School District of City of Pontiac v. Secretary of US Department of Education

SUTTON, Circuit Judge,

concurring in the order.

Judge Cole and Judge McKeague have written well-reasoned opinions about this difficult case, but I find myself unable to join either one in full.- I write separately to explain my position, which comes down to embracing Judge Cole’s standing and justiciability conclusion and Judge McKeague’s merits conclusion at the panel stage.

I.

Standing. I agree with Judge Cole that the school districts have Article III standing to file this lawsuit. Cole Op. at 260-62. The school districts have alleged a sufficient “injury in fact”: the Secretary’s threatened withholding of federal funds to which they are otherwise entitled. In a declaratory judgment action, “injury in fact” turns on whether the threatened harm is real and imminent, as opposed to speculative and distant. See MedImmune, Inc. v. Genentech, Inc., 549 U.S. 118, 127-28, 127 S.Ct. 764, 166 L.Ed.2d 604 (2007). This threatened harm is real and sufficiently imminent. The Secretary has consistently denied waivers seeking to evade the Act’s requirements due to insufficient federal funding, and the school districts fairly allege that “noncompliance with [the Act’s] mandates due to lack of funds will result in the withholding of ... federal funds.” Compl. ¶ 16; see Lujan v. Defenders of Wildlife, 504 U.S. 555, 561-62, 112 S.Ct. 2130, 119 L.Ed.2d 351 (1992). Putting the school districts to the choice of abandoning their legal claim or risking sanctions “is a dilemma that it was the very purpose of the Declaratory Judgment Act to ameliorate.” MedImmune, 549 U.S. at 128, 127 S.Ct. 764 (internal quotation marks omitted). South Dakota v. Dole, 483 U.S. 203, 205, 107 S.Ct. 2793, 97 L.Ed.2d 171 (1987), involved a similar threat — that the government would withhold federal funds — and the Court resolved the merits of that declaratory judgment action without saying a word about standing.

The injury also is fairly traceable to the challenged action of the Secretary, who has authority to withhold the funds and who has declined to relax the testing and assessment requirements of the Act, even if the school districts cannot meet the requirements with federal funds alone. See Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 528 U.S. 167, 180, 120 S.Ct. 693, 145 L.Ed.2d 610 (2000). And enjoining the Secretary from withholding federal funds for failing to comply with the Act’s requirements will redress the threatened injury. See id. at 180-81, 120 S.Ct. 693; Cole Op. at 262.

(As an aside, the school districts separately claim injury in fact due to their ongoing expenditure of local funds to comply with the Act. See also Cole Op. at 261-62. But this theory does not necessarily *279establish a redressable injury. An injunction against the Secretary will stop him from cutting off federal funds for noncompliance, but it will not require the States to roll back their statutes and regulations implementing the Act’s requirements. A State’s alteration of its own statutes and regulations turns on “choices made by independent actors not before the courts and whose exercise of broad and legitimate discretion the courts cannot presume either to control or to predict.” ASARCO Inc. v. Kadish, 490 U.S. 605, 615, 109 S.Ct. 2037, 104 L.Ed.2d 696 (1989). The pleadings do not disclose how Michigan, Texas and Vermont “implement[] a single, statewide State accountability system” ensuring adequately yearly progress, 20 U.S.C. § 6311(b)(2)(A), but their implementing regulations almost assuredly affect the school districts. Yet because one redressable injury suffices to establish standing, the validity of this alternative theory makes no difference.)

Ripeness. The claims also are fit for judicial review, and the districts would suffer hardship if we declined to consider them. See Abbott Labs. v. Gardner, 387 U.S. 136, 149, 87 S.Ct. 1507, 18 L.Ed.2d 681 (1967); Cole Op. at 262-63. Because this dispute raises an unvarnished question of law that will not benefit from further factual development, it presumptively is ripe for review. See Dixie Fuel Co. v. Comm’r of Social Sec., 171 F.3d 1052, 1058 (6th Cir.1999), overruled on other grounds by Barnhart v. Peabody Coal Co., 537 U.S. 149, 157, 123 S.Ct. 748, 154 L.Ed.2d 653 (2003). “Nothing would be gained by postponing a decision” when “[t]he issue presented in this ease is purely legal, ... will not be clarified by further factual development” and hardship to the parties exists. Thomas v. Union Carbide Agric. Prods. Co., 473 U.S. 568, 581-82, 105 S.Ct. 3325, 87 L.Ed.2d 409 (1985); see Cole Op. at 263-64.

While further administrative proceedings might sharpen the nature of some of the school districts’ claims, they would not alter or make more concrete the nature of the legal question. The Secretary has made his position about the meaning of the Act clear in public statements, in his stance in this case and in related litigation pending in the Second Circuit. Requiring the school districts to seek waivers or propose plan revisions that the Secretary has confirmed he will not grant — because his interpretation of the Act prevents him from doing so — would merely prolong the litigation, already entering its fifth year. Nor would it help any of the participants in this case, least of all the students attending the affected schools, which presumably is why no party raised this issue on its own. See Union Carbide, 473 U.S. at 581-82, 105 S.Ct. 3325. No Lawyer Left Behind is not the name of the Act.

Civil Rule 19(a). I do not agree that Michigan, Texas and Vermont are necessary parties, now called “required” parties, under Rule 19(a) and cannot agree that we must dismiss the complaint in their absence. Let me start by resisting the idea that we must, or should, suddenly invoke Rule 19 to resolve this case — a rule that would appear for the first time in the record and pleadings of this case in our decision dismissing the complaint. I accept that the federal courts may, as a matter of discretion, raise required-party issues under Rule 19(a) on their own and may do so for the first time on appeal— even when the parties have forfeited the issue. Provident Tradesmens Bank & Trust Co. v. Patterson, 390 U.S. 102, 111, 88 S.Ct. 733, 19 L.Ed.2d 936 (1968); see also Bowling Transp., Inc. v. N.L.R.B., 352 F.3d 274, 281 (6th Cir.2003). But that does not make the issue one of subject matter jurisdiction, which it is not. Lincoln Property Co. v. Roche, 546 U.S. 81, *28090, 126 S.Ct. 606, 163 L.Ed.2d 415 (2005). And it does not establish that we should insert Rule 19(a) into a case whenever it occurs to us, only when it is prudent. This case, and the equally high-profile case pending in the Second Circuit, have been going on since 2005. See Connecticut v. Spellings, 453 F.Supp.2d 459, 465 (D.Conn.2006). The three affected States — Michigan, Texas and Vermont — surely know about this litigation and just as surely have made considered decisions not to join it. Indeed, it is quite possible that, even if we required the Governors of these States or the Secretaries of their respective departments of education to join the case, they still would opt not to take a position on the issue at hand. I am not sure we can make a State take a stand on the point, and I know we cannot make it take a coherent one. What we might gain from such a ruling thus is far from clear.

What is more, after the parties filed supplemental briefs before the en banc court and after the en banc oral argument, we sua sponte raised three “justiciability” questions. While the third question asked whether we should proceed in the absence of the States, we did not ask the parties to brief Rule 19 — not classically thought of as a justiciability doctrine — and we still have not done so. Not surprisingly, in their supplements to the supplemental en banc briefs, no party- — and not one of the four amid in their merits briefs — identified a lurking Rule 19 problem, and it remains quite possible that all of the participants in this case disagree that there is one. Cf. Huber v. Taylor, 532 F.3d 237, 249 (3d Cir.2008). While no party to the case would be surprised if we resolved the case on ripeness grounds, relying in part on the absence of the States, all parties to the case (I suspect) would be surprised to see us dismiss the case under Rule 19. No doubt, had we discovered an absence of subject matter jurisdiction at this late stage of the case, we would have to dismiss the dispute, no matter how lamentable that development would be, no matter how astonished the parties would be. But I am aware of no precedent that compels us sua sponte to insert Rule 19 into the case at this point and to dismiss the case in its fifth year of litigation. In my view, we should answer the question of statutory interpretation that the school districts’ complaint asked us to answer and leave Rule 19 out of it.

