Opinion for the Court filed by Circuit Judge ROGERS.
Opinion concurring in part and dissenting in part filed by Circuit Judge GRIFFITH.
ROGERS, Circuit Judge.The principal issue in this appeal is whether the Religious Freedom Restoration Act- (“RFRA”), 42 U.S.C. § 2000bb et seq., requires strict scrutiny of a federal agency’s approval of an airport layout plan incident to a determination of eligibility for federal funding if the plan, when implemented by a subdivision of a state, may burden religious exercise. Because O’Hare International Airport, by some measures “the busiest airport in the world,” Suburban O’Hare Comm’n v. Dole, 787 F.2d 186, 196 (7th Cir.1986), has been plagued by delays in recent years, the City of Chicago plans to expand and reconfigure the airport. The petitioners, two Chicago suburbs, a church, and several individuals, challenge the Federal Aviation Administration’s approval of the City’s plan on the ground that the approval violates RFRA because the approved runway configuration, which requires the relocation of a church cemetery, is not the least restrictive means of satisfying the government’s compelling interest in reducing delays.
We hold that any burden on the exercise of religion caused by the City’s airport expansion plan is not fairly attributable to the FAA. The Supreme Court has recognized that even in instances in which the federal government plays some role, constitutional standards do not attach to conduct by third parties in which the federal government merely acquiesces. So too, a federal agency’s determination that a City’s expansion plan is eligible for federal funding does not render the City’s implementation of the plan tantamount to federal action that is the source of the burden on the free exercise of religion. The expansion plan for the airport, which is owned by the City, was prepared and will be implemented by the City, which is prepared to proceed without federal funds if necessary, and RFRA does not apply to burdens imposed by states or their subdivisions. Hence, the court need not reach the question whether the FAA has shown a compelling governmental interest in imposing a burden on the free exercise of religion.
Additionally, the court lacks jurisdiction to consider the petitioners’ challenge to the FAA letter expressing a non-binding intention to obligate federal funding for the expansion because the letter is not a final order. Accordingly, because the petitioners’ remaining contentions are without merit, we deny the petitions for review.
I.
A.
The Airport and Airway Improvement Act, 49 U.S.C. §§ 47101 et seq. (“AAIA”), authorizes federal funding for airport improvement projects and establishes the prerequisites for a project to be eligible for funding. Congress established a national transportation policy aimed at the efficient transportation of passengers and property to ensure “the expanding wealth of the United States, the competitiveness of the industry of the United States, the standard *58of living, and the quality of life.” Id. § 47101(b)(2)49USCAS47101. To those ends, the AAIA requires that “airport construction and improvement projects that increase the capacity of facilities to accommodate passenger and cargo traffic be undertaken to the maximum feasible extent so that safety and efficiency increase and delays decrease.” Id. § 47101(a)(7).
To establish a “safe, efficient, and integrated system of public-use airports,” the Secretary of Transportation must maintain a public airport development plan that includes the “kind and estimated cost of eligible airport development.” Id. § 47103(a). More pertinently, pursuant to a delegation of authority from the Secretary, see id. § 106(g), the FAA may make project grants to a State, public agency, or private owner of a public-use airport from the Airport and Airway Trust Fund for airport development. See id. § 47104(a); see also id. § 47105(a), id. § 47102(19). To be eligible for federal grants, the airport development must comply with standards set by the FAA. See id. § 47105(b)(3). The FAA may approve an application only if satisfied that there are funds to cover costs not paid by the federal government, that the sponsor has authority to carry out the project, and that the project is consistent with state agency plans for the areas surrounding the airport, will contribute to carrying out the AAIA’s purposes, and will be completed without unreasonable delay. See id. § 47106(a).
In addition to the statutory requirements for specific projects, a grant application may not be approved unless the airport itself operates according to certain standards. See id. § 47107. Among these standards, the airport owner must “maintain a current layout plan of the airport” that is approved by the FAA. Id. § 47107(a)(16). The FAA must approve any modified airport layout plan (“ALP”) before the owner of the airport implements any changes. See id. § 47107(a)(16)(B); id. § 47104. When the approval of such a plan constitutes a “major Federal action[ ] significantly affecting the quality of the human environment” under the National Environmental Policy Act of 1969, 42 U.S.C. §§ 4321-4347 (“NEPA”), the FAA must prepare an environmental impact statement (“EIS”) determining the plan’s effect on the environment and considering reasonable alternatives. See id. § 4332(C); Communities Against Runway Expansion v. FAA, 355 F.3d 678, 681 (D.C.Cir.2004); see also 49 U.S.C. § 47106(c).
Once an airport owner has an approved ALP, it may apply for a Letter of Intent (“LOI”) to provide AAIA funding for the project. Upon such application, the FAA “may issue a letter of intent to the sponsor stating an intention to obligate from future budget authority an amount.” Id. § 47110(e)(1). Any such statement of intention, however, is non-binding on the federal government. Id. § 47110(e)(3).
B.
