tools of regulatory construction is not a bar to establishing substantial justification. Id. at 9-10 (citing cases), 12. He asserts in addition that this case is distinguishable from Patrick, where the Federal Circuit based its decision on this Court’s failure to consider the “totality of the circumstances” and found the Secretary’s position contrary to the plain language of the statute, the legislative history, VA’s own precedent opinion, and the Federal Circuit’s opinions on the issue. Id. at 11 (quoting Patrick, 668 F.3d at 1331). Finally, he argues that this case concerns a regulation that has been held to be confusing and in need of reform, a factor the Secretary considered when choosing to litigate the issue of whether the regulation was ambiguous. Id.

Further, because we must analyze the totality of the circumstances, not just some of the circumstances, we are not limited to those factors specifically identified by *149counsel. See Cline, 26 Vet.App. at 327 (Court must consider the Stillwell factors, “as well as any other circumstances that may apply”); Felton, 7 Vet.App. at 286 (“[W]e do not limit our examination of [substantial justification] to a vacuum surrounded by only the four corners of the pleadings and the underlying opinion by the Court.... [T]he Secretary argued that VA’s position has always been substantially justified, and the Court has merely put that argument to the test.”); see also Patrick, 668 F.3d at 1332 (“We have repeatedly made clear that the substantial justification inquiry requires an analysis of the ‘totality of the circumstances’ surrounding the government’s adoption of a particular position.”); White, 412 F.3d at 1317 (substantial justification analysis “does not exclude any valid issue from consideration”).

II. ANALYSIS

A. Administrative Phase

I agree with the majority that substantial justification must be measured at both the administrative and litigation phases. See Locher v. Brown, 9 Vet.App. 535, 537 (1996). Looking at the totality of the circumstances, including but not limited to the circumstances specifically identified by counsel, there are strong arguments for finding substantial justification at both phases. In adjudicating the issue of ex-traschedular consideration, the Board cited Thun v. Peake, 22 Vet.App. 111 (2008), which states: “The governing norm for extraschedular consideration is a finding that the evidence presents such an exceptional disability picture that the available schedular evaluations for the service-connected disability at issue are inadequate.” Thun v. Peake, 22 Vet.App. 111, 115 (2008) (citing M21-1MR, pt. III, subpt. iv, ch. 6, sec. B.5.c). Moreover, at the time of the May 2010 Board decision in this case, as the majority notes, there was “silence” in the precedential opinions of this Court as to whether § 3.321(b)(1) could be applied to multiple disabilities collectively or only to disabilities individually.3 Accordingly, although the Federal Circuit ultimately invalidated the Secretary’s interpretation of § 3.321(b)(1), that action was taken in a “case of first impression,” see White, 412 F.3d at 1316-17; (Leamon) Johnson, 17 Vet.App. at 442; Felton, 7 Vet.App. at 283. This factor weighs in favor of substantial justification in this case.

In both White and (Leamon) Johnson, the Board in the underlying appeals relied on then-current regulation 38 C.F.R. § 19.9(a)(2), but in the subsequently decided Disabled American Veterans (DAV) v. Secretary of Veterans Affairs, 327 F.3d 1339 (Fed.Cir.2003), a “case of first impression,” the Federal Circuit found that regulation inconsistent with 38 U.S.C. § 7104(a) and invalidated it. See White, 412 F.3d at 1315. Both eases were remanded after DAV was decided, and in the EAJA phases of both White and (Leamon) Johnson, the Federal Circuit and this Court, respectively, held that the Secretary’s actions were substantially justified, under “the totality of the circumstances,” because “DAV v. Sec’y was a case of first impression and there had been no prior adverse reaction to” the now-invalidated regulation. (Leamon) Johnson, 17 Vet.App. at 442; see also White, 412 F.3d at 1316-17. Similarly, here, this Court’s (and *150the Federal Circuit’s) silence on the issue meant that at the time of Mr. Johnson’s May 2010 Board decision, there had been no adverse reaction to the method of adjudicating the issue of extraschedular consideration that the Board employed.