Be that as it may, Michigan, Texas and Vermont are not “required” parties under Rule 19(a)(1). An entity is a “required party” if:

(A) in that person’s absence, the court cannot accord complete relief among existing parties; or
(B) that person claims an interest relating to the subject of the action and is so situated that disposing of the action in the person’s absence may:
(i) as a practical matter impair or impede the person’s ability to protect the interest; or
(ii) leave an existing party subject to substantial risk of incurring double, multiple, or otherwise inconsistent obligations because of the interest.

Fed.R.CivJP. 19(a)(1).

The first ground for treating the three States as required parties — that the court “cannot accord complete relief’ without them, Fed.R.Civ.P. 19(a)(1)(A) — does not exist. The school districts seek to enjoin the Secretary “from withholding from states and school districts any federal funds ... because of a failure to comply with the mandates of the NCLB that is attributable to a refusal to spend non-NCLB funds.” Compl. at 58. Enjoining the Secretary from withholding federal funds from the school districts and the States does not require the presence of the *281States. It requires only the presence of the Secretary.

It also is far from clear that the three States claim the kind of “interest,” Fed. R.Civ.P. 19(a)(1)(B), that is cognizable under the other two grounds for treating the States as required parties. The States do not have a legally protected interest that resolution of the Pennhurst question would threaten, a missing ingredient that is fatal in several circuits. See, e.g., United Keetoowah Band of Cherokee Indians of Okla. v. United States, 480 F.3d 1318, 1324-25 (Fed.Cir.2007); Liberty Mut. Ins. Co. v. Treesdale, Inc., 419 F.3d 216, 230 (3d Cir.2005); Makah Indian Tribe v. Verity, 910 F.2d 555, 558 (9th Cir.1990). The States instead have an “interest in promoting what [they] regard[] as enlightened public policy,” Am. Maritime Transp., Inc., 870 F.2d 1559, 1562 (Fed.Cir.1989)— namely, their vision of how the Act should be interpreted — and have an interest concerning the obligations that come with accepting funding under the Act, see Liberty Mut. Ins. Co., 419 F.3d at 230 (holding “merely a financial interest” is not enough under Rule 19(a)(1)(B)). I recognize that we have not yet spoken clearly on what types of interests Rule 19 encompasses, and I see no need to take a stand today. I note only that the States’ presumed “interests” in this case are a distant cry from the types of interests that normally implicate Rule 19(a)(1)(B), and that they are more reminiscent of the kinds of interests that require just notice, not mandatory joinder. See Fed R. Civ. P. 5.1 (requiring a party to notify the state attorney general when challenging a state law but not requiring a court to consider whether the state attorney general is a necessary party); see also Republic of Philippines v. Pimentel, — U.S. —, 128 S.Ct. 2180, 2185-86, 2189, 171 L.Ed.2d 131 (2008) (finding the Republic of the Philippines a required party in case involving claims against the property of Ferdinand Marcos); Minnesota v. United States, 305 U.S. 382, 386-88, 59 S.Ct. 292, 83 L.Ed. 235 (1939) (condemnation dispute); Hooper v. Wolfe, 396 F.3d 744, 747-48 (6th Cir.2005) (recovery of diverted assets); Keweenaw Bay Indian Community v. Michigan, 11 F.3d 1341, 1343-44, 1347 (6th Cir.1993) (tribal fishing rights).

Even if the States have a cognizable interest, however, they are not required parties under other requirements of Rule 19(a)(1)(B). Proceeding without them will not “as a practical matter impair or impede” their ability “to protect the[ir] interest[s].” Fed.R.Civ.P. 19(a)(l)(B)(i). When States stick their heads in the sand for nearly five years of litigation about a high-profile lawsuit, it is difficult to say that proceeding without them will impair their interests — which so far seem focused above all on not being forced to take a public stand on the issues presented.

Nor is there any risk of prejudice to the States if we proceed to the merits without them. See Republic of the Philippines v. Pimentel, 128 S.Ct. at 2189. The matter at hand is not a challenge to the constitutionality of a state law or even to the meaning of a state law. It concerns the meaning of a federal law: Namely, did Congress satisfy the Pennhurst dear-statement rule by “clearly” describing “the conditions that go along with the acceptance of ... funds” under the Act? Arlington Cent. Sch. Dist. Bd. of Educ. v. Murphy, 548 U.S. 291, 304, 126 S.Ct. 2455, 165 L.Ed.2d 526 (2006). We know that States need not invariably be parties to Penn-hurst clear-statement cases because there are many decisions from the Supreme Court involving just a local government as a party, but not a State or even the Federal Government, including one decided a few months ago. See, e.g., Forest Grove Sch. Dist. v. T.A., — U.S. —, 129 S.Ct. 2484, 174 L.Ed.2d 168 (2009); Winkelman *282v. Parma City Sch. Dist., 550 U.S. 516, 127 S.Ct. 1994, 167 L.Ed.2d 904 (2007); Arlington Cent. Sch. Dist., 548 U.S. 291, 126 S.Ct. 2455, 165 L.Ed.2d 526; Jackson v. Birmingham Bd. of Educ., 544 U.S. 167, 125 S.Ct. 1497, 161 L.Ed.2d 361 (2005); Barnes v. Gorman, 536 U.S. 181, 122 S.Ct. 2097, 153 L.Ed.2d 230 (2002); Davis v. Monroe County Bd. of Educ., 526 U.S. 629, 119 S.Ct. 1661, 143 L.Ed.2d 839 (1999); Gebser v. Lago Vista Indep. Sch. Dist., 524 U.S. 274, 118 S.Ct. 1989, 141 L.Ed.2d 277 (1998); Franklin v. Gwinnett County Pub. Schs., 503 U.S. 60, 112 S.Ct. 1028, 117 L.Ed.2d 208 (1992).

The facts of this case illustrate why that often will be so. Whichever side Michigan, Texas or Vermont ultimately take on this discrete legal question, there is no reason to think that the present parties will not protect it. Perhaps the States (silently) agree with the policy underlying the Act— that the threat of reduced federal funds is precisely the incentive that the States and school districts need to make “adequate yearly progress” in the achievement tests required under the Act. If so, there is no reason to think that the Secretary will not ably advance that position, and in fact he (and she) already has done so in the five years of litigation. Or perhaps the States (silently) side with the school districts, believing that they should not have to meet achievement test standards that they cannot reach with federal funds alone. Here, too, the school districts have ably advanced this position throughout this litigation, and they have ample incentives to continue doing so. In the end, the Pennhurst question at the heart of this case turns on an issue of statutory meaning, one that will not change — it cannot change — based on equitable or other considerations that a State might or might not choose to raise.

In the absence of any unprotected interests, no prejudice — -not even a risk of prejudice — exists. What we have instead is a frustrating reality: How could the three States, all deeply involved in the implementation of the Act, not take a public stance on how this significant piece of legislation should be construed? Whether as intervening parties or as amici, the three States would have done well to offer their views. Yet whatever the explanation may be for their resounding silence, Rule 19(a) does not kick in whenever an entity should take a public stand in litigation; it applies only when its absence irreparably prejudices the entity’s interests. That simply is not the case here.

