OPINION AND ORDER

WOLSKI, Judge.

On June 1, 2004, the plaintiffs Miguel A. Contreras, Jorge Balderrama and Anita Trujillo (“plaintiffs”), on behalf of themselves and a proposed class of Hispanic Customs Service Criminal Investigator Special Agents, filed the instant action seeking money damages from the defendant United States (“government”). On June 22, 2004, the plaintiffs filed an amended complaint (“Am.Compl.”) and an amended motion for class certification.

This case is currently before the Court on the government’s motion to dismiss the amended complaint under Rules 12(b)(1) and 12(b)(6) of the Rules of the United States Court of Federal Claims (“RCFC”), and plaintiffs’ cross-motion for partial summary judgment under RCFC 56(c). For the reasons that follow, the government’s motion to dismiss for lack of subject matter jurisdiction is GRANTED.

I. BACKGROUND

A. The Present Action

The plaintiffs are members of a proposed class of current and former series 1811 criminal investigator special agents formerly within the U.S. Customs Service (“Customs”), now under the Department of Homeland Security.1 Am. Compl. ¶¶7, 16. They seek money damages based upon Customs’s alleged violation of 5 U.S.C. § 4523 (2000) (“Section 4523”), part of the Federal Law Enforcement Pay Reform Act (“FLEPRA”). Am. Compl. ¶ 1.

Passed in 1990, Section 4523 provides:

(a) An agency may pay a cash award, up to 5 percent of basic pay, to any law enforcement officer employed in or under such agency who possesses and makes substantial use of 1 or more foreign languages in the performance of official duties.

*585(b) Awards under this section shall be paid under regulations prescribed by the head of the agency involved (or designee thereof). Regulations prescribed by an agency head (or designee) under this subsection shall include:

(1) procedures under which foreign language proficiency shall be ascertained;

(2) criteria for the selection of individuals for recognition under this section; and

(3) any other provisions which may be necessary to carry out the purposes of this subchapter.

5 U.S.C. § 4523. Customs did not provide a foreign language pay award (“FLPA”) to any of its law enforcement officers until 1998.2 Am. Compl. ¶ 2. In 1994, Congress passed the Customs Officers Pay Act, 19 U.S.C. § 267a (2000) (“section 267a”), which provides:

Cash awards for foreign language proficiency may, under regulations prescribed by the Secretary of the Treasury, be paid to customs officers (as referred to in section 267(e)(1) of this title) to the same extent and in the same manner as would be allowable under subchapter III of chapter 45 of title 5 with respect to law enforcement officers (as defined by section 4521 of such title).

Customs began issuing FLPAs to customs officers in 1996, following a threatened boycott of the use of the Spanish language by Miami customs officers. Am. Compl. ¶¶ 23-25. Customs began issuing FLPAs to law enforcement officers in 1998. The standards to qualify for an FLPA as a customs officer were, however, less burdensome than those to qualify for an FLPA as a law enforcement officer. See id. ¶ 3. For FLPA eligibility, a customs officer need only have obtained an annual certification from a supervisor that the officer used a foreign language more than ten percent of the time in his non-overtime regularly scheduled duty. Id. ¶¶ 3, 30. In contrast, to qualify for an FLPA as a law enforcement officer required meticulous documentation of case numbers, informant numbers and duties, as well as a narrative report. Also, no credit was given for partial hours or for hours outside the standard work week.3 Id. ¶¶3, 31. Therefore, the plaintiffs, who frequently performed official duties using the Spanish language outside of the standard work week, had a much more difficult time accruing the necessary number of hours of foreign language use mandated by FLEPRA’s “substantial use” requirement. See id. ¶¶3-4. In 2002, Customs relaxed the FLPA qualifications for law enforcement officers to require less documentation. Id. ¶34.

The plaintiffs divide their complaint into three counts. In Count One, the plaintiffs seek FLPAs from the period 1996 through 1998, during which time Customs provided FLPAs to customs officers but not to law enforcement officers. Id. ¶¶ 48-52. In Count Two, the plaintiffs contest the standards by which Customs determined FLPA eligibility for law enforcement officers. They argue that the standards should have been identical to those used for customs officer FLPAs. They seek additional FLPA money computed using the less stringent customs officer standards. Id. ¶¶ 53-56. In Count Three, the plaintiffs argue that the law enforcement FLPA standards violated section 4523 because they did not credit, toward the FLPA “substantial use” requirement, time spent using the foreign language outside of the standard work week or for partial hours. They seek additional FLPA money based on standards that would credit language use outside of the standard work week and partial hours. Id. ¶¶ 57-61.

B. Procedural History

In January 1995, plaintiff Contreras filed a class action complaint with the U.S. Equal Employment Opportunity Commission (“EEOC”), alleging that Customs discriminated against Hispanic law enforcement officers. See Contreras v. Ridge, 305 F.Supp.2d 126, 128-29 (D.D.C.2004). See also Contreras v. Rubin, 1998 WL 253533, *1 (E.E.O.C. *586May 11, 1998). An EEOC administrative law judge (“ALJ”) recommended class certification. Ridge, 305 F.Supp.2d. at 129. The Treasury Department rejected the ALJ’s recommendation. Id. In May 1998, the EEOC vacated the Treasury’s rejection and certified the class. Id. at 130. In June 2000, the ALJ issued an order defining the class, and the parties began discovery. Id. Contreras subsequently withdrew from the EEOC proceeding, and filed a complaint in the U.S. District Court for the District of Columbia on May 10, 2002, seeking relief based on Title VII theories. Am. Compl. ¶13.

The district court granted summary judgment for the government on most of Contreras’s claims, including the claim for FLPAs that were denied for allegedly discriminatory reasons. Ridge, 305 F.Supp.2d at 134. The court held that Contreras had failed to exhaust the administrative remedies for the claims. Id. Some of Contreras’s Title VII claims, however, survived summary judgment. Id. at 133-34. Although the district court dismissed the FLPA claim on summary judgment, because the dismissal was not on the merits, res judicata does not bar Contreras here.4

II. DISCUSSION

The government moves to dismiss under both RCFC 12(b)(1) and 12(b)(6). It contends that the Civil Service Reform Act (“CSRA”)5 applies to plaintiffs’ claims, depriving this Court of jurisdiction. The government also argues that section 4523 is not money-mandating and thus this case is either beyond this Court’s jurisdiction or fails to state a claim for relief. The plaintiffs cross-move for partial summary judgment on Count Three of their Amended Complaint, arguing that the use of a foreign language for partial hours and during hours worked outside the standard work week must be included, as a matter of law, in determining awards under section 4523.

A. Legal Standard

The granting of a motion to dismiss for failure to state a claim under RCFC 12(b)(6) “is appropriate when the facts asserted by the claimant do not entitle him to a legal remedy.” Lindsay v. United States, 295 F.3d 1252, 1257 (Fed.Cir.2002). When considering a motion to dismiss under RCFC 12(b)(6), “the allegations of the complaint should be construed favorably to the pleader.” Scheuer v. Rhodes, 416 U.S. 232, 236, 94 S.Ct. 1683, 40 L.Ed.2d 90 (1974). A complaint should not be dismissed “unless it appears beyond doubt that the plaintiff can prove no set of facts in support of his claim which would entitle him to relief.” Conley v. Gibson, 355 U.S. 41, 45-46, 78 S.Ct. 99, 2 L.Ed.2d 80 (1957).