B. Litigation Phase

1. Secretary’s Interpretation of His Own Regulation

At the litigation phase, the Secretary offered cogent reasons for the Board’s application of § 3.321(b)(1) that the Federal Circuit did not accept on the merits but that a reasonable person could have found persuasive. See Underwood, 487 U.S. at 566 n. 2, 108 S.Ct. 2541; Stillwell, 6 Vet.App. at 302 (analyzing “reasons given” for government’s position). Specifically, he argued that the language in the regulation referring to “schedular evaluations” and “disability or disabilities” was ambiguous and could refer to either a combined ex-traschedular evaluation, as Mr. Johnson contended, or “a recognition that a veteran may receive extraschedular ratings for one or more individual disabilities,” as the Government contended. Johnson I, 26 Vet.App. at 243. The reasons given for the Secretary’s position also included that the Secretary was interpreting his own regulation and that that interpretation was entitled to a high degree of deference. Id. (citing, inter alia, Auer v. Robbins, 519 U.S. 452, 461-62, 117 S.Ct. 905, 137 L.Ed.2d 79 (1997)).

The fact that the Agency was interpreting its own regulation, and not a statute, has a direct bearing on how reasonable its position was. Interpreting the language of his own regulation, the Secretary would know that his interpretation would be accepted unless that interpretation were contrary to the regulation’s plain meaning. See Thun v. Shinseki, 572 F.3d 1366, 1369 (Fed.Cir.2009); see also Decker v. N.W. Envntl. Def. Ctr., 568 U.S. 597, 133 S.Ct. 1326, 1337, 185 L.Ed.2d 447 (2013) (“It is well established that an agency’s interpretation need not be the only possible reading of a regulation—or even the best one— to prevail. When an agency interprets its own regulation, the Court, as a general rule, defers to it ‘unless that interpretation is “plainly erroneous or inconsistent with the regulation.” ’ ” (quoting Auer, 519 U.S. at 461, 117 S.Ct. 905) (other internal citation omitted)); Martin v. Occupational Safety and Health Review Comm’n, 499 U.S. 144, 151, 111 S.Ct. 1171, 113 L.Ed.2d 117 (1991). However, if he were interpreting a statute, he would know that, even if his interpretation were not contrary to the statute’s plain meaning, and the Court agreed with him that the statute was ambiguous, he would still have to demonstrate the overall persuasiveness of his position. See Wanless v. Shinseki, 618 F.3d 1333, 1338 (Fed.Cir.2010) (Secretary’s interpretation of statutory language is entitled to deference “only in so far as it has the ‘power to persuade’” (quoting Skidmore v. Swift & Co., 323 U.S. 134, 140, 65 S.Ct. 161, 89 L.Ed. 124 (1944))). Accordingly, the Secretary was on stronger footing proposing an interpretation of his own regulation than he would have been if he were proposing the same interpretation of a statute. See Decker, 133 S.Ct. at 1337; Skidmore, 323 U.S. at 140, 65 S.Ct. 161; Saysana v. Gillen, 614 F.3d 1, 4 (1st Cir.2010) (holding the Government’s choice to litigate “the question of whether [a provision] was ambiguous, and therefore whether courts had to defer to a reasonable administrative interpretation of [that provision]” was itself reasonable).4