The last ground for invoking Rule 19(a)(1) — subjecting an existing party to multiple or otherwise inconsistent obligations — also is missing. Inconsistent obligations arise only when a party cannot simultaneously comply with the orders of different courts. See Delgado v. Plaza Las Americas, Inc., 139 F.3d 1, 3 (1st Cir.1998); cf. PaineWebber, Inc. v. Cohen, 276 F.3d 197, 201 (6th Cir.2001). Here, the Secretary risks only a possible injunction prohibiting him from withholding federal funds if the affected school districts cannot comply with the Act due to inadequate funding. He can readily comply with this order and any subsequent order in cases involving Michigan, Texas and Vermont.

One of Judge McKeague’s primary disagreements with me turns on whether we can grant “complete relief’ without the States. McKeague Op. at 302-03. Because the districts must follow the Act and the existing plans of their respective States, he points out, those plans would have to be amended for the school districts to obtain the relief they seek. But Rule 19 does not turn on the relief that the claimants could have sought but on the relief they did seek. “Rule 19 can be utilized only to bring in parties necessary to a *283complete and just adjudication of the issues presently before the court.” Ross. v. Houston Indep. Sch. Dist., 699 F.2d 218, 230 (5th Cir.1983) (emphasis added); see also Local 670, United Rubber, Cork, Linoleum, and Plastic Workers of Am., AFL-CIO v. Armstrong Rubber Co., 822 F.2d 613, 620 (6th Cir.1987); 7 Charles Alan Wright, Arthur R. Miller & Mary Kay Kane, Federal Practice & Procedure § 1604 (3d ed.2009). The claimants do not seek to be relieved from a specific mandate of the Act but only to enjoin the Secretary from withholding funds. That relief is not contingent on actions of the States, and it is relief we have the power to grant. See LaChemise Lacoste v. General Mills, Inc., 487 F.2d 312, 314 (3d Cir.1973); cf. 4 James Wm. Moore et al., Moore’s Federal Practice, § 19.03[2][d] (3d ed. 2009) (“Joinder should not be compelled when meaningful relief can be granted without the absentee.”).

That the three States retain independent authority to withhold funds “[i]n order to enforce the Federal requirements” of the Act, 20 U.S.C. § 1232c(b), or perhaps impose some other form of sanction, does not change things. Rule 19(a)(1)(A) “focuses on relief between the parties and not on the speculative possibility of further litigation” or administrative proceedings “between a party and an absent person.” Sales v. Marshall, 873 F.2d 115, 122 (6th Cir.1989) (internal quotation marks omitted). The States have not threatened or even hinted that they would invoke their § 1232c(b) authority if the school districts won this case, and there is no indication that a State has ever invoked that authority, even though it applies to a wide variety of education programs. Under these circumstances, indeed, it is quite likely that the school districts would not have had standing to name the States as party defendants to enjoin them from this speculative, ill-defined and distant threat. Surely Rule 19(a)(1) does not require what Article III prevents.

II.

That takes me to the merits — the meaning of the No Child Left Behind Act and the nature of the obligation that the school districts and the States undertook when they accepted federal funds under the Act starting in 2002. As the school districts see it, “states and school districts that accept NCLB funding are not required to use their own funds for NCLB compliance,” Pontiac Supp. Br. at 12, and accordingly any “failure to comply with the NCLB mandates for this reason does not provide a basis for withholding any federal funds to which they are otherwise entitled under the NCLB,” Compl. at 58. As the Secretary sees it, “a State’s obligation to implement its plan is not contingent upon a particular appropriation of federal funds or capped at a particular level of state expenditures.” Final Br. for the Appellee at 17.

A.

A few rules set the stage for deciding who is right. Congress passed the Act under the Spending Clause. U.S. Const, art. 1, § 8, cl. 1. Congress’s spending authority permits it to condition the allocation of federal funds to the States on their compliance with federal regulations, including most notably regulations that Congress otherwise lacks the power to impose. See Dole, 483 U.S. at 207, 107 S.Ct. 2793. Put another way, the Tenth Amendment, the Eleventh Amendment and the Constitution’s other structural limitations on congressional authority do not limit properly enacted spending-clause legislation.

Yet other limitations constrain Congress’s spending authority: two constitutional limits and a statutory one. As a *284constitutional matter, Congress may not impose conditions “unrelated to the federal interest” in enacting spending legislation, id. at 207-08, 107 S.Ct. 2793 (internal quotation marks omitted), and it may not “coere[e]” the States into accepting funds and the regulations that come with them, id. at 211, 107 S.Ct. 2793. These restrictions, however, need not detain us here. Surely there is a legitimate connection between the Act’s funding and the conditions imposed on the States who accept it. Congress did not give the States federal money for education, then insist that they move the location of their capitals or rename their state birds. Congress asked them to meet a series of educational requirements in return for receiving education funding.

Perhaps more plausibly, the school districts’ complaint could be read to include a claim that the Act is unconstitutionally “coercive,” a choice-bending contract of adhesion. After all, what State in these fiscally challenging times would have the fortitude to turn down hundreds of millions of dollars in education funding? (Perhaps suggesting that there is something to the point, no State refused aid under the Act, notwithstanding the conditions that came with it.) But in their briefs in the district court and on appeal, the school districts have not claimed that they were coerced into accepting this bargain. Because they have not developed this claim in any way and because they have trained their arguments not on invalidating the Act but on limiting their responsibilities under it, they have forfeited any claim of unconstitutionality.

The statutory limitation on Congress’s spending power, by contrast, lies at the core of this dispute. Given the breadth of Congress’s power to impose conditions on States that accept federal money, Arlington Cent. Sch. Dist., 548 U.S. at 296, 126 S.Ct. 2455, given its authority under the Spending Clause to regulate the States beyond the limited and enumerated powers the Constitution otherwise gives it and given that the States are not represented in the Halls of Congress, the federal courts have required Congress to state those conditions “unambiguously” in the text of the statute, Pennhurst State Sch. & Hosp. v. Halderman, 451 U.S. 1, 17, 101 S.Ct. 1531, 67 L.Ed.2d 694 (1981). Analogizing spending clause legislation to a contract that “requires offer and acceptance of its terms,” Barnes v. Gorman, 536 U.S. 181, 186, 122 S.Ct. 2097, 153 L.Ed.2d 230 (2002), the Court has explained that Congress’s authority to impose conditions on a State “rests on whether the State voluntarily and knowingly accepts the terms of the ‘contract,’ ” Pennhurst, 451 U.S. at 17, 101 S.Ct. 1531. Just as parties to a contract “cannot knowingly accept conditions ... which they are unable to ascertain,” neither can the States. Arlington Cent. Sch. Dist., 548 U.S. at 296, 126 S.Ct. 2455 (internal quotation marks omitted). Spending clause conditions thus bind the States only when Congress spells them out clearly in the text of the law.