In contrast, on a 12(b)(1) motion to dismiss for want of jurisdiction, facts bearing on jurisdiction may be disputed, necessitating the consideration of items outside of the pleadings. Englewood Terrace L.P. v. United States, 61 Fed.Cl. 583, 584 (2004). See Reynolds v. Army & Air Force Exchange Service, 846 F.2d 746, 747-748 (Fed.Cir.1988). While the defendant has the burden to show that the plaintiff has failed to state a claim upon which relief can be granted, the plaintiff has the burden of establishing subject matter jurisdiction over his action. See Thomson v. Gaskill, 315 U.S. 442, 446, 62 S.Ct. 673, 86 L.Ed. 951 (1942).

Summary judgment is appropriate only “if the pleadings, depositions, answers to interrogatories, and admissions on file, together with affidavits, if any, show that there is no genuine issue as to any material fact and that the moving party is entitled to a judgment as a matter of law.” RCFC 56(c); see Celotex Corp. v. Catrett, 477 U.S. 317, 322-23, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986); Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 247-48, 106 S.Ct. 2505, 91 L.Ed.2d 202 (1986); Sweats Fashions, Inc. v. Pannill Knitting Co., 833 F.2d 1560, 1562-63 (Fed.Cir.1987).

*587 B. The Civil Service Reform Act

Where a matter falls within the competence of the Merit Systems Protection Board (“MSPB”) (a CSRA creation), this Court is without jurisdiction. In Lindahl v. Office of Personnel Management, 470 U.S. 768, 105 S.Ct. 1620, 84 L.Ed.2d 674 (1985), the Supreme Court noted that the Federal Circuit has exclusive jurisdiction over appeals from the MSPB, but did not speak directly to the issue of the Claims Court’s jurisdiction. See id. at 775, 105 S.Ct. 1620. The same year that Lindahl was decided, the Federal Circuit held in McClary v. United States, 775 F.2d 280 (Fed.Cir.1985), that where a federal employee has an appeal to the MSPB, his cause of action is not cognizable in the Claims Court. Id. at 282. In United States v. Fausto, 484 U.S. 439, 108 S.Ct. 668, 98 L.Ed.2d 830 (1988), appealed from the Federal Circuit, the Supreme Court held that after the CSRA, the Claims Court did not have jurisdiction to determine whether a personnel action was unwarranted, because that task is within the MSPB’s authority. Id. at 454-55,108 S.Ct. 668. The Federal Circuit held in Worthington v. United States, 168 F.3d 24 (Fed.Cir.1999), that the jurisdictional test for the Court of Federal Claims in CSRA-related cases begins with an analysis of the MSPB’s jurisdiction; the Circuit cautioned that not every adverse personnel action falls within the Board’s jurisdiction. Id. at 26-27. In Read v. United States, 254 F.3d 1064 (Fed.Cir.2001), the Federal Circuit interpreted Fausto as holding that the Court of Federal Claims lacks jurisdiction over a claim within the MSPB’s jurisdiction, even if the plaintiff is barred from asserting his particular claim before the Board. Id. at 1067.

The Federal Circuit has held that premium pay — as distinguished from “basic pay” — falls outside of the MSPB’s jurisdiction, such that actions for premium back pay may be brought under the Tucker Act in the Court of Federal Claims. See Nigg v. MSPB, 321 F.3d 1381, 1385 (Fed.Cir.2003); Martinez v. MSPB, 126 F.3d 1480, 1482 (Fed.Cir.1997).

The CSRA jurisdictional analysis applied to this case involves two questions. First, are the plaintiffs covered by the CSRA? Second, is an action for FLPAs within the MSPB’s jurisdiction? If the answer to both of these question is yes, then the Court lacks jurisdiction over the plaintiffs’ claims. As to question one, the plaintiffs concede that they are employees covered by the CSRA. As to question two, the defendant argues that the plaintiffs have framed their complaint as an action seeking damages based on an illegal reduction in pay; assuming that to be true, the defendant concludes that the action is within the MSPB’s jurisdiction. The plaintiffs counter that they do not frame their complaint in those terms. They argue that FLPAs are premium pay, not basic pay, and therefore not subject to the CSRA.

Title 5, section 7512 lists the personnel actions covered by the CSRA. Subsection (4) includes within CSRA’s scope “a reduction in pay.” 5 U.S.C. § 7512(4). Hence, if the failure to provide an FLPA is a reduction in pay, then jurisdiction is wanting. Section 4522 of FLEPRA states: “An award under this subchapter [5 U.S.C. §§ 4521 — 4523] is in addition to the basic pay of the recipient.” By statute, an FLPA is something beyond pay, a premium. To be deprived of an FLPA is not to have one’s pay reduced; it is to lose a bonus. Therefore, FLPAs fall within the premium pay precedents of the Federal Circuit. And because the personnel action plaintiffs are challenging — FLPA denials — is not within the MSPB’s jurisdiction, the CSRA does not deprive this Court of jurisdiction.

C. Money-Mandating Statute

Our Court’s jurisdiction is derived principally under the Tucker Act, which provides in relevant part:

The United States Court of Federal Claims shall have jurisdiction to render judgment upon any claim against the United States founded either upon the Constitution, or any Act of Congress or any regulation of an executive department, or upon any express or implied contract with the United States, or for liquidated or unliquidated damages in cases not sounding in tort.

*58828 U.S.C. § 1491(a)(1) (2000). The Tucker Act makes the government susceptible to suit in this Court, provided that the matter comes within the description of claims under our jurisdiction. In this regard, a claim “founded ... upon ... any Act of Congress or any regulation of an executive department” must be based on a law or regulation that either entitles the plaintiff to a payment of money from the government, or places a duty upon the government, the breach of which gives the plaintiff a money damages remedy. See United States v. Mitchell, 463 U.S. 206, 216-18, 103 S.Ct. 2961, 77 L.Ed.2d 580 (1983) (“Mitchell II”). Such a statute or regulation is commonly called “money-mandating.”

The government argues that this case is not within this Court’s jurisdiction, because FLEPRA vests the decision whether to grant an FLPA entirely within the agency’s discretion. It contends that the plain meaning of section 4523 proves that the statute is not money-mandating because the provision begins, “[a]n agency may pay a cash award” (emphasis added). Because FLPAs are neither automatic nor competitively awarded, the defendant concludes that section 4523 is not money-mandating.

The plaintiffs respond with several arguments. First, they argue that section 4523 standing alone is money-mandating, even though couched in discretionary terms, because it establishes discrete conditions precedent to an employee’s entitlement to an FLPA, and a sum certain for the award. Pis.’ Reply at 1-2, 4-6. Second, they argue that once Customs exercised its section 267a power to provide FLPAs to customs officers, it was required by the statute also to authorize FLPAs for law enforcement officers. See Tr. (Jan. 6, 2005) at 75; Pis.’ Corr. Opp. at 11-12. And finally, they claim that once Customs established an FLPA program under section 4523, its discretion to refuse to award FLPAs to otherwise qualified law enforcement officers ceased. See Pis.’ Corr. Opp. at 8-9; Pis.’ Reply at 4-5.

1. What is the test to determine if a statute is “money-mandating”?

Each side argues that it would prevail on the government’s motion to dismiss regardless of the test used to determine whether the statute is money-mandating, see Def.’s Corr. Reply at 2 (“the outcome of plaintiffs’ case will be the same”); id. at 3-4; Tr. (Jan. 6, 2005) at 49 (plaintiffs’ counsel stating “I think clearly any measure of statutory interpretation means they’re going to be moneymandat[ing] statutes”). But plaintiffs rely heavily on the since-vacated decision of the Federal Circuit in Fisher v. United States, 364 F.3d 1372 (Fed.Cir.2004) (“Fisher I”), vacated, Fisher v. United States, 2005 WL 548257, 403 F.3d 1307 (Fed.Cir. Mar.9, 2005), and the since-overruled Gollehon Farming v. United States, 207 F.3d 1373 (Fed.Cir.2000), overruled, Fisher v. United States, 402 F.3d 1167, 1172 (Fed.Cir.2005) (“Fisher II”). See, e.g., Pls.’ Corr. Opp. at 6, 11, 13; Pis.’ Reply at 3^4. Since Fisher I was based in part on the Supreme Court’s decision in United States v. White Mountain Apache Tribe, 537 U.S. 465, 472-73, 123 S.Ct. 1126, 155 L.Ed.2d 40 (2003), the interpretation of which was left unsettled by Fisher II, see 402 F.3d at 1174-75, the Court believes it is important to reach a conclusion as to the proper test, regardless of whether the outcome of the instant motion would be affected.