*151 2. Consistency with Regulatory Scheme

The reasons given also included that the Secretary, rather than proposing his interpretation for the first time in litigation, had long construed the regulation to mean that a veteran was entitled to extrasche-dular consideration on a disability-by-disability basis and had recorded that construction in his Adjudication Procedures Manual. Johnson I, 26 Vet.App. at 243-44 (citing VA Adjudication Procedures Manual M21-1MR, pt. II, subpt. iv, ch. 6, sec. B.5.c (claims are to be submitted for extraschedular consideration “if the sche-dular evaluations are considered inadequate for an individual disability”)). See Stillwell, 6 Vet.App. at 302 (identifying “consistency with ... VA policy with respect to such position” as factor weighing in favor of substantial justification); cf. Salmi v. Sec’y of HHS, 712 F.Supp. 566, 569-70 (W.D.Mich.1989) (no substantial justification where Secretary “fail[ed] to apply his own long-standing de minimus construction of the severity requirement” (emphasis in original)). The reasons given included the Secretary’s argument that this Court in Thun v. Peake cited with approval the Manual M21-1MR provision stating that claims be submitted for ex-traschedular consideration if the rating schedule is inadequate “for an individual disability” and that the Court noted that the analysis must begin with “a comparison between the level of severity and symptomatology of the claimant’s service-connected disability with the established criteria found in the rating schedule for that disability.” Johnson I, No. 10-1785, Appellee’s Memorandum of Law Responding to Court Order dated May 16, 2012 (Appellee’s June 2012 Memorandum), at 10, 12 (Jun. 15, 2012) (quoting Thun, 22 Vet.App. at 115). The Secretary argued, moreover, that the Federal Circuit, in affirming this Court’s Thun decision, deferred to VA’s interpretation of § 3.321(b)(1) as set forth in the M21-1MR. Id. at 15 (quoting Thun v. Shinseki, 572 F.3d 1366, 1369 (2009)).5

The reasons given further included practical considerations that took into account the provision’s place in the regulatory scheme. In particular, the Secretary argued that the VA rating schedule assigns disability evaluations based on the impact of individual disabilities, not collective disabilities; and, thus, “it is unworkable for VA to compare a Veteran’s disabilities collectively to the criteria of an individual disability in the rating schedule to determine whether the rating schedule is adequate and extraschedular consideration is warranted.” Appellee’s June 2012 Memorandum at 10. Additionally, § 3.321(b)(3) *152states that effective dates for extraschedu-lar evaluations will be determined in accordance with 38 C.F.R. § 3.400, and § 3.400 describes how to provide an effective date for an individual claim only, not collective claims. Id. at 6.

Moreover, the Secretary represented, and this Court has no cause to doubt, that “a sampling of Board decisions demonstrates that the Board generally follows VA’s long-standing interpretation” of the regulation to require individualized extras-chedular consideration. Id. at 8; see Fed. Express Corp. v. Holowecki, 552 U.S. 389, 399-400, 128 S.Ct. 1147, 170 L.Ed.2d 10 (2008) (judicial deference was due to the Equal Employment Opportunity Commission’s (EEOC’s) “interpretive statement, embodied in the compliance manual and memoranda, [which] has been binding on EEOC staff for at least five years” even though, “as the Government concedes, the agency’s implementation of this policy has been uneven”).6

3. This Court Upholds the Secretary’s Approach

Five out of eight Judges on this Court deemed the Secretary’s interpretation reasonable. See Johnson I, 26 Vet.App. at 244 (noting, inter alia, that the Secretary was interpreting his own regulation and that “the VA disability rating schedule is designed to consider the disabling effect of disabilities separately”). Although the majority points out that this Court’s agreement with the Secretary’s interpretation does not “resolve the substantial justification inquiry,” ante at 141 (citing Patrick, 668 F.3d at 1332), that fact is both obvious and a red herring. Of course this Court’s merits determination does not “resolve” the question of whether the Government was substantially justified; however, neither does the Federal Circuit’s merits determination. Neither court adjudicated the issue of whether the Government’s position was substantially justified, and neither court was asked. See Felton, 7 Vet.App. at 286 (“Indeed, our underlying decision on the merits addressed an altogether different question: whether the regulation at issue was valid rather than whether the steps that led to VA’s promulgation of the regulation (i.e., the prelitigation conduct) was substantially justified.”); see also Toro Co. v. White Consol. Indus., Inc., 383 F.3d 1326, 1335 (Fed.Cir.2004) (only “issues that were actually decided, ... in the earlier litigation” represent the law of the case); Grantham v. Brown, 114 F.3d 1156, 1158 (Fed.Cir.1997) (“It is axiomatic that the language in [any case] must be read in light of the facts and issues that were before the court when the language was written.”); Young v. Shinseki, 25 Vet.App. 201, 204 (2012) (en banc) (issue preclusion requires that (1) the issues previously adjudicated were “identical to” those before the current tribunal; (2) the issues were “actually litigated” in the prior proceeding; (3) the prior tribunal’s resolution of those issues was “necessary to the resulting judgment”; and (4) the litigant was fully represented in the prior proceeding).