Even though this dear-statement rule has constitutional roots, it remains a rule of statutory interpretation, one constrained by other canons of statutory interpretation. A State or a school district cannot escape a federal regulation merely by showing possible ways in which a law may be unclear; it must identify a plausible alternative interpretation of the law consistent with its theory of ambiguity. See Bell v. New Jersey, 461 U.S. 773, 783 n. 8, 103 S.Ct. 2187, 76 L.Ed.2d 312 (1983) (rejecting a reading of a statute as “no more than remotely plausible” in favor of a better reading of the law even though it imposed additional obligations on the States); Bennett v. Ky. Dep’t of Educ., 470 U.S. 656, 672, 105 S.Ct. 1544, 84 L.Ed.2d *285590 (1985) (finding no ambiguity in a provision of the predecessor of the NCLB, the Elementary and Secondary Education Act of 1965, Pub.L. No. 89-10, 79 Stat. 27, because “[n]o plausible reading of the statute or regulations suggests that” Kentucky’s actions comported with the statute). A State cannot tenably complain about a congressional bait and switch when the alleged “bait” is premised on an implausible reading of the statute. As is true in other areas of the law, the implausibility of an alternative interpretation of a statute defeats a claim of threshold ambiguity. See Auto-Owners Ins. Co. v. Redland, Ins. Co., 549 F.3d 1043, 1047 (6th Cir.2008) (insurance contract); Zirnhelt v. Michigan Cons’l Gas Co., 526 F.3d 282, 287 (6th Cir.2008) (ERISA plan); cf. Cuomo v. Clearing House Ass’n, L.L.C., — U.S. —, 129 S.Ct. 2710, 2715, 174 L.Ed.2d 464 (2009) (Chevron).

In gauging statutory ambiguity, the courts also apply a wide-angle, not a telephoto, lens. What matters is not whether a provision is ambiguous when read in isolation but whether it is ambiguous when read in context. See, e.g., Pennhurst, 451 U.S. at 19, 101 S.Ct. 1531 (Spending Clause); FDA v. Brown & Williamson Tobacco Corp., 529 U.S. 120, 132, 120 S.Ct. 1291, 146 L.Ed.2d 121 (2000) (Chevron); Prater v. Ohio Educ. Ass’n, 505 F.3d 437, 441 (6th Cir.2007) (contract).

B.

Measured by these yardsticks, the No Child Left Behind Act clearly requires the States (and school districts) to comply with its requirements, whether doing so requires the expenditure of state and local funds or not. A contrary interpretation is implausible and fails to account for, and effectively eviscerates, numerous components of the Act.

The basic bargain underlying the Act, works like this. On the federal side, Congress offers to allocate substantial funds to the States on an annual basis — nearly $14 billion in 2008 for Title I, Part A, a 60% increase in releyant federal funding since 2001 — exercising relatively little oversight over how the funds are spent. On the State side, the States agree to test all of their students on a variety of subjects and to hold themselves and their schools responsible for making adequate yearly progress in the test scores of all students. In broad brush strokes, the Act thus allocates substantial federal funds to the States and school districts and gives them substantial flexibility in deciding how and where to spend the money on various educational “inputs,” but in return the schools must achieve progress in meeting certain educational “outputs” as measured by the Act’s testing benchmarks. As the Supreme Court recently explained:

NCLB marked a dramatic shift in federal educational policy. It reflects Congress’ judgment that the best way to raise the level of education nationwide is by granting state and local officials flexibility to develop and implement educational programs that address local needs, while holding them accountable for the results. NCLB implements this approach by requiring States receiving federal funds to define performance standards and to make regular assessments of progress toward the attainment of those standards. 20 U.S.C. § 6311(b)(2). NCLB conditions the continued receipt of funds on demonstrations of “adequate yearly progress.” Ibid.

Horne v. Flores, — U.S. —, 129 S.Ct. 2579, 2601, 174 L.Ed.2d 406 (2009). The school districts’ position-that they can accept the federal dollars, spend them largely as they wish, yet exempt themselves from the Act’s requirements if compliance *286.would require any local money — undoes this bargain by nullifying some provisions of the Act and undermining several others.

Accountability. Accountability is the centerpiece of the Act, and a plausible interpretation of the legislation cannot ignore that reality. Instead of focusing on how much money school districts spend on each child or “dictating funding levels,” the Act “focuses on the demonstrated progress of students through accountability reforms.” Id. at 2603. The Act begins with a “Statement of Purpose” that drives home Congress’s interest in establishing accountable public schools: “ensuring ... high-quality academic assessments [and] accountability systems”; “holding schools, local education agencies, and States accountable for improving the academic achievement of all students”; “improving and strengthening accountability”; and “providing ... greater responsibility for student performance.” 20 U.S.C. §§ 6301(1), (4), (6), (7). See Appendix (containing the full Statement of Purpose).

Title I, Part A of the Act carries out this objective by requiring participating States to test their students and, over time, to establish and meet certain benchmarks in doing so. Today, the Act requires all public school students in participating States to take seventeen standardized tests over the course of their school careers, see id. § 6311(b)(3)(C), and requires the States to grade schools and school districts on their ability to make “adequate yearly progress” in the test results, see id. §§ 6311(b)(2)(B), 6316(c)(1). For those participating schools that repeatedly fail to make progress, the Act requires an escalating series of sanctions: (1) starting with “improvement,” which gives students the right to transfer into more successful schools and forces schools to develop better practices; (2) moving to “corrective action,” which may include a new curriculum, extended hours, or personnel discharges; and (3) ending with “restructuring,” which may include discharging large portions of the staff or converting the school into a charter school. Id. § 6316(b); see also id. § 6316(c).

The Act provides limited exceptions to these accountability measures, and none of them applies here. See, e.g., id. §§ 6311(b)(7), 6311(c)(1), 6311(h)(2)(A)(i), 6316(b)(7)(D), 6316(c)(10)(F), 6337(e)(3). By cabining the Secretary’s discretion to excuse failure, the Act furthers an essential objective — the source, indeed, of its title — that no child, whether living in inner-city school districts or not, whether suffering from learning disabilities or not, whether English is their second language or not, whether otherwise disadvantaged or not, would be left behind when it came to ensuring not just that more resources were devoted to their education but that objectively measurable progress would be made in their education. See, e.g., id. § 6301(3) (setting goal of “closing the achievement gap between high- and low-performing children, especially the achievement gaps between minority and nonminority students, and between disadvantaged children and their more advantaged peers”). That is why the Act requires each State to set up an annual system of academic assessments in reading, science and math, id. § 6311(b)(3)(A), using “the same academic assessments ... to measure the achievement of all children,” id. § 6311(b)(3)(C)(i) (emphasis added), providing for “the participation in such assessments of all students,” id. § 6311(b)(3)(C)(ix)(I) (emphasis added), and requiring “each local educational agency” to collect appropriate data and reports on all student achievement, id. at § 6311(h)(2)(B) (emphasis added).

The school districts’ interpretation would break the accountability backbone of the Act. Excusing school districts from *287compliance with the Act whenever federal funding fell short would make it hard if not impossible to hold them accountable for meeting the Act’s goals. If school districts decided they were not given enough money to test all children, they could test just some children. If school districts decided they were not given enough money to fix all underperforming schools, they could fix just some schools. Because the school districts have alleged that virtually every major requirement of the Act is underfunded, see Compl. ¶¶ 32-86, their interpretation would excuse them from all of these requirements, transforming a no-exceptions accountability system into a non-existent one.

The NAACP supports the Secretary in this case, as well as in the Second Circuit case, not because it is satisfied with the levels of federal funding under the Act (few are) but because it does not want the Act’s accountability measures violated with impunity — letting schools filled with disadvantaged students off the hook. NAACP Br. at 12 (“The panel’s decision invites States and school districts to evade their obligations to poor and minority children.”). The NAACP’s concern is reflected in the Act itself, which begins by saying that the Act is designed “to ensure that all children ... reach ... proficiency on challenging State academic achievement standards and State academic assessments,” “especially ... disadvantaged” students. 20 U.S.C. § 6801 (emphasis added). With the Act’s accountability system in place, these goals have a chance of success; without them, they have no chance of success, risking a return to (or a continuation of) a system of lower standards for higher-poverty schools.