To the extent that plaintiffs have argued that all they need to do is make “a non-frivolous allegation of a money-mandating statute” in order for this Court to have jurisdiction, this approach was clearly rejected by the Federal Circuit in Fisher II. Id. at 1172. The en banc portion of Fisher II makes plain that the “non-frivolous allegation” concept, teased from Banks v. Garrett, 901 F.2d 1084 (Fed.Cir.1990), is not relevant to the determination of whether a statute is money-mandating, but instead applies to the facts alleged to support a claim under a statute that has already been determined to be money-mandating. Fisher II, 402 F.3d at 1172. Plaintiffs, however, also argue, citing Fisher I, that White Mountain Apache “established] a new test for determining whether a statute is money-mandating,” and that “[t]he new test clearly lowers the threshold fo[r] establishing that a statute or regulation is money-mandating, because it replaces a normal ‘fairly interpreted’ test with a less-demanding test of ‘reasonable amenability’ based on fair inferences.” Pis.’ Corr. Opp. at 6 n. 1 (citing *589Fisher I, 364 F.3d at 1377.). On this point, the Federal Circuit has stated, “[w]hether White Mountain alters the Mitchell test, ... and whether the new test is less stringent in some respects or is the same, ... is less than clear.” Fisher II, 402 F.3d at 1174.

Upon careful consideration of the issue, the Court concludes that the Supreme Court did not adopt a new test employing a lower threshold in the White Mountain Apache opinion. The issue arises because the Supreme Court, after stating the long-held test that “a statute creates a right capable of grounding a claim within the waiver of sovereign immunity if, but only if, it can fairly be interpreted as mandating compensation by the Federal Government for the damage sustained,” White Mountain Apache, 537 U.S. at 472, 123 S.Ct. 1126 (internal quotations and citations omitted), loosely paraphrased the test in the following dictum:

This “fair interpretation” rule demands a showing demonstrably lower than the standard for the initial waiver of sovereign immunity. “Because the Tucker Act supplies a waiver of immunity for claims of this nature, the separate statutes and regulations need not provide a second waiver of sovereign immunity, nor need they be construed in the manner appropriate to waivers of sovereign immunity.” Mitchell II, supra, at 218-19, 103 S.Ct. 2961. It is enough, then, that a statute creating a Tucker Act right be reasonably amenable to the reading that it mandates a right of recovery in damages. While the premise to a Tucker Act claim will not be “lightly inferred,” 463 U.S., at 218, 103 S.Ct. 2961, a fair inference will do.

Id. at 472-73, 123 S.Ct. 1126 (quoting Mitchell II, 463 U.S. at 218-19, 103 S.Ct. 2961). In its subsequent analysis of the claim at issue, the White Mountain Apache Court proceeds to discuss whether “it is fair to infer a damages remedy,” repeatedly states that it is deciding whether a “fair inference” supports a claim, see id. at 473-75, 477, 123 S.Ct. 1126, and contrasts “a plain and explicit statement standard [with] the less demanding requirement of fair inference,” id. at 477, 123 S.Ct. 1126.

As is noted in Fisher II, the four justices who dissented from the White Mountain Apache decision characterized the majority’s language in that case as “a newly devised approach” and “a new test.” Fisher II, 402 F.3d at 1174 (quoting White Mountain Apache, 537 U.S. at 482, 487, 123 S.Ct. 1126 (Thomas, J., dissenting)). Significantly, the four dissenters squarely stated that “the test to determine if Congress has conferred a substantive right enforceable against the Government in a suit for money damages is whether an Act can fairly be interpreted as mandating compensation by the Federal Government for the damage sustained.” White Mountain Apache, 537 U.S. at 481-82, 123 S.Ct. 1126 (citations and internal quotation marks omitted). Those four justices dissented, then, “[b]ecause the [Act in question] ‘cannot fairly be interpreted as mandating compensation by the Federal Government for damage sustained’ by the White Mountain Apache Tribe.” Id. at 482, 123 S.Ct. 1126 (quoting United States v. Testan, 424 U.S. 392, 400, 96 S.Ct. 948,47 L.Ed.2d 114 (1976)).

As the Federal Circuit also notes in Fisher II, two justices in a concurring opinion believed “that the majority opinion in White Mountain was not inconsistent with the opinion decided that same day in United States v. Navajo Nation,” in which “the majority spoke only in terms of the established ‘fairly be interpreted’ test.” Fisher II, 402 F.3d at 1174 (citing White Mountain Apache, 537 U.S. at 479, 123 S.Ct. 1126 (Ginsburg, J., concurring); U.S. v. Navajo Nation, 537 U.S. 488, 123 S.Ct. 1079, 155 L.Ed.2d 60 (2003)). The two concurring justices squarely stated that for the White Mountain Apache decision, “[t]he dispositive question, accordingly, is whether [the Act in question] ... is fairly interpreted to mandate compensation for the harm caused by maladministration of the property.” White Mountain Apache, 537 U.S. at 480, 123 S.Ct. 1126 (emphasis added). Thus, no matter how one spins the language employed in the majority opinion, it is clear that six justices of the Court stated that the applicable rule of decision for the case was the traditional “fairly interpreted” standard.

Moreover, the reason that the Supreme Court in White Mountain Apache had cause *590to use the term “fair inference” repeatedly throughout its analysis was the nature of the source of the alleged money mandate. As is discussed above, there are two basic types of statutes that are scrutinized as potential bases for a money mandate: statutes, like the one at issue here, that concern money payments, which require a determination whether payments are mandatory or discretionary; 6 and statutes, like the one at issue in White Mountain Apache, which place certain duties on the government. In the latter category, even if the promise to pay money damages is not expressly in the statute, these damages may nevertheless be implied by the statute as a necessary remedy for the government’s breach of its duties. See Mitchell II, 463 U.S. at 217 n. 16, 103 S.Ct. 2961 (“the substantive source of law may grant the claimant a right to recover damages either ‘expressly or by implication’ ”) (quoting East-port Steamship Corp. v. United States, 178 Ct.Cl. 599, 605, 372 F.2d 1002 (1967)). As people who determine if a law implies a damages remedy are inferring whether this is so, it was perfectly natural (if a tad confusing) for the Supreme Court to have discussed its interpretive process in terms of “inferences.” 7

A close reading of the confusion-causing passage from White Mountain Apache demonstrates that the Supreme Court could not have intended to change the legal test for determining whether a statute is money-mandating. It starts out innocently enough, with the uncontroversial description of the “fairly be interpreted” test as “demanding] a showing demonstrably lower than the standard for the initial waiver of sovereign immunity.” White Mountain Apache, 537 U.S. at 472, 123 S.Ct. 1126. This innocuous statement, however, was the springboard for the somewhat-loose statement that it is “enough” that a statute, to be money-mandating, “be reasonably amenable to the reading that it mandates a right of recovery in damages.” Id. at 473, 123 S.Ct. 1126. But the intervening quote from Mitchell II, that a money-mandating statute does not need to be “construed in the manner appropriate to waivers of sovereign immunity,” sheds light on the meaning of “reasonably amenable.” Id. at 472-73, 123 S.Ct. 1126 (quoting Mitchell II, 463 U.S. at 218-19, 103 S.Ct. 2961). All that the Supreme Court seems to be saying here is that, unlike the “unequivocally expressed” language that is required for a waiver of sovereign immunity, see, e.g., United States v. King, 395 U.S. 1, 4, 89 S.Ct. 1501, 23 L.Ed.2d 52 (1969), a statute may by implication supply the basis for a money-mandate.