Nor would it resolve the reasonableness inquiry if the Supreme Court unanimously reversed the Federal Circuit’s merits determination tomorrow. The Supreme *153Court binds the Federal Circuit as to merits determinations actually reached, just as the Federal Circuit binds this Court, see Toro, 383 F.3d at 1335; however, for EAJA substantial justification purposes, that hierarchy of authority does not mean, ipso facto, that disagreements with merits determinations are unreasonable.

However, the fact that this Court agreed with the Secretary’s interpretation of § 3.321(b)(1) is evidence in favor of a finding that the case presented “a dispute over which reasonable minds could differ.” Norris, 695 F.3d at 1265; League of Women Voters of Calif. v. F.C.C., 798 F.2d 1255, 1260 (9th Cir.1986) (denying EAJA fees because, inter alia, “[t]he constitutionality of the amended statute is a question upon which reasonable minds could differ, as evidenced by the five-four division by the Supreme Court”); Gonzales, 408 F.3d at 621 (noting differing views of Government’s position taken by judges at previous phases and concluding, “[t]hus, ‘reasonable minds’ could and did differ about” the government’s position). Moreover, the en banc approval of the Secretary’s position by this Court, unique amongst courts as the only court established to hear exclusively appeals from the Board of Veterans’ Appeals, is arguably a stronger factor in the Secretary’s favor than judicial approval generally. See 38 U.S.C. § 7252(a) (giving this Court exclusive jurisdiction over appeals from the Board); Shinseki v. Sanders, 556 U.S. 396, 129 S.Ct. 1696, 1707, 173 L.Ed.2d 532 (2009) (noting this Court’s exclusive jurisdiction over veterans benefits cases and its special expertise in veterans benefits law); Toiler v. Derwinski, 2 Vet.App. 8, 11-12 (1991) (“Congress has made the United States Court of Veterans Appeals the national ‘statutory court of review’ of decisions on veterans’ benefits by the Secretary and the Department of Veterans Affairs.” (citation omitted)); cf. Underwood, 487 U.S. at 569, 108 S.Ct. 2541 (exploring aspects of court approval or disapproval that may bear on weight to be given to either, including that “a string a losses can be indicative; and even more so a string of successes”).

C. The Majority’s Analysis

I. Scant Analysis Beyond Repeating the Merits Determination

As for factors weighing against substantial justification, the majority states that it applies a totality of the circumstances test but effectively substitutes the Federal Circuit’s merits determination, which was that the Secretary’s interpretation was contrary to the regulation’s language and public policy, for its own analytic responsibility.7 Johnson II, 762 F.3d at 1366. While mindful that the exercise of determining substantial justification has been described as “a judgment call” that is “quintessentially discretionary in nature,” Chiu v. United States, 948 F.2d 711, 715 and n. 4 (Fed.Cir.1991), the majority makes the one error that will doom to failure any substantial justification analysis: to focus on one factor, such as the merits determination, without considering all the pertinent *154factors surrounding the Government’s conduct. See Patrick, 668 F.3d at 1382 (“Here, although the Veterans Court acknowledged this ‘totality of the circumstances’ standard, it improperly focused on only one factor—the fact that the court itself had previously upheld the VA’s erroneous interpretation....”); Essex, 757 F.2d at 253 (rejecting prevailing party’s argument that the Government is per se not substantially justified where it acts in contravention of a regulation because “[t]he reasonableness of the government’s litigation position is determined by the totality of the circumstances, and we eschew any single-factor approach”); Felton, 7 Vet.App. at 286 (“[W]e have gone to great lengths to make clear that the resolution of EAJA issues depends on many factors.”); Stillwell, 6 Vet.App. at 302 (identifying the “merits” as one factor out of many for consideration).