Flexibility. The school districts’ interpretation is inconsistent not only with the Act’s accountability requirements but also with the flexibility the Act gives States and school districts in return for increased responsibility for student achievement. As the Act’s Statement of Purpose makes clear, that is the central tradeoff of the Act: “providing greater decisionmaking authority and flexibility to schools and teachers in exchange for greater responsibility for student performance.” id. § 6301(7) (emphasis added); see also Horne, 129 S.Ct. at 2601 (the Act “reflects Congress’ judgment that the best way to raise the level of education nationwide is by granting state and local officials flexibility to develop and implement educational programs that address local needs, while holding them accountable for the results”). Unlike most spending programs, this one comes with few strings telling the States how they should comply with its conditions. Under the Act, States develop their own curricula and standards, 20 U.S.C. § 6311(b)(1), their own tests to assess whether students are meeting those standards, id. § 6311(b)(3), and their own definitions of progress under those standards, id. § 6311(b)(2)(B), so long as the progress culminates in near-universal proficiency by 2014, id. § 6311(b)(2)(F).

This flexibility extends to spending as well. As the school districts rightly acknowledge, the Act “provide[s] school districts with unprecedented new flexibility in their allocation of Title I funds.” Final Reply Br. of Pontiac Sch. Dist. at 3 (internal quotation marks omitted). Some federal funds, to be sure, must be spent in certain ways. See, e.g., 20 U.S.C. § 6303 (reserving some Title I, Part A funds for school improvement); id. § 6317(c)(1) (same); id. § 6318(a)(3)(A) (reserving some funds for parental involvement programs); id. § 6319(1) (reserving some funds for professional development). And the Act strictly confines the use of Title I funds to geographic areas with heavy concentrations of low-income students. See id. § 6313(a). But within these areas and *288with respect to these priority students, the Act gives States and school districts substantial flexibility in choosing how to spend the money. For instance: Section 6314 gives school districts wide discretion to consolidate funds from various sources and to focus them on certain schools in whatever ways will improve student performance there; § 6313(b) gives school districts discretion to transfer funds between schools within certain guidelines; and § 7305b allows States and school districts to transfer up to 50% of the funds allotted to other education programs to supplement their funds under Title I, Part A.

The substantial flexibility the Act gives recipients over federal funds is surpassed by the near-complete flexibility they retain over their own funds. The only limitation is that participating States cannot reduce their own spending and offset it with federal funding but must use the Act’s federal dollars to supplement, not supplant, their own. 20 U.S.C. §§ 6321, 7901. Beyond that basic requirement — a prohibition on fiscal cheating, really — the States can use their dollars however they see fit, whether for teachers or for computers or for facilities or for whatever else they think will help their students the most.

The express and unprecedented flexibility the Act gives to the States in prioritizing the spending of federal dollars — especially in Title I, Part A — cannot co-exist with an interpretation of the statute that allows school districts to exempt themselves from the accountability side of the bargain whenever their spending choices do not generate the requisite achievement. Were the school districts correct, a State could use this flexibility to focus its federal and local resources almost exclusively on improving, say, teacher quality — a legitimate goal no doubt, but one that would allow the State to sidestep the Act’s mandatory assessment requirements by contending that it lacked the funds to administer them or to make progress under them. Sch. Dist. of City of Pontiac v. Sec’y of United States Dep’t of Educ., 512 F.3d 252, 284 (6th Cir.2008) (McKeague, J. dissenting). That is not what Congress had in mind. It gave the States a clear and consequential choice: between taking the bitter (accountability) with the sweet (unprecedented flexibility in spending federal and state dollars) or leaving the money on the table.

Costs of Compliance. Not surprisingly, in view of the expansive flexibility that the Act gives States in spending federal and local funds, the Act says nothing about the bill of particulars at the heart of the school districts’ complaint: the costs of complying with the Act’s requirements. How could it be otherwise? The Act’s spending flexibility necessarily makes it impossible to calculate or even define the costs of complying with the Act’s requirements.

The primary formula for allocating Title I, Part A grant money does not say a word about costs of compliance. See 20 U.S.C. §§ 6313(c), 6333(a), 6334(a), 6335(a)-(c), 6337. While the Act asks States to submit plans to the Secretary, id. § 6311, and asks school districts to submit plans to the States, id. § 6312, it does not require either entity to estimate the cost of compliance. Nor, in fulfilling their various reporting responsibilities under the Act, must the States or school districts estimate the costs of compliance. See, e.g., id. §§ 6311(h), 6316(a)(1)(C). If, as the Supreme Court recently explained, the Act “expressly refrains from dictating funding levels,” Horne, 129 S.Ct. at 2603, why would Congress exempt failing school districts from the accountability requirements based on inadequate “funding levels”? The school districts have no answer.

But even if Congress wished to make costs of compliance a legitimate excuse for, *289say, inadequate yearly progress, how would it do so? Once Congress decided to measure accountability by educational outputs (gauged by tests scores), as opposed to educational inputs (gauged by dollars), it made objective measurements of compliance costs virtually impossible. Any effort to measure these costs surely would vary from school to school, if" not from student to student, and they surely would vary from year to year. The phrase “costs of compliance” has no discernible meaning in this context, as the Act leaves it to the States, no matter how little or how much funding Congress provides, to make discretionary cost choices about how to make meaningful achievement-related progress.

Take a cost estimate for adding an extra hour to the school day, for lengthening the school year or for hiring more math or reading teachers — all plausible ways to improve a school’s achievement scores. Each innovation has an estimable cost, to be sure. But that does not establish that the estimate would lead to the requisite progress. And if it did not, then what? Perhaps extending the school day by one more hour, extending the school year by one more week or hiring one more math or reading teacher would do the trick. But maybe not. What works for one school district might not work for another. What, indeed, works for one classroom might not work for the classroom next door, given the correlation between great teachers and great teaching — and the occasional operation of that principle in reverse. Even more discrete costs like developing and administering tests cannot be accounted for in advance given the considerable flexibility States have under the Act in implementing those requirements. Within certain general limits, a State may develop whatever curricular standards and tests it wants. 20 U.S.C. § 6311(b). The State may use pre-existing standards that meet the Act’s requirements, id. § 6311(b)(1)(F), or it may create new ones.

In their complaint, to use one example, the school districts say that Brandon Town School District “estimates that ... it needed to spend $390,000 more than it received in NCLB Title I funding to ensure that the school makes [adequate yearly progress].” Compl. ¶ 65. The school district may be right, and we have no license to say that it is not at this Rule 12(b)(6) stage of the case. The issue, however, is not whether the school districts can fairly say that compliance with “adequate yearly progress” requires more federal dollars than the Secretary has allocated to them. It is whether a State could tenably think that the Act excuses non-compliance whenever a school district maintains that it has insufficient resources to make the required progress. Surely every school district could do more with more money. And if that is the case, every failing school district could do more with more federal money — and maybe enough to make adequate yearly progress. It is hard to imagine when — or, for that matter, why — -a failing school would ever concede that it was getting sufficient federal funds to make such progress.