The Supreme Court then concludes the passage with the statement that, “[w]hile the premise to a Tucker Act claim will not be ‘lightly inferred,’ 463 U.S., at 218, 103 S.Ct. 2961, a fair inference will do.” 537 U.S. at 473, 123 S.Ct. 1126 (quoting Mitchell II, 463 U.S. at 218, 103 S.Ct. 2961). Two aspects of this sentence add to the confusion — the contrast between inferences that are “lightly” made and those that are “fair.” This sentence might give the impression that an inference that is weaker than is normally used to interpret a statute would suffice. But the “lightly inferred” quote from Mitchell II is ripped from its context, where it had nothing to do with determining if a statute were money-mandating, and instead concerned “the general scope of the Tucker Act.” See Mitchell II, 463 U.S. at 218, 103 S.Ct. 2961. The Supreme Court was using those words to describe its reluctance to wander from the express waiver of sovereign immunity accomplished through that act. See id. (discussing the rejection of Tucker Act waivers for declaratory judgments, for the tolling of statutes of limitations, and for contracts implied in law). The Mitchell II Court was citing language in cases holding that “a waiver cannot be implied but must be unequivocally expressed,” King, 395 U.S. at 4, 89 S.Ct. 1501 (citing United States v. Sherwood, 312 U.S. 584, 61 S.Ct. 767, 85 L.Ed. 1058 *591(1941)), and “that limitations and conditions upon which the Government consents to be sued must be strictly observed and exceptions thereto are not to be implied,” Soriano v. United States, 352 U.S. 270, 276, 77 S.Ct. 269, 1 L.Ed.2d 306 (1957) (citing Sherwood, 312 U.S. at 590-91, 61 S.Ct. 767). These are all cases in which no inferences at all were used. Thus, the words “lightly inferred” were clearly meant to be an understatement — hardly the stuff from which to mold a new test. See Mitchell II, 463 U.S. at 218, 103 S.Ct. 2961 (citing King and Soriano).

The use of “fair,” as a counterpoint to “ligh[t],” compounds the confusion. One who is unfamiliar with the precedents shaping our Court’s jurisdiction might well take “fair” to mean “so-so,” or “okay,” or “arguable”— stricter than “lightly,” but maybe not by much. What the Supreme Court has constructed, though, is the intended tautology that a “fair interpretation” of a statute implying a money damages remedy must be based on “fair inferences.” This only becomes confusing because the Supreme Court had, at the outset of the passage, converted the adverb “fairly” to the adjective “fair,” translating the “can fairly be interpreted” test to the “ ‘fair interpretation’ rule.” See White Mountain Apache, 537 U.S. at 472, 123 S.Ct. 1126. Because of this converted modifier, and the focus on the inferential process of interpretation, a reader of the White Mountain Apache opinion could easily lose sight of just what the words of the traditional test meant.

The Supreme Court in White Mountain Apache provided no new analysis or explanation for the test it employed, but merely relied on Mitchell II as the source of the test. This in and of itself should be enough for the conclusion that no new test was intended, for the Supreme Court would hardly make a dramatic change in our Court’s jurisdiction by sleight-of-hand, or sotto voce. It was clearly referring to the “can fairly be interpreted” test when it coins the “ ‘fair interpretation’ rule.” See id. The Supreme Court led into the curious passage by properly quoting Mitchell II for the test of whether a statute “can fairly be interpreted as mandating compensation.” White Mountain Apache, 537 U.S. at 472, 123 S.Ct. 1126 (quoting Mitchell II, 463 U.S. at 217, 103 S.Ct. 2961).

The Mitchell II Court stated the test in terms of whether a statute “can fairly be interpreted,” 463 U.S. at 217-19, 226, 228, 103 S.Ct. 2961, and never once uses the formulation “fair interpretation.” The “can fairly be interpreted” language comes from United States v. Testan, 424 U.S. 392, 400, 96 S.Ct. 948, 47 L.Ed.2d 114 (1976), which in turn rested on our predecessor court’s decision in Eastport, 178 Ct.Cl. at 607, 372 F.2d 1002. See Mitchell II, 463 U.S. at 217, 103 S.Ct. 2961.8 Tracing the test back to East-port reveals conclusively that the adverb “fairly” was used as a synonym for “rightly,” “justly,” or “correctly.” When the test was first stated as whether the statute “can be fairly be interpreted as mandating compensation,” this was a paraphrase (in the converse) of the following sentence appearing earlier in the same paragraph:

Where the claimant is not suing for money improperly exacted or retained (the first class defined above), the historical boundaries of our competence have excluded those instances in which the basis of the federal claim — be it the Constitution, a statute, or a regulation — cannot be held to command, in itself and as correctly interpreted, the payment of money to the claimant, but in which some other principle of damages has to be invoked for recovery.

Eastport, 178 Ct.Cl. at 607, 372 F.2d 1002 (emphasis added). Things sometimes get lost in translation, but this Court can avoid such error by remembering the meaning of “fairly.” If a “fair interpretation” is the result when a statute is “fairly ... interpreted,” as the Supreme Court announces in White Mountain Apache, then this means a “correct interpretation.” And if this be done through inference, these inferences are to be correct ones.

*592To read “fair inference” to mean anything less than the normal inference used in interpreting a statute, moreover, would make little sense, particularly in light of Fisher IPs elimination of the “two-step process.” See 402 F.3d at 1172. The meaning of a statute when this Court determines if a case is within its jurisdiction is the same as its meaning when the Court determines the merits of the case. Id. at 1173. How could it be that a statute would require the government to pay money damages merely because it arguably can be read to require the government to pay money damages? To be close to something is not the same as being it. Surely, “good enough for government liability” is not the measure of our Court’s jurisdiction. The test for whether a statute is money-mandating has not changed — our Court must still determine whether the statute, correctly interpreted, would require a money damages remedy.

For all of the above reasons — the fact that six justices thought the traditional test applied; the Supreme Court’s use of “inference” as synonymous with “interpretation”; the recognition that any use of inferences at all is a lower standard than the express text needed to waive immunity; the lack of any explanation or analysis indicating a change in standards; the long-standing meaning of “fairly” as “correctly;” and the absurd result if one construes any law which could possibly be read as creating money damages as actually creating this remedy — the Court concludes that White Mountain Apache did not change the standard for determining whether a law is money-mandating.