Although the majority devotes a few brief paragraphs to the Secretary’s arguments on substantial justification during the administrative phase, the running theme even in that brief discussion is that the Secretary’s interpretation of § 3.321(b) was eventually determined to be erroneous. Ante at 141 (“The Secretary contends that his position at the administrative stage was substantially justified in part because he has consistently and long applied his erroneous interpretation of § 3.321(b)(1).” (emphasis added)); ante at 141 (“Moreover, the Secretary’s argument of a consistent and long-standing interpretation finds little support in the majority decision in Johnson I, which was predicated on the majority’s [ ] view that the regulation was ambiguous—a view subsequently rejected by the Federal Circuit.... ” (emphasis added)); ante at 142 (“Here, the Board never addressed the scope of § 3.321(b) and instead applied the regulation in a manner wholly contrary to the plain wording of the regulation.” (emphasis added)).

Not every misinterpretation of a regulation by an agency “stamps its position as unreasonable,” Meinhold, 123 F.3d at 1278, and “the fact that one court agreed or disagreed with the Government does not establish whether or not its position is substantially justified,” Underwood, 487 U.S. at 565, 108 S.Ct. 2541. It follows that meaningful analysis of the reasonableness of the Government’s position is required here. Unsurprisingly, courts are unanimous on this point. See, e.g., Gonzales, 408 F.3d at 620; Taucher, 396 F.3d at 1174. The majority’s dismissive and superficial analysis fails to provide the meaningful review of the totality of the circumstances that substantial justification requires.

2. Using the Federal Circuit’s Plain Language Analysis as a Trump Card as to Substantial Justification

The majority’s view is, apparently, that a more detailed analysis is not necessary because the Federal Circuit held at the merits stage that the Secretary’s reading of the regulation was “contrary to the plain language of that regulation and without policy support,” and this poses a “significant hurdle” as to substantial justification. Ante at 143 (citing Patrick, 668 F.3d at 1331). But it does not follow that, because the Federal Circuit used a “plain language” analysis to arrive at a particular result, the analysis used to arrive at that result is plain, or straightforward, or the only reasonable result that the court could have reached, or that parties who failed to foresee that result were unreasonably wrong. Plain language, otherwise stated as language unadorned by other methods of determining meaning, is a canon of statutory and regulatory construction and not a trump card in the substantial justification analysis. See Wash. Legal Found. v. *155U.S. Sentencing Comm’n, 17 F.3d 1446, 1449-50 (D.C.Cir.1994) (“‘Plain meaning’ analysis is merely a tool of construction; it does not perform a talismanic function in statutory interpretation.”); see also Watt v. Alaska, 451 U.S. 259, 266, 101 S.Ct. 1673, 68 L.Ed.2d 80 (1981) (“[T]he plain-meaning rule is ‘rather an axiom of experience than a rule of law, and does not preclude consideration of persuasive evidence if it exists.’ ” (quoting Boston Sand Co. v. United States, 278 U.S. 41, 48, 49 S.Ct. 52, 73 L.Ed. 170 (1928))). Reasonable minds can and do differ on whether the meaning of a regulation can be divined from its plain meaning, just as they can differ on any other legal question.