“Reflecting a growing consensus in education research that increased funding alone does not improve student achievement,” the Act moves from a dollars-and-cents approach to education policy to a results-based approach that allows local schools to use substantial additional federal dollars as they see fit in tackling local educational challenges in return for meeting improvement benchmarks. Horne, 129 S.Ct. at 2603 & n. 17. The Act, in short, rejects a money-over-all approach to education policy, making it implausible that the heartland accountability measures of the law could be excused whenever schools, exercising their flexibility over *290how to spend federal and local dollars, decided they cost too much.

Express waiver authority in some areas and silence in others. Congress knew how to allow the Secretary to waive obligations under the Act, and it did so in discrete circumstances. The Act’s general waiver provision allows States and school districts to seek waivers on just two grounds: that the waiver will “(i) increase the quality of instruction for students; and (ii) improve the academic achievement of students.” 20 U.S.C. § 7861(b)(1)(B). Neither ground excuses schools from the accountability measures due to insufficient federal funds.

The Act also specifically describes two types of exceptions from the § 6311 accountability requirements. One applies only in case of “exceptional or uncontrollable circumstances, such as a natural disaster or a precipitous and unforeseen decline in ... financial resources.” Id. §§ 6311 (b)(3)(C)(vii), 6311(b)(7), 6311(c)(1), 6311(h)(2)(A)(i). Not only does neither exemption turn on the sufficiency of federal funding, but one of them — the “unforeseen decline in ... financial resources” — creates an exemption for sharp declines in local funding, which implies that the Act contemplates local spending.

The other type of exception excuses compliance if federal funding is not sufficient, but again only in limited circumstances. The Act, for instance, allows States to “suspend the administration of, but not cease the development of,” annual tests if federal funding falls below certain levels. Id. § 6311(b)(3)(D). Section 6311(c)(2) requires States to “participate in biennial state academic assessments of 4th and 8th grade reading and mathematics under the National Assessment of Educational Progress,” but only if “the Secretary pays the costs of administering such assessments.” Under the school districts’ interpretation, that final qualification is unnecessary and indeed pointless. See Astoria Fed. Sav. & Loan Ass’n v. Solimino, 501 U.S. 104, 112, 111 S.Ct. 2166, 115 L.Ed.2d 96 (1991) (“[W]e construe statutes, where possible, so as to avoid rendering superfluous any parts thereof.”). Congress likewise capped a school district’s costs for supplemental educational services at the amount of federal money allocated. Id. § 6316(e)(6). As each of these examples show, Congress knew how to excuse schools from compliance based on inadequate funding, but did so only in discrete circumstances — none of them applicable here. See Hamdan v. Rumsfeld, 548 U.S. 557, 578, 126 S.Ct. 2749, 165 L.Ed.2d 723 (2006) (“A familiar principle of statutory construction ... is that a negative inference may be drawn from the exclusion of language from one statutory provision that is included in other provisions of the same statute.”).

Inconsistency between the school districts’ interpretation and other provisions of the Act. Besides conflicting with the hallmark features of the Act and ignoring the implications and inferences that follow from Congress’s express waiver provisions in some circumstances but not in others, the school districts’ interpretation conflicts with, or is at least in tension with, other provisions of the Act. The Act explicitly anticipates that funding to meet the Act’s requirements will come from a variety of sources, not all federal. It requires the Secretary, for instance, to examine how States, school districts and schools have used “Federal, State, and local educational agency funds and resources to support schools and provide technical assistance to improve the achievement of students in low-performing schools.” 20 U.S.C. § 6491(a)(2)(E)(iv) (emphases added); see also id. § 6491(a)(2)(E)(v) (requiring a similar inquiry to determine how they have *291“used State educational agency and local educational funds to help [certain] schools ... meet the requirements described in § 6319 of this title of having all teachers highly qualified not later than the end of the 2005-2006 school year”). This provision contradicts the notion that States and school districts need only meet the Act’s requirements to the extent federal funds can do the trick.

Some of the Act’s accountability measures apply to every school in a participating State regardless of whether the school receives any federal funding at all. See, e.g., id. § 6311(b)(2). Other requirements apply to every school in a school district receiving funds, whether or not the school itself receives funds. See id. § 6319. Yet these universal accountability requirements cannot be squared with an interpretation of the Act that demands accountability only to the extent a school receives federal funding. See Pontiac, 512 F.3d at 274-75 (McKeague, J., dissenting). Otherwise, an unfunded school district would have no accountability responsibilities under the Act, a notion the Act expressly contradicts.

In passing the Act, Congress also knew how to allocate funds for specific purposes — in the nature of traditional input-based funding programs. Some provisions of the Act tell the States exactly how to “use the [federal] funds.” See, e.g., 20 U.S.C. §§ 6313, 6362(c)(7)(A), 6372, 6381c, 6383, 6393, 7114. But these provisions make all the more conspicuous the contrast with other provisions of the Act that say nothing about how to spend the money but mention only the accountability benchmarks the States must achieve in using it. See, e.g., id. §§ 6311(a), 6316(a). Congress showed that it knew how to write input-based provisions, limiting the States’ responsibility to spend certain funds on discrete items, and output-based provisions, allowing the States to spend the money however they wished so long as they achieved the federal benchmarks. We should respect the reality that Congress knew how to distinguish between the two. See BP America Prod. Co. v. Burton, 549 U.S. 84, 92, 127 S.Ct. 638, 166 L.Ed.2d 494 (2006).

Ongoing implementation of the Act. If for some reason the States or school districts had any doubt about the nature of the bargain they were undertaking when they first accepted federal funds under the Act, time has cleared things up. Since 2002, when it passed the Act, Congress has made annual appropriations to the States, and each year the appropriations have not been linked to, or premised on, any effort to ascertain the funds needed to make adequate yearly progress. See Compl. ¶¶ 25-31. Yet each year the Department of Education has not wavered: The Secretary consistently has denied attempts to evade the Act’s requirements due to insufficient federal funding. By the time the school districts filed this lawsuit in 2005, they plainly were on notice that there was no linkage between the appropriated federal funds and the States’ duty to comply with the accountability measures. Notwithstanding that notice, the States continued to participate in the program — continued to accept the spending-legislation offer of funding, as it were, in exchange for the continued obligation to meet the Act’s achievement requirements. See, e.g., Jackson v. Birmingham Bd. of Educ., 544 U.S. 167, 183, 125 S.Ct. 1497, 161 L.Ed.2d 361 (2005) (States had sufficient notice of their responsibility under Title IX because the regulations had “been on the books” for some time); Davis v. Monroe County Bd. of Educ., 526 U.S. 629, 643, 119 S.Ct. 1661, 143 L.Ed.2d 839 (1999) (same); cf. Bennett, 470 U.S. at 669, 105 S.Ct. 1544 (“the fact that Title I [of the Elementary and Secondary Education Act of 1965] was *292an ongoing, cooperative program meant that grant recipients had an opportunity to seek clarification of the program requirements”); Brown & Williamson Tobacco, 529 U.S. at 143, 120 S.Ct. 1291 (noting in the context of Chevron that “[o]ver time, ... subsequent acts can shape or focus” the initial “range of plausible meanings” that a statute may have). The Act does not permit the school districts to accept one part of the bargain and discard the other, least of all when the passage of time confirms the two sides of the bargain.

C.

Resisting this conclusion, the school districts argue that 20 U.S.C. § 7907(a), a provision that arrives on page 559 of this 674-page Act, 115 Stat. 1425, 1983, changes everything. Appearing in a section dealing with general rules under the Act, the provision reads in full:

GENERAL PROHIBITION. — Nothing in this Act shall be construed to authorize an officer or employee of the Federal Government to mandate, direct, or control a State, local educational agency, or school’s curriculum, program of instruction, or allocation of State or local resources, or mandate a State or any subdivision thereof to spend any funds or incur any costs not paid for under this Act.