2. Does the FLEPRA provision concerning FLPAs mandate the payment of money?

The critical question for our jurisdiction, then, is whether section 4523, either alone, under the relevant regulations, or through its interplay with section 267a, is fairly (meaning correctly) interpreted to require the payment of money damages.9 The Court starts, and usually ends, with the text of the laws or regulations that are in question, as it is presumably the intention of Congress and agencies, in the absence of scrivener’s errors or the like, to enact and adopt the actual laws and regulations that are respectively enacted or adopted. See, e.g., Hartford Underwriters Ins. Co. v. Union Planters Bank, N.A., 530 U.S. 1, 6, 120 S.Ct. 1942, 147 L.Ed.2d 1 (2000); Hughes Aircraft Co. v. Jacobson, 525 U.S. 432, 438, 119 S.Ct. 755, 142 L.Ed.2d 881 (1999); Chicago v. Environmental Defense Fund, 511 U.S. 328, 337, 114 S.Ct. 1588, 128 L.Ed.2d 302 (1994). In other words, the Congress does what it says and means what it says. As the Supreme Court has explained:

Legislative history can be a legitimate guide to a statutory purpose obscured by ambiguity, but “[i]n the absence of a ‘clearly expressed legislative intention to the contrary,’ the language of the statute itself ‘must ordinarily be regarded as conclusive.’ ” United States v. James, 478 U.S. 597, 606, 106 S.Ct. 3116, 92 L.Ed.2d 483 (1986) (quoting Consumer Product Safety Comm’n v. GTE Sylvania, Inc., 447 U.S. 102, 108, 100 S.Ct. 2051, 64 L.Ed.2d 766 (1980)). Unless exceptional circumstances dictate otherwise, “[w]hen we find the terms of a statute unambiguous, judicial inquiry is complete.” Rubin v. United States, 449 U.S. 424, 430, 101 S.Ct. 698, 66 L.Ed.2d 633 (1981).

Burlington Northern Railroad Co. v. Oklahoma Tax Comm’n, 481 U.S. 454, 461, 107 S.Ct. 1855, 95 L.Ed.2d 404 (1987).

Section 4523 plainly states that “[a]n agency may pay a cash award, up to 5 percent of basic pay, to any law enfoi'cement officer employed in or under such agency who possesses and makes substantial use of 1 or more foreign languages in the performance of official duties.” 5 U.S.C. § 4523(a) (emphasis added). Unlike the use of “shall” *593or “must,” which very obviously connotes mandatory action, laws employing “may” have been held on occasion to admit of some ambiguity. Because the provision does not use mandatory language, a very strong, but rebuttable, presumption arises that the PLPA provision is non-money-mandating. The Supreme Court has stated in this regard that:

The word “may,” when used in a statute, usually implies some degree of discretion. This common-sense principle of statutory construction is by no means invariable, however, and can be defeated by indications of legislative intent to the contrary or by obvious inferences from the structure and purpose of the statute.

United States v. Rodgers, 461 U.S. 677, 706, 103 S.Ct. 2132, 76 L.Ed.2d 236 (1983) (emphasis added) (footnote and citations omitted). The Court notes that while inferences may be used to convert “may” to “shall” through the interpretive process, these inferences are to be obvious. The Federal Circuit has recently reiterated this canon of statutory interpretation:

We may thus presume that when Congress used the word “may” in the statute in suit, we should use common sense and presume that the word conveys some degree of discretion. But we must proceed to test that presumption against the intent of Congress and other inferences that we may rationally draw from the structure and purpose of the statute at hand.

McBryde v. United States, 299 F.3d 1357, 1362 (Fed.Cir.2002). See also Hoch v. United States, 33 Fed.Cl. 39, 42-43 (1995) (citing Huston v. United States, 956 F.2d 259, 261 (Fed.Cir.1992), and Allen v. United States, 229 Ct.Cl. 515, 518, 1981 WL 22043 (1981)).

The strong presumption that “may” is permissive and discretionary, and not mandatory, has long been established. See, e.g., Rodgers, 461 U.S. at 706, 103 S.Ct. 2132; Haig v. Agee, 453 U.S. 280, 294 n. 26, 101 S.Ct. 2766, 69 L.Ed.2d 640 (1981) (“‘may expressly recognizes substantial discretion”); Fanners’ & Merchants’ Bank of Monroe v. Federal Reserve Bank of Richmond, 262 U.S. 649, 662-63, 43 S.Ct. 651, 67 L.Ed. 1157 (1923) (“It is true that in statutes the word ‘may is sometimes construed as ‘shall’. But that is where the context, or the subject-matter, compels such construction.”); Thompson v. Lessee of Carroll, 63 U.S. (22 How.) 422, 434, 16 L.Ed. 387 (1859) (regarding the construction of “may’ as “must”: “It is only where it is necessary to give effect to the clear policy and intention of the Legislature, that such a liberty can be taken with the plain words of a statute.”); Minor v. Mechanics Bank of Alexandria, 26 U.S. (1 Pet.) 46, 64, 7 L.Ed. 47 (1828) (“The ordinary meaning of the language, must be presumed to be intended, unless it would manifestly defeat the object of the provisions.”).

Here, the inference of discretion is confirmed by the structure of the statute. In subsection (a) of section 4523, Congress used the word “may’ to define the agencys power to provide FLPAs; in subsection (b), Congress used the word “shall” to define how the agencies are to administer FLPA programs. 5 U.S.C. § 4523(a)-(b). Given the textual proximity of “may and “shall” in FLEPRA, these words should be interpreted according to their ordinary meaning. See Huston, 956 F.2d at 262 (“When, within the same statute, Congress uses both ‘shall’ and ‘may,’ it is differentiating between mandatory and discretionary tasks.”).

Indeed, throughout FLEPRA, Congress distinguished between payments and pay adjustments that “shall” be made, and those that “may’ be made. See, e.g., Title IV, Pub.L. No. 101-509, 104 Stat. 1465-69 (1990), § 403(a) (“higher minimum rates and corresponding increases in all step rates ... shall be established”); § 403(c) (“higher minimum rates ... shall apply ... in the same manner ... and may be increased”); § 404(a) (“shall be paid any applicable special pay adjustment ... but such special pay adjustment shall be reduced”); § 404(b) (“shall receive an adjustment, which shall be a percentage”); § 406 (“shall reduce the rate of periodic payments”); § 407 (“may receive a relocation payment”); § 410(b) (“may be paid premium pay”) (adding 5 U.S.C. § 5547(c)(2)). The proper inference drawn from the distinction between “may’ and “shall” in the same statute further strengthens the presumption that “may’ is discretionary. See Huston, 956 *594F.2d at 262; Anderson v. Yungkau, 329 U.S. 482, 485, 67 S.Ct. 428, 91 L.Ed. 436 (1947) (“And when the same Rule uses both ‘may’ and ‘shall,’ the normal inference is that each is used in its usual sense — the one act being permissive, the other mandatory.”); Farmers & Merchants, 262 U.S. at 663, 43 S.Ct. 651 (“This statute appears to have been drawn with great care. Throughout the act the distinction is clearly made between what the Board and the reserve banks ‘shall’ do and what they ‘may’ do.”); United States ex. rel. Siegel v. Thoman, 156 U.S. 353, 359-60, 15 S.Ct. 378, 39 L.Ed. 450 (1895) (“In the law to be construed here it is evident that the word ‘may’ is used in special contradistinction to the word ‘shall,’ and hence there can be no reason for ‘taking such a liberty.’ ... In the first the word ‘shall’ and in the latter provision the word ‘may1 is used, indicating command in the one and permission in the other.”); see also Hopi Tribe v. United States, 55 Fed.Cl. 81, 87 (2002).

Nor does the purpose of FLEPRA give rise to any inferences, obvious or otherwise, that foreign language pay awards are mandatory. The parties agree that the “purpose of FLEPRA was to ‘make Federal enforcement agencies competitive with their state and local counterparts.’” Pls.’ Corr. Opp. at 10 (citing Def.’s Mot. to Dismiss at 22 (quoting 136 Cong. Rec. S2310-03, *S2326)). To this end, as recounted above, FLEPRA, among other things, required an increase in the rates of pay for law enforcement officers, § 403, and special pay adjustments for officers in certain localities, § 404; and authorized discretionary relocation payments, § 407, and foreign language awards, § 408. Congress plainly and clearly used permissive language for some provisions and mandatory language for others, and it can hardly be said that construing the FLPA as mandatory “is necessary to give effect to the clear policy and intention of the Legislature,” Thompson, 63 U.S. (22 How.) at 434, or that following the natural reading of “may” “would manifestly defeat the object of the provisions,” Minor, 26 U.S. (1 Pet.) at 64. In making federal agencies competitive with other law enforcement forces, Congress chose to require some pay enhancements, and left others as optional tools in the hands of the various agencies.