This Court and others have held numerous times and without difficulty that, even though the government’s interpretation was rejected by a court using the plain language canon, that position was substantially justified because the plain language question was close, or because the law in question was complex, or because the case presented an issue of first impression, or, as in this case, because a combination of those factors was present. See White, 412 F.3d at 1316 (Secretary’s position was substantially justified even where Board applied a regulation found to be “facially invalid on the basis of contradicting the overarching legislation” because the case was “a case of first impression” where there had been “no prior adverse reaction” to the regulation); Bates, 20 Vet.App. at 192 (although Federal Circuit held that Secretary’s interpretation of 38 U.S.C. § 511(a) was contrary to statute’s “plain language” and had “no basis in the statutory language, legislative history, or case authority,”8 Secretary’s position was substantially justified because the issue was “a matter of first impression” and the Federal Circuit’s analysis “could not have been easily divined from existing caselaw”); Ozer, 16 Vet.App. at 478 (Secretary’s enactment of regulation that “contravened the plain language of the statute and was thus invalid” was substantially justified where the Secretary’s conduct represented “a good faith effort to interpret an evolving area of the law,” did not conflict with then-existing precedent, and “was not questioned” until proceedings before the Court in Felton, when it was invalidated (quoting Felton, 7 Vet.App. at 285)); see also Saysana, 614 F.3d at 6 (Government was wrong to contend that 8 U.S.C. § 1226(c) “was ambiguous,” such that deference under Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984), was appropriate; however, Government’s position was “substantially justified” because case presented a novel issue on which there was little precedent); Pottgieser, 906 F.2d at 1324 (Government’s reading of Social Security statute contrary to its plain language was substantially justified because “the Secretary’s litigation position had a reasonable basis and the Social Security statutes [are] complex”); Yoffe, 775 F.2d at 450 (although Government misread Treasury regulations to allow for the private, rather than public, sale of abandoned merchandise, reviewing court did “not think that the government should be found to have acted without substantial justification because it did not foresee how the court of appeals would interpret the regulations”); S. Or. Citizens Against Toxic Sprays, Inc. v. Watt, 556 F.Supp. 155, 157 (D.Ore.1983) (government was substantially justified where contravened regulation was “a convoluted regulation, difficult to interpret” and there was “little case law to guide the Government”), aff'd sub nom. Southern Oregon Citizens Against Toxic Sprays v. Clark, 720 F.2d *1561475, 1481 (9th Cir.1983) (affirming denial of EAJA fees because, although the appeals court found that the regulation was not “difficult to interpret” but was rather “straightforward and means what it says, ... [t]he analysis required by the regulation is exceptional ... [and] the only reported case interpreting the ... regulation before this decision supported the government’s position”).

Similarly, here, the regulation at issue is complex; the question of whether a plain language analysis would suffice to interpret the regulation was a close one, especially given that 5 out of 11 Judges to explore the issue determined that the regulation was not clear;9 no court had ever interpreted the regulation in a manner contrary to the way the Secretary interpreted it; and the Secretary’s interpretation presented a reasonable attempt to resolve a difficult question. See, e.g., White, 412 F.3d at 1316. That the Federal Circuit has now provided an authoritative interpretation of that regulation does not change the fact that, prior to Johnson II, the Government could be understood if it found the regulation both “complex” and “exceptional.” See Stillwell, 6 Vet.App. at 303; see also Meinhold, 123 F.3d at 1278; Brambley, 17 Vet.App. at 24-25 (noting the “ambiguity of the disability ratings regulations” relating to extraschedular and TDIU considerations and the “need to streamline and clarify” these regulations).

The Federal Circuit’s citation to an additional reason—public policy—for disagreeing with the Secretary’s interpretation does not rob this Court of the ability to hold that interpretation reasonable when analyzing the Government’s position for EAJA purposes, for all of the reasons listed above, nor does it absolve this Court of the responsibility to make an independent determination of reasonableness. Compare Bates, 398 F.3d at 1361 (Federal Circuit holds that the Secretary’s interpretation had “no basis in the statutory language, legislative history, or case authority”), with Bates, 20 Vet.App. at 192 (Secretary’s interpretation was substantially justified); see also Gonzales, 408 F.3d at 620 (merely stating that the merits court “soundly rejected the government’s arguments” does not establish a lack of substantial justification). The majority has simply failed to do that here.

3. The Majority’s Patrick Dicta

The holding of Patrick, as noted above, was that this Court had applied an erroneous legal test, because it “improperly focused on only one factor.” 668 F.3d at 1332. The observation cited by the majority, regarding plain language and legislative history of statutes, was not necessary to the determination that the case must be remanded for a totality-of-the-circumstances analysis and thus is not binding on this Court. See McDaniel v. Sanchez, 452 U.S. 130, 141, 101 S.Ct. 2224, 68 L.Ed.2d 724 (1981) (“[DJictum unnecessary to the decision in [a] case ... [is] not controlling in this case[.]”).