20 U.S.C. § 7907(a). According to the school districts, the provision means that the Federal Government may not require the States (or school districts) to spend their own money to meet the Act’s requirements — that if the appropriated federal funds do not suffice to meet the accountability measures, the States are free to ignore them. That is not true — for several reasons.

First, § 7907(a) by its terms is a rule of construction, which explains how other sections of the Act should not be “construed.” What the school districts urge, however, is something different — to “construe” § 7907(a) itself to create a no-unfunded-mandate exception, to use § 7907(a) to lift § 7907(a) into a sweeping exception to the Act. No one boot straps a boot strap.

Section 7907(a)’s rule of construction has no job to do here. The Act’s express trade-off between local flexibility to spend federal funds and local responsibility to obtain output-based progress in doing so has no ambiguity to speak of and thus no ambiguity to be “construed.” There is nothing unclear, and no shortage of detail in the 674-page piece of legislation, about this hallmark of the Act. There is no other provision of the Act — at least as far as this dispute is concerned — that calls out for ambiguity-clarifying “constru[ction].”

Second, text is context, and a reading of § 7907(a) in its immediate surroundings confirms the modest role it plays. The section merely re-enforces the flexibility that the Act gives to school districts in developing them own local programs and spending their own funds in tackling local education matters. It functions as an anti-commandeering rule of construction, nothing more. As such, it prevents the Secretary from construing the Act to “mandate” States or school districts (1) to adopt a particular “curriculum” or “program of instruction”; (2) to “allocate]” state or school district resources in a particular way; or (3) “to spend any funds or incur any costs not paid for under this Act.” All three limitations parallel the Act’s theme of local flexibility — that the States and school districts must be permitted to make their own decisions about how to spend local resources in return for taking on the duty to make progress in a results-driven way. Thus, while the Act requires the schools to make adequate yearly progress, it does not tell them how to do so or mandate the spending of any local money *293to do so. If satisfying the Act’s requirements takes additional money, so be it, but the Act is equally satisfied whether additional costs are incurred or not. Section 7907(a) simply fortifies the Act’s focus on results over spending levels, on outputs over inputs. See Horne, 129 S.Ct. at 2603 (the Act “expressly refrains from dictating funding levels”).

Judge Cole notes that the Federal Government cannot point to “any provision” of the Act “that explicitly spells out the States’ [fiscal] obligations under this Act.” Cole Op. 26. That is right, but it is consistent with this feature of the Act. The Act does not tell States to spend; it tells them to do. It does not spell out fiscal obligations; it spells out performance obligations, reporting obligations, parental-involvement obligations, teacher-qualification obligations — all of which the Act makes perfectly clear through hundreds of pages of statutory text. No doubt, these performance obligations cost money, but the Act in general — and § 7907(a) in particular — leave it to the States to make these spending and allocation choices for themselves.

Third, as signaled by § 7907(a)’s appearance in a “general” set of provisions, it does not trump specific directives found elsewhere in the Act. “[I]t is a commonplace of statutory construction that the specific governs the general,” Morales v. Trans World Airlines, Inc., 504 U.S. 374, 384, 112 S.Ct. 2031, 119 L.Ed.2d 157 (1992), and accordingly the school districts cannot leverage § 7907(a)’s “general” rule of construction to destroy the specific bargain offered to the States: the exchange of federal funds with flexibility to spend them in return for a no-excuses commitment to meeting the Act’s accountability requirements. The school districts’ contrary argument not only ignores this “commonplace” rule of construction but it also undermines other provisions of the Act. To ensure that the Act does not lead local schools to devote fewer local resources to education, the Act requires each participating State to preserve a level of local effort — no less than 90% of the prior year’s expenditures. 20 U.S.C. §§ 6321, 7901. But if we read § 7907(a) in isolation in the manner that the school districts urge, it would bar this requirement as well — as a violation of the prohibition against mandating States to “spend any funds ... not paid for under this” Act. The school districts correctly acknowledge that § 7907(a)’s general rule of construction does not overrule this specific provision, Pontiac Supp. Br. at 20, but they persist in claiming that it overrules the specific bargain at the heart of the Act. They cannot have it both ways. The specific governs the general across the board, not just in some places as opposed to others.

Pennhurst, the school districts’ featured case, embraced this form of analysis, paying attention to the difference between general and specific statutory provisions. The statute at hand provided that “[p]er-sons with developmental disabilities have a right to appropriate treatment, services, and habilitation for such disabilities,” 42 U.S.C. § 6010(1) (current version at 42 U.S.C. § 15009(a)(1)), and the question was whether this language imposed mandatory duties or hortatory goals. After comparing the general language of § 6010 to other provisions that imposed express conditions on federal funding, the Court concluded that “[t]he existence of explicit conditions throughout the Act, and the absence of conditional language in § 6010, manifest the limited meaning of § 6010.” Pennhurst, 451 U.S. at 23, 101 S.Ct. 1531. “[A] brief comparison of the general language of § 6010 with the conditions Congress explicitly imposed on the States,” the Court reasoned, “demonstrates that Congress did not intend to place either abso*294lute or conditional obligations on the States.” Id. at 25, 101 S.Ct. 1531. The same principle applies here. What was sauce for the States in Pennhurst is sauce for the school districts here.

Fourth, § 7907 as a whole supports this interpretation. Labeled “Prohibitions on Federal Government and Use of Federal Funds,” the section contains other provisions that all limit federal officials in imposing more conditions on the States than those mentioned in the Act. The other subsections prohibit federal officials from requiring any particular curricula, academic standards or building standards. 20 U.S.C. § 7907(b)-(d). Because words are “known by the company [they] keep[],” Gustafson v. Alloyd Co., 513 U.S. 561, 575, 115 S.Ct. 1061, 131 L.Ed.2d 1 (1995), these statutory neighbors reenforce the notion that § 7907(a) is an anti-commandeering rule of construction.

Fifth, it strains credulity to think that Congress, via a single half-sentence 559 pages into the Act, suddenly blinked, changing the hallmark bargain at the core of this legislation. “Congress ... does not alter the fundamental details of a regulatory scheme in vague terms or ancillary provisions.” Whitman v. Am. Trucking Ass’ns, 531 U.S. 457, 468, 121 S.Ct. 903, 149 L.Ed.2d 1 (2001). The National Legislature “does not ... hide elephants in mouseholes,” id., yet that is precisely what the school districts purport to have found hidden in § 7907(a). See also Brown v. Gardner, 513 U.S. 115, 118, 115 S.Ct. 552, 130 L.Ed.2d 462 (1994) (“Ambiguity is a creature not of definitional possibilities but of statutory context.”).

D.