Despite the clarity of the FLEPRA language, there is always legislative history. But “[wjhere the plain language of the statute would settle the question before the court, the legislative history is examined with hesitation to determine whether there is a clearly expressed legislative intention contrary to the statutory language.” Madison Galleries, Ltd. v. United States, 870 F.2d 627, 629 (Fed.Cir.1989) (citing INS v. Cardoza Fonseca, 480 U.S. 421, 432 n. 12, 107 S.Ct. 1207, 94 L.Ed.2d 434 (1987)). Plaintiffs argue that “[djefendant’s reading of [the] statute would render such awards wholly discretionary and there is no support for that in the legislative history or elsewhere.” Pis.’ Corr. Opp. at 10. But the only legislative history available on the topic of FLPAs confirms that these were discretionary in nature, as the Conference Report read:

Agencies will have the discretion to pay sums up to 5% of base pay to eligible Federal law enforcement employees whom the agency determines have demonstrated a level of proficiency in a foreign language and where a need exists to use that language in the performance of their duties.

H.R. Conf. Rep. No. 101-906, at 91 (1990) (emphasis added). Thus, payment of FLPAs may not be “wholly” discretionary, in the sense that they cannot be given unless the recipient is proficient in a foreign language that is needed to perform his or her duties. But whether or not to give FLPAs to eligible individuals is wholly within the agency’s discretion.

With the language of the text, the structure and purpose of the law, and the legislative history all against them, plaintiffs nevertheless maintain that FLEPRA is money-mandating based on cases concerning the only two statutes which the Federal Circuit has held to contain a money-mandatory “may” — the moiety statute found at Title 19, section 1619, and the statute providing for payment of certain attorneys’ fees of federal judges, 28 U.S.C. § 463. See Pls.’ Reply at 2, 5 (discussing Doe v. United States, 100 F.3d 1576 (Fed.Cir.1996)); Pls.’ Corr. Opp. at *5959 (discussing McBryde, 299 F.3d 1357).10

The latter case may well be sui generis. In McBryde, the Federal Circuit determined that to construe a statute as conferring upon the Administrative Office of the United States Courts a discretionary power to pay judges’ attorneys’ fees would “raise serious constitutional questions,” because this would pose a “threat to the independence of the Judiciary.” McBryde, 299 F.3d at 1364 (internal quotation marks and citations omitted); see Hopi Tribe, 55 Fed.Cl. at 91. No such concerns are present in this case. The Court has, however, followed the guidance of McBryde, quoted above, and has “test[ed]” the “common-sense ... presumption” that “may” connotes discretion “against the intent of Congress and other inferences that we may rationally draw from the structure and purpose of the statute at hand.” McBryde, 299 F.3d at 1362. As discussed above, nothing has been found to undermine the discretionary nature of section 4523.

The various versions of the moiety statute have had a long history before this Court, the Federal Circuit, and our mutual predecessor*, the Court of Claims. In Tyson v. United States, 91 Ct.Cl. 139, 32 F.Supp. 135 (1940), the Court of Claims explained that its jurisdiction could extend over petitions challenging the Secretary of the Treasury’s denial of a claim for an informer’s fee. The moiety statute then in effect provided that certain informers whose “original information” led to a recovery of custom duties or associated fines, penalties, or forfeitures “may be awarded and paid by the Secretary of the Treasury a compensation of 25 per centum of the net amount recovered, but not to exceed $50,000 in any case.” Id. at 140, 32 F.Supp. 135 (quoting Section 619 of the Tariff Act of 1930). The Court of Claims rejected the argument that “the Secretary’s action in refusing to pay the award is final and conclusive, and that [the Court of Claims] has no jurisdiction to review it.” Id at 141, 32 F.Supp. 135.

The Court of Claims explained:

Congress____intended to confer upon an informer an absolute right to demand the payment of the award when he had met the conditions precedent thereto laid down by Congress____[W]hen the information was the first information which the Secretary had had, and when that information led to the recovery of duties, or of a fine, penalty, or forfeiture, then the informer was entitled as of right to the payment of the award, and if the Secretary of the Treasury arbitrarily or capriciously refused to pay it, the informer had the right to file suit in court to compel that payment.

Id at 141-42, 32 F.Supp. 135. The Court based its decision on two precedents construing “shall” statutes, Ramsay v. United Stales, 21 Ct.Cl. 443, 450, 1800 WL 1542 (1886), aff'd 120 U.S. 214, 30 L.Ed. 582 (1887) , and United States v. Laughlin, 249 U.S. 440, 442, 443, 39 S.Ct. 340, 63 L.Ed. 696 (1919). See Tyson, 91 Ct.Cl. at 142-43, 32 F.Supp. 135. The relevant statute in Ramsay read, “[t]he Secretary of the Treasury, under general regulations to be by him prescribed, shall determine whether any claimant is entitled to such share as above limited, and to whom the same shall be paid, and shall make payment accordingly.” Ramsay, 21 Ct.Cl. at 449, 1800 WL 1542 (emphasis added). The Laughlin case concerned a statute that provided, “[t]hat in all cases where it shall appear to the satisfaction of the Secretary of the Interior that any person has heretofore or shall hereafter make any payments to the United States under the public land laws in excess of the amount he was lawfully required to pay under such *596laws, such excess shall be repaid to such person or to his legal representatives.” Laughlin, 249 U.S. at 442, 39 S.Ct. 340 (emphasis added).

The Tyson Court concluded that, “as in the Laughlin and Ramsey [sic11] cases, supra, we think Congress intended that the determination of the facts set out by Congress as a prerequisite to the right to claim the award should be committed to the Secretary of the Treasury.” Tyson, 91 Ct.Cl. at 143, 32 F.Supp. 135. The Court thus interpreted the phrase, “may be awarded and paid by the Secretary of the Treasury,” as not placing the decision at his “uncontrolled judgment or discretion,” but requiring that the Secretary make a factual determination that could then be reviewed by the Court to see if it was “supported by no evidence, if it was arbitrary or capricious, or if the proceedings were fatally defective.” Id.

Apparently, the informer compensation statute at issue in Ramsay was based on the same type of “conditions precedent” as the moiety statute in Tyson. See Ramsay, 21 Ct.Cl. at 449 (noting claimant was found to be the “first informer,” and “the information given by him led to the recovery” of tax penalties). The statute in Laughlin similarly was based on a clear condition precedent: excess payments under public land laws. See Laughlin, 249 U.S. at 443, 39 S.Ct. 340. From Tyson, our Court has identified two factors to consider in deciding whether to override the normal (very strong) presumption that Congress, in using permissive language, intended that money awards to informants were to be discretionary, not mandatory: whether the “statutes enumerate specific requirements that must be fulfilled before an informant acquires an absolute right to compensation,” Hoch v. United States, 33 Fed.Cl. 39, 43 (1995); and whether they “state a sum certain” to be paid, id. at 44. See id. at 44-45 (finding 28 U.S.C. § 524(c) not to be money-mandating because it imposed only general conditions precedent to an award, and the amount of the award was not a sum certain); Confidential Informant v. United States, 46 Fed. Cl. 1, 5-6 (2000) (holding 26 U.S.C. § 7623 was not money-mandating, as it did not provide specific eligibility requirements for informants, and it only authorized a ceiling for awards).