*157First, the Federal Circuit did not hold that the government’s interpretation lacked substantial justification because it interpreted a statute contrary to its plain meaning and legislative history. Indeed, the Federal Circuit has determined that it cannot make such a holding. See Bowey, 218 F.3d at 1378 (“[DJetermining substantial justification requires the application of law to facts. Since such inquiries are specifically excluded from our jurisdictional grant, see 38 U.S.C. § 7292(d)(2), we must remand this case to allow the Court of Appeals for Veterans Claims to decide whether the government’s position was, in fact, substantially justified at the time it was adopted, based on the record at the time.”).

Second, the majority misreads the Patrick dicta. The language cited by the majority reads, “Where ... the government interprets a statute in a manner that is contrary to its plain language and unsupported by its legislative history, it will prove difficult to establish substantial justification.” 668 F.3d at 1330-31 (emphasis added). In other words, where the situation is similar to that in Bates, involving an agency’s misconstruing Congress’s words and misreading congressional history, it will prove “difficult”—though, clearly, not impossible, see Bates, 20 Vet.App. at 192— to establish that the agency’s misconstruction was reasonable. See Patrick, 668 F.3d at 1333 (“[I]t would be hard to imagine how it could be held that one had been substantially justified in defying the will of Congress by interpreting a statute in a manner contrary to its plain language and legislative history.” (citations and internal quotation marks omitted)). The Federal Circuit said nothing in Patrick, even in passing, about interpretations that contradict statutes but do not contradict legislative history. And here, there has been no holding that the Government’s interpretation of § 3.321(b)(1) was contrary to legislative history because there is no legislative history as to VA regulation § 3.321.10 And, unlike the Secretary in Patrick, the Secretary here was not interpreting a statute; he was interpreting his own regulation. Although the canons of regulatory construction are the same as those of statutory construction, as noted above, an agency is on much more stable ground in asserting that its construction of its own words is “reasonable” than in stating that its construction of Congress’s words is reasonable. See Decker, 133 S.Ct. at 1337; Skidmore, 323 U.S. at 140, 65 S.Ct. 161. A reasonable Secretary might well make a different calculation when faced with his own regulatory language that he believes to be ambiguous than when faced with statutory language that he believes possesses the same ambiguity. See Felton, 7 Vet.App. at 284; see also Saysana, 614 F.3d at 4. Accordingly, whatever one thinks of the accuracy of the Federal Circuit’s prediction in Patrick, which is that, when two specific factors are against the Secretary’s interpretation, that interpretation will have a significant hurdle to clear before being deemed substantially justified, ante at 142-43, neither of those factors is present in this case.

III. CONCLUSION

For the reasons stated above, I believe that the balance of factors in the totality of *158the circumstances analysis weighs in the Secretary’s favor on the question of whether the Secretary’s position was unreasonable in fact and law and that he should thus pay the appellant’s legal fees. The Secretary’s reasonable arguments on a novel question about his regulation, which included practical considerations and his longstanding interpretation, and this Court’s en banc decision in his favor citing those same arguments, made both the Government’s administrative and litigation positions in this case “justified to a degree that could satisfy a reasonable person.” Underwood, 487 U.S. at 565, 108 S.Ct. 2541. I dissent from the majority’s attempt to short-circuit the totality-of-the-circumstances analysis by citation to dicta from the Federal Circuit about what factors might make a substantial justification analysis more or less difficult. Additionally, I dissent from the majority’s reliance on the Federal Circuit’s merits determination, even though that merits determination did not purport to address substantial justification or the reasonableness of the Government’s positions at the time those positions were taken.

. Although the majority notes Judge Stein-berg’s concurrence in Brambley v. Principi, 17 Vet.App. 20 (2003), his view that extraschedu-lar consideration should be conducted on a collective basis was not shared by the majority in Brambley, see Maryland v. Wilson, 519 U.S. 408, 412-13, 117 S.Ct. 882, 137 L.Ed.2d 41 (1997) (noting that statements in concurring opinions do not constitute binding precedent), and the Court never adopted that view in the subsequent 11 years of jurisprudence leading up to Johnson II.