Stray remarks in the legislative history and offhand comments by former Secretary of Education Rod Paige do not alter this conclusion. Various Senators and Representatives made statements for and against the Act, arguing in some places that it would impose increased costs on the States and in some places that it would not. See Pontiac, 512 F.3d at 269-271; see also id. at 282-284 (McKeague, J., dissenting). But that makes no difference. The school districts’ legislative-history-based arguments about the meaning of § 7907(a) are no more plausible than their text-based ones. In either case, they ask us to embrace an interpretation of § 7907(a) that cannot be reconciled with the hallmark features of the Act. Legislative history also cannot alter the outcome in a clear-statement case. The Pennhurst clear-statement rule turns on textual ambiguity, not ambiguity in the legislative history. If there is textual ambiguity and a plausible state-friendly way to read the statute, that ends the matter. But if there is no such ambiguity, there is nothing for the legislative history to clarify. See Arlington, 548 U.S. at 304, 126 S.Ct. 2455 (“In a Spending Clause case, the key is not what a majority of the Members of both Houses intended but what the States are clearly told regarding the conditions that go along with the acceptance of those funds.”); cf. United States v. Nordic Village, Inc., 503 U.S. 30, 37, 112 S.Ct. 1011, 117 L.Ed.2d 181 (1992) (“legislative history has no bearing on the ambiguity point” in applying a clear-statement rule).

The same goes for Secretary Paige’s statements, which no one claims deserve deference, whether under Chevron or any other doctrine. In one speech, he said that the Act “contains language that says that things that are not funded are not required,” Compl. ¶ 15 (quoting Paige speech of Sept. 4, 2003), and in another speech that “if it is not funded, it’s not required. There is language in the bill that prohibits requiring anything that is not paid for,” id. (quoting Paige speech of *295Dec. 2, 2003). These comments, however, conflict with others in which Secretary Paige “repeatedly” emphasized that “[i]f a state decides to accept the federal funds [offered under the Act], then it’s required to implement the law in its entirety.” Compl. ¶ 16 (quoting Paige speech of Mar. 25, 2004). More importantly, these words conflict with the actions of the Department of Education, which has consistently held States accountable under the Act, whether those States are content with the level of federal funding or not. See Compl. ¶ 17-19 (noting the “uniform rejection of requests for waivers from the NCLB mandates based upon a lack of federal funding”). Whatever Secretary Paige meant in his comments nearly six years ago (and it is not clear that they are inconsistent with my reading of § 7907(a)), they are not binding on the Department of Education and cannot create ambiguity where none otherwise exists.

The school districts, lastly, invoke the 1984 Perkins Vocational Education Act, 20 U.S.C. §§ 2301-2471. Their chain of reasoning goes like this: § 7907(a) parallels the first part of § 2306a(a) of the Perkins Act; § 2306a(a), unlike § 7907(a), proceeds to exempt several Perkins Act provisions requiring the expenditure of specific non-federal funds; this exemption implies that these provisions otherwise would conflict with the prohibition; because the exempted provisions require the expenditure of non-federal funds, § 2306a(a) should be read to forbid requiring the expenditure of non-federal funds except as to the specific provisions exempted, and § 7907(a) should be read the same way.

The argument proves too much. Two of the provisions excepted in the Perkins Act — those requiring that States and school districts maintain their financial effort rather than supplanting their previous spending with federal funds, 20 U.S.C. §§ 2391(a), 2413 — have counterparts in No Child Left Behind as well. See id. §§ 6321, 7901. And although § 7907(a) includes no exceptions for these maintenance-of-effort provisions, no one, the school districts included, see Pontiac Supp. Br. at 20, claims that § 7907(a) nullifies the Act’s maintenance-of-effort provisions. See United States v. Atl. Research Corp., 551 U.S. 128, 137, 127 S.Ct. 2331, 168 L.Ed.2d 28 (2007) (statutes should not be read to make any provision “a dead letter”). Better, it seems to me, to read § 7907(a) in the context of No Child Left Behind, not the Perkins Act, which was first enacted 18 years earlier and which adds nothing that supports a plausible alternative interpretation of § 7907(a).

í¡: ^4 Hs H*

Depending on whom you ask, the No Child Left Behind Act might be described in many ways: bold, ground-breaking, noble, naive, oppressive, all of the above and more. But one thing it is not is ambiguous, at least when it comes to the central tradeoff presented to the States: accepting flexibility to spend significant federal funds in return for (largely) unforgiving responsibility to make progress in using them. The theme appears in one way or another in virtually every one of the Statements of Purpose of the Act, and it comes across loud and clear in the remaining 674 pages of legislation. That § 7907(a) suddenly transformed the Act into a no-strings-attached grant program, or for that matter an outright gift program, not only ignores the pages of legislation that precede it, but it also ignores the words of § 7907(a) itself and one of the eternal prerogatives of power: control follows money. San Antonio Indep. Sch. Dist. v. Rodriguez, 411 U.S. 1, 53 n. 109, 93 S.Ct. 1278, 36 L.Ed.2d 16 (1973). Here, unlike prior education funding programs, Congress did not exercise that control by telling the *296schools how to spend the money but by telling them to get results with it. Time will tell whether Congress was wise to move from conditioning federal funds on “adequate” additional local funding to conditioning federal funds on “adequate” local progress. But no state official who read the Act could plausibly think that Congress intended to impose neither condition.

That said, I have considerable sympathy for the school districts, many of whom may well be unable to satisfy the Act’s requirements in the absence of more funding and thus may face the risk of receiving still less funding in the future. Yet two Presidents of different parties have embraced the objectives of the Act and committed themselves to making it work. So have a remarkably diverse group of legislators. If adjustments should be made, there is good reason to think they will be. But, for now, it is hard to say that the judiciary will advance matters by taking the teeth out of the hallmark features of the Act. It is the political branches, not the judiciary, that must make any changes, because the Act’s requirements are clear, making them enforceable upon participating States and their school districts.

III.

For these reasons, I concur in the order affirming the district court’s judgment.

APPENDIX

The purpose of this subchapter is to ensure that all children have a fair, equal, and significant opportunity to obtain a high-quality education and reach, at a minimum, proficiency on challenging State academic achievement standards and state academic assessments. This purpose can be accomplished by—

(1)ensuring that high-quality academic assessments, accountability systems, teacher preparation and training, curriculum, and instructional materials are aligned with challenging State academic standards so that students, teachers, parents, and administrators can measure progress against common expectations for student academic achievement;
(2) meeting the educational needs of low-achieving children in our Nation’s highest-poverty schools, limited English proficient children, migratory children, children with disabilities, Indian children, neglected or delinquent children, and young children in need of reading assistance;
(3) closing the achievement gap between high- and low-performing children, especially the achievement gaps between minority and nonminority students, and between disadvantaged children and their more advantaged peers;
(4) holding schools, local educational agencies, and States accountable for improving the academic achievement of all students, and identifying and turning around low-performing schools that have failed to provide a high-quality education to their students, while providing alternatives to students in such schools to enable the students to receive a high-quality education;
(5) distributing and targeting resources sufficiently to make a difference to local educational agencies and schools where needs are greatest;
(6) improving and strengthening accountability, teaching, and learning by using State assessment systems designed to ensure that students are meeting challenging State academic achievement and content standards and increasing achievement overall, but especially for the disadvantaged;
(7) providing greater decisionmaking authority and flexibility to schools and *297teachers in exchange for greater responsibility for student performance;
(8) providing children an enriched and accelerated educational program, including the use of schoolwide programs or additional services that increase the amount and quality of instructional time;
(9) promoting schoolwide reform and ensuring the access of children to effective, scientifically based instructional strategies and challenging academic content;
(10) significantly elevating the quality of instruction by providing staff in participating schools with substantial opportunities for professional development;
(11) coordinating services under all parts of this subchapter with each other, with other educational services, and, to the extent feasible, with other agencies providing services to youth, children, and families; and
(12) affording parents substantial and meaningful opportunities to participate in the education of their children.

20 U.S.C. § 6301, Pub.L. 89-10, Title I, § 1001.