In Doe v. United States, 100 F.3d 1576 (Fed.Cir.1996), the Federal Circuit construed 19 U.S.C. § 1619 (“section 1619”), whose predecessor was at issue in Tyson. A decision of this Court had held the provision to be non-money-mandating, based on the statute’s 1986 amendments. Doe v. United States, 32 Fed.Cl. 472, 474-75 (1994). Prior to the amendments, a section 1619 award was set at twenty-five percent of the net amount recovered due to the informant’s original information, but in no event could it exceed $50,000. See Doe, 100 F.3d at 1579. Following the amendments, a section 1619 award was no longer set at twenty-five percent, but instead could “not exceed 25 percent of the net amount so recovered.” Id. (quoting 19 U.S.C. § 1619(a)). The interjection of discretion in the determination of the size of the award was held by our Court to change the money-mandating nature of the moiety statute. Doe, 32 Fed.Cl. at 474-75; but see Lewis v. United States, 32 Fed.Cl. 59, 64 (1994).

The Federal Circuit reversed. Given that for several decades the moiety statute had been construed to come within the jurisdiction of the Court of Claims, see Doe, 100 F.3d at 1580 (collecting cases), the Circuit noted that it did “not write on a clean slate, or come to this provision of law in the abstract.” Id. at 1581. Because section 1619 “has long been understood to mandate the payment of an award” when “claimants satisfied] the statutory conditions,” the normal presumption was flipped, as “Congress may be presumed to have been aware of this construction when it amended the statute in 1986.” Id. at 1580. Starting, then, from the presumption that the statute had been intended to mandate award payments, the Federal Circuit found the introduction of “some discretion” in the amount of the award to not *597deprive our Court of jurisdiction, as “Congress provided no indication in either the statutory language or legislative history that it intended to reject or alter this construction and to render awards wholly discretionary.” See Doe, 100 F.3d at 1581-82. Thus, section 1619 “continues to require the payment of some award to claimants who have met the statutory conditions for recovery.” Id. at 1582.

Whether the informant award precedents are applicable in construing statutes concerning awards to federal employees is debatable. It might be more logical to presume that Congress intends the stronger incentive of a mandatory payment to induce people to come forward with information concerning illegal activity, compared to an award for using a second language at work, which, after all, is already part of one’s job. Under FLEPRA, law enforcement officers eligible for awards are already employed by the federal government; the employment relationship provides an incentive for the officers to use their Spanish-language skills in the performance of their official duties regardless of eligibility for an FLPA. Presumably, work evaluations and promotions are enhanced by such activity-

To draw an analogy, FLEPRA is much closer to the civil service incentive award program, 5 U.S.C. § 4503 (“section 4503”), than it is to the informant award statutes in the Tyson line of cases. Like FLEPRA and generally unlike the informant statutes, section 4503 applies expressly to federal employees. And like FLEPRA, persons whose deeds merit cash recognition under section 4503 would already be motivated by the desire to receive favorable performance evaluations and promotions.

Section 4503 states:

The head of an agency may pay a cash award to, and incur necessary expense for the honorary recognition of, an employee who—

(1) by his suggestion, invention, superior accomplishment, or other personal effort contributes to the efficiency, economy, or other improvement of Government operations or achieves a significant reduction in paperwork; or

(2) performs a special act or service in the public interest in connection with or related to his official employment.

Our Court has held that section 4503 is not money-mandating because it is phrased in discretionary terms. Rosano v. United States, 9 Cl.Ct. 137, 144-45 (1985). See McGee v. United States, 5 Cl.Ct. 480, 481-82 (1984). Cf. Adair v. United States, 227 Ct.Cl. 345, 350-51, 648 F.2d 1318 (1981) (holding that variable incentive pay statute not money-mandating).12

In any event, even applying the factors identified in Tyson and Hoch to FLEPRA does not change the Court’s conclusion that Congress intended FLPAs to be discretionary payments. In Tyson, the requirements necessary for award eligibility were clear and did not admit of agency interpretation. The informant had to provide original information, and that information had to lead to a successful enforcement action. In contrast, eligibility under FLEPRA requires “substantial” use of a foreign language. “Substantial” can have a varying content; Congress recognized that fact by providing the agencies with regulatory authority in section 4523(b) to define its content. No such authority was provided for moiety awards. The agency’s role under FLEPRA is not restricted to “the determination of the facts set out by Congress,” Tyson, 91 Ct.Cl. at 143, 32 F.Supp. 135, but also involves the creation of the very standard to which the facts will be applied.

Second, FLEPRA does not establish a sum certain, but instead caps any award at five percent of basic pay, and thus resembles the *598capped award found beyond our jurisdiction in Hoch, 33 Fed.Cl. at 44-45. Unlike the 1930 Tariff Act moiety statute, which in addition to setting a cap also set awards below the cap at twenty-five percent of recovered amounts, section 4523 would allow any award from five percent of basic pay or lower. Compare Tyson, 91 Ct.Cl. at 140, 32 F.Supp. 135 with 5 U.S.C. § 4523(a). And unlike the moiety statute considered in Doe, FLEPRA was “writ[ten] on a clean slate,” Doe, 100 F.3d at 1581, freshly minted by Congress without decades of precedent establishing that it was money-mandating.

Since FLPAs are not a sum certain, and FLEPRA does not provide specific eligibility criteria, the Tyson analysis does not provide a basis for disregarding Congress’ use of permissive language in the FLPA provision. Further, unlike the moiety statutes, FLEPRA contains both “may” and “shall,” as is discussed above, which further reduces the likelihood that Congress could have meant “may” to connote something other than discretion. See Huston, 956 F.2d at 262; see also Anderson, 329 U.S. at 485, 67 S.Ct. 428; Thoman, 156 U.S. at 359-60, 15 S.Ct. 378. Section 4523 is not, in and of itself, money-mandating.

It is still, of course, possible that Customs could have made FLPAs money-mandating, by adopting regulations that would create an entitlement if certain specific requirements are met. But plaintiffs are not arguing that they have been deprived of FLPAs to which they were entitled under the regulations adopted by Customs under section 4523. Instead, they argue that the regulations adopted by Customs under section 267a, to apply to other Customs officers, should apply also to them. Thus, the money-mandate, according to plaintiffs, was triggered by the 1996 adoption of FLPAs for other Customs officers. See Pis.’ Corr. Opp. at 11-12.

Section 267a authorizes the Secretary to make FLPAs available to customs officers “to the same extent and in the same manner as would be allowable under [FLEPRA] with respect to law enforcement officers.” 19 U.S.C. § 267a. The error in the plaintiffs’ analysis is their interpretation of the verbal “would be.” The plaintiffs’ construction requires that the Court read the statute to permit the award of FLPAs to Customs officers as FLPAs “are awarded ... to law enforcement officers.” If section 267a did read so, one could plausibly argue that a FLPA program could only be established for customs officers if the same program were established for law enforcement officers working for Customs. But the statute does not read that way. The use of the conditional verbal “would be” means that the Secretary has under section 267a the same FLPA power with regard to customs officers as he has with regard to law enforcement officers. Contrary to the plaintiffs’ contentions, nothing in the text forbids the Secretary from having an FLPA program for one type of officer and not for another. The Secretary’s decision so to limit FLPA availability might violate the Administrative Procedure Act (“APA”), 5 U.S.C. §§ 701-706, if arbitrary or capricious, but that claim is not brought to this Court. For the same reasons, the plaintiffs’ contention that the regulations pursuant to section 267a must be substantially indistinguishable from those pursuant to section 4523 is without merit.