. The Federal Circuit noted in passing that Mr. Johnson’s interpretation of § 3.321(b)(1) was "consistent with” 38 U.S.C. § 1155. However, the Federal Circuit expressed no *151opinion about whether the Secretary’s contrary reading of § 3.321(b)(1) was also reasonable in light of the statute. See Cody, 631 F.3d at 144 ("[B]oth the [EAJA] order and the record reflect that the district court considered the arguments of both parties to be reasonable. We agree. In this case of first impression, the Government made reasonable arguments based on statutory interpretation and analogous cases.”).

. The majority ignores language from the M21-1 MR stating that extraschedular evaluations will be performed for "an individual disability” but cites a different section of the Manual and holds against the Secretary that this different section does not discuss extras-chedular evaluation of "a disability” but instead discusses extraschedular evaluation “due to disability.” Ante at 142. The majority concedes that the Secretary’s manual would provide evidence that extraschedular consideration was intended to be limited to an individual disability if the indefinite article had been used prior to the word "disability”— which is precisely the language the Secretary did use and precisely the language in place at the time of the Board decision in this case. See Thun, 572 F.3d at 1369 (quoting "an individual disability” language from manual as of 2009).

. Although the majority contends that the Secretary’s acknowledgment that some Board decisions might have mistakenly made collective extraschedular referral decisions over the years shows that he was "unable to confirm a consistent VA practice regarding his interpretation of § 3.321(b)(1),” ante at 141, the majority fails to explain how lack of perfection in applying the Secretary’s interpretation in hundreds of Board decisions over the years demonstrates that the Secretary’s interpretation was not longstanding. See Fed. Express Corp., 552 U.S. at 400, 128 S.Ct. 1147 (noting that "[s]ome degree of inconsistent treatment is unavoidable when the agency processes over 175,000 inquiries a year”).

. As the Federal Circuit appeared to recognize, its policy observation was dicta. See Myore v. Nicholson, 489 F.3d 1207, 1211 (Fed.Cir.2007) ("If the statutory language is clear and unambiguous, the inquiry ends with the plain meaning.”); Cacciola v. Gibson, 27 Vet.App. 45, 52 (2014) ("If the meaning of the regulation is clear from its language, that is the end of the matter.” (citation and internal quotation marks omitted)); see also Johnson II, 762 F.3d at 1366 ("We further note that, while policy arguments would not, in any case, persuade us to depart from the plain language of the regulation....” (emphasis added)). And although we are bound by the Federal Circuit’s plain language interpretation of the regulation, we are not bound by its dicta as to public policy. See Toro, 383 F.3d at 1335,

. Bates v. Nicholson, 398 F.3d 1355, 1361 (Fed.Cir.2005).

. I was one of three Veterans Court Judges who found the regulation’s language clear, joining Judge Davis in dissenting in Johnson I. I find nothing inconsistent or remarkable about viewing the regulatory language as clear and unambiguous, as I expressed in Johnson I, but viewing the Secretary's adopting and arguing a position that the language was ambiguous, and his interpretation thereof due deference, as reasonable, all things considered, and an issue worthy of the lengthy briefing and opinion-drafting process that characterized the Johnson I decision. See Saysana, 614 F.3d at 4 (although Government was wrong about whether statute was ambiguous, "the question of whether § 1226(c) was ambiguous, and therefore whether courts had to defer to a reasonable administrative interpretation of the statute, was one of first impression,” and, "as such, it was appropriate for the government to seek specific instruction from the court on this issue”).

. The Federal Circuit in Johnson II did not address regulatory history, and I note that the issue of regulatory history is a murky one. See Johnson I, 26 Vet.App. at 257 (Kasold, J., dissenting) (noting that the language of the original 1930 VA administrative rule "is unclear—ambiguous—as to whether an extras-chedular disability rating could be predicated on a multiple-disability basis, because the rule simply indicates that such a rating would be based on the reduction in average earning capacity compared to the average worker ‘suffering a similar disability' ").