The plaintiffs also argue that section 4523 becomes money-mandating once the Secretary has created an FLPA program. According to plaintiffs, “Congress intended to grant to agencies the discretion on whether or not to have a foreign language proficiency award program for law enforcement officers.” Pis.’ Corr. Opp. at 8. But once this discretion is exercised to adopt some FLPA program, no matter how discretionary that program appears to be, plaintiffs argue there is a money-mandate that allows them into this Court. According to this theory, although FLEPRA does not require the adoption of an FLPA program, if one is adopted, this Court becomes a forum in which the promulgated regulations may be challenged. See Pis.’ Corr. Opp. at 8-9; Pis.’ Reply at 4-5.

Instead of identifying a regulation promulgated pursuant to section 4523(b) mandating that they receive FLPAs, the plaintiffs argue that regulations should entitle to them to FLPAs, but that the actual regulations in force from 1998 to at least 2002 were overly *599restrictive. The plaintiffs contend that these regulations were contrary to the text of section 4528, in that Customs declined to credit non-standard work week hours toward the FLPA substantial use requirement. But under the deference this Court must give such agency interpretations of the statutes they administer, see Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984), it cannot be said that this conflicts with section 4523. That provision gives broad discretion to the agencies to adopt “procedures under which foreign language proficiency shall be ascertained,” and “criteria for the selection of individuals for recognition.” 5 U.S.C. § 4523(b)(1),(2). Moreover, the plaintiffs recognize that one of the chief concerns influencing whether an agency wished to establish an FLPA program is whether it would have sufficient funds. See Am. Compl. Exs. 1, 3. Regardless of whether or not this Court would have counted non-standard work week hours toward the measure of “substantial use” of a foreign language “in the performance of official duties,” Customs had the discretion to tailor its FLPA program to fit both its needs and its budget.

What plaintiffs seek in this regard is a declaratory judgment that non-money-mandating regulations are inconsistent with a non-money-mandating statute. Perhaps this complaint could be brought in the district court under the APA. District courts are empowered to hold unlawful and set aside any agency action that is “in excess of statutory ... authority.” 5 U.S.C. § 706(2)(C). See, e.g., Pender Peanut Corp. v. United States, 20 Cl.Ct. 447, 452 (1990); NLRB Union v. Federal Labor Relations Authority, 834 F.2d 191, 195 (D.C.Cir.1987). If the regulations were struck down as unlawful, then Customs would have to decide whether it would re-write them, or, if it determines that following the district court interpretation of the statute would be too costly, discontinue the FLPA program. As noted above, plaintiffs concede that whether to have such a program is discretionary. But this Court cannot re-write the regulations to provide a money mandate where one does not exist. See Adair, 227 Ct.Cl. at 351-53, 648 F.2d 1318. All courts may have the power to decide if their jurisdiction extends over a matter, but no court has the power to manufacture it.

In sum, there exists no law or regulation that can fairly be interpreted to mandate the payment of money, upon which jurisdiction over plaintiffs’ claims may be based. The text of FLEPRA uses discretionary language in the FLPA provision, but mandatory language elsewhere. Nothing in the structure or purpose of the law, or even its legislative history, supports the suggestion that Congress intended to create an entitlement to foreign language pay awards. Nor does the establishment of pay awards under section 267a trigger any such mandate. Indeed, even under “fair inferences,” section 4523 is not reasonably amenable to the reading that it mandates money damages.

III. CONCLUSION

For the reasons stated above, the Court GRANTS the defendant’s motion to dismiss for want of subject matter jurisdiction pursuant to RCFC 12(b)(1). Plaintiffs’ cross-motion for partial summary judgment is DENIED as moot. The Clerk is directed to close the case.

IT IS SO ORDERED.

. Customs had been a bureau within the Treasury Department up until March 9, 2003, when the relevant functions and workforce were transferred to the Department of Homeland Security’s Bureau of Immigration and Customs Enforcement. Am. Compl. ¶ 17; Contreras v. Ridge, 305 F.Supp.2d 126, 128 n. 1 (D.D.C.2004).

. For purposes of the motion to dismiss, the Court assumes that the facts as alleged in the Amended Complaint are true. Perez v. United States, 156 F.3d 1366, 1370 (Fed.Cir.1998).

. The standard work week was defined as Monday through Friday, 9 a.m. to 5 p.m. See Am. Compl. ¶ 3.

. The Clerk entered judgment on the grant of summary judgment on February 26, 2004. Contreras moved for reconsideration or 28 U.S.C. § 1292(b) certification on March 11, 2004. The District Judge denied that motion May 21, 2004. Am. Compl. ¶15. Thus, the Court is not presented with a 28 U.S.C. § 1500 issue.

. Pub.L. No. 95-454, § 202(a), 92 Stat. 1111 (1978).

. As is discussed in the next section below, construing these statutes also involves the use of inferences, but to a lesser degree.

. Indeed, the Supreme Court must have meant "inference” to mean the same thing as "interpretation.” If it did not, the following sentence from the opinion would be utterly inexplicable: "Together they show that a fair inference will require an express provision, when the legal current is otherwise against the existence of a cognizable claim.” White Mountain Apache, 537 U.S. at 478, 123 S.Ct. 1126.

. Testan never employs the "fair interpretation” formulation, but states the test as whether the statute "can be fairly interpreted” both places it is mentioned. 424 U.S. at 400, 402, 96 S.Ct. 948.

. Plaintiffs identify the Back Pay Act, 5 U.S.C. § 5596, as the technical basis for jurisdiction, Am. Compl. ¶ 8, but concede that this statute is "merely derivative in application,” Pis.’ Corr. Opp. at 7 (citing United States v. Connolly, 716 F.2d 882, 887 (Fed.Cir.1983)), and is “a 'money-mandating’ statute when based on violations of statutes or regulations covered by the Tucker Act.” Worthington v. United States, 168 F.3d 24, 26 (Fed.Cir.1999) (citing Connolly, 716 F.2d at 887).

. The plaintiffs also cite to Sawyer v. United States, 930 F.2d 1577 (Fed.Cir.1991). See Pis.' Reply at 6. Sawyer concerned a serviceman who had become disabled while on active duty but was otherwise entitled to basic pay. The plaintiff sought disability retirement under 10 U.S.C. § 1201, which permits a military Secretary to remove from active duty a disabled serviceman and provide him retirement pay. The government contended that section 1201 was discretionary and therefore not money-mandating, but the Federal Circuit disagreed; it determined that the "may” in section 1201 refers to the Secretary's decision whether to retire a disabled active-duty serviceman and not to the Secretary’s statutory obligation to provide retirement pay to the serviceman who has been removed because of disability. Sawyer, 930 F.2d at 1580. Thus, the "may” issue in Sawyer was simply a red herring offered by the government; the statute on its face clearly mandated money.

. The informer’s surname was spelled "Ramsey” in the opinion concerning his initial claim, but was spelled "Ramsay" in the subsequent opinions concerning his widow’s claim, upon which the court relied. Compare Ramsey v. United States, 14 Ct.Cl. 367, 1800 WL 1055 (1878), with Ramsay, 21 Ct.Cl. 443, 1800 WL 1542.

. The Court of Claims in Griffin v. United States, 215 Ct.Cl. 710, 1978 WL 5754 (1978), implicitly recognized the error in taking jurisdiction of an incentive award dispute when it held that if the government accepted the employee’s suggestion, an implied-in-fact contract — not a claim founded directly upon the award statute — would arise between the employee and the government supporting jurisdiction; the implied contract claim would therefore consist of the difference between the amount actually awarded and the amount the plaintiff sought. See id. at 714-15, 1978 WL 5754 (also noting that without the published procedures for award qualification, the Secretary would have non-reviewable discretion whether to give an award).