In December 2002, the City, which owns and operates O’Hare, submitted for FAA review an ALP designed to increase capacity and decrease costly delays that were interfering with O’Hare’s role as a major connecting hub. The City’s plan called for realigning three of the seven existing runways and adding an eighth runway. To accomplish the expansion, the plan would require the City to acquire 440 acres of adjacent property, including businesses and homes in the neighboring Villages of Bensenville and Elk Grove. Further, the plan would require relocation of two cemeteries: St. Johannes and Rest Haven.
*59Upon receiving the City’s application, the FAA prepared an EIS that initially screened fifteen alternatives. After rejecting many of the alternatives as implausible or insufficient means of addressing the delays at O’Hare, and after a second screening eliminated three more, the FAA compared the four remaining alternatives, using computer software to perform simulations that modeled how well each alternative would enhance capacity and reduce delays. The FAA concluded that the City’s plan — Alternative C — with the shortest average delay and $150 million savings in the five years following construction, was clearly preferable to all others.
Members of St. Johannes Church and descendants of those buried at the cemeteries objected that the relocation of the cemeteries would substantially burden their exercise of religion because of their belief in the physical resurrection of the bodies of Christian believers. Citing their rights under RFRA, they asked the FAA to demonstrate that Alternative C was the least restrictive means of meeting the government’s compelling interests in reducing delay and enhancing capacity. Although expressing uncertainty over whether it was required to comply with RFRA in this instance because the City was ultimately responsible for designing and implementing the expansion plan, the FAA proceeded as if RFRA did apply in order to avoid litigation over the project. It found that the religious practices of some petitioners would be substantially burdened if the cemeteries were acquired and the bodies were relocated by the City. It concluded-— after examining the petitioners’ proposals for avoiding the relocation of the cemeteries, each of which the FAA characterized as derivative of the no-build and limited build alternatives it had already considered and rejected, and examining derivatives of its own that would limit effects on the cemeteries, each of which the FAA determined posed significant difficulties that would interfere with the goal of reducing delays — that Rest Haven Cemetery could remain at its current location by repositioning certain cargo facilities.
Therefore, in the final EIS, the FAA proposed to conclude that Alternative C, as modified to protect Rest Haven Cemetery, was the least restrictive means of achieving the federal government’s compelling interest in increasing capacity and reducing delay. After receiving additional comments, the FAA reviewed the conclusions of the EIS in the Record of Decision (“ROD”) and concluded that the approval of Alternative C as modified satisfied RFRA because “it clearly performs so much better than any other alternative.” The issuance of the ROD on September 30, 2005 cleared the way for the City to implement its plan without relocating Rest Haven Cemetery. Additionally, in response to the City’s February 15, 2005 request for a LOI for $300 million of Airport Improvement Program (“AIP”) funds and $55.8 million of entitlement funds to go toward the first construction the City would be doing to implement the ALP, the FAA, upon making statutory findings, on November 21, 2005 issued a LOI expressing its intention to fund about $300 million in discretionary funds and $37.2 million in entitlement funds over a 15-year period. These petitions for review followed.
II.
The petitioners’ central challenge to the FAA’s approval of the City’s ALP depends upon RFRA. In 1990, the Supreme Court held in Employment Division v. Smith, 494 U.S. 872, 110 S.Ct. 1595, 108 L.Ed.2d 876 (1990), that the Free Exercise Clause of the First Amendment to the Constitution does not prohibit burdens on the exer*60cise of religion imposed by neutral laws of general applicability. See id. at 879, 110 S.Ct. 1595. In so doing, the Court declined to apply the compelling interest balancing test set forth in Sherbert v. Verner, 374 U.S. 398, 83 S.Ct. 1790, 10 L.Ed.2d 965 (1963), which required that governments demonstrate that laws substantially burdening religious exercise are supported by a compelling interest. See Smith, 494 U.S. at 882-85, 110 S.Ct. 1595. Congress, seeking to secure a wider berth for religious exercise, enacted RFRA, which aimed to reinstate the compelling interest test in place of the neutrality standard pronounced by the Court. See Religious Freedom Restoration Act of 1993, Pub.L. No. 103-141, 107 Stat. 1488 (1993) (prior to 2000 amendment); see also S. Rep. NO. 103-111, at 8 (1993), as reprinted in 1993 U.S.C.C.A.N. 1892, 1897-98; H.R. Rep. 103-88 (1993). In City of Boerne v. Flores, 521 U.S. 507, 117 S.Ct. 2157, 138 L.Ed.2d 624 (1997), the Supreme Court held that RFRA was unconstitutional as applied to the states because it was beyond Congress’s remedial power to regulate states under Section 5 of the Fourteenth Amendment to the Constitution. See id. at 536, 117 S.Ct. 2157. RFRA’s compelling interest test remained in effect as to the federal government.1
RFRA provides that “Government shall not substantially burden a person’s exercise of religion,” 42 U.S.C. § 2000bb-l(a), unless application of the burden “is the least restrictive means of furthering [a] compelling governmental interest,” id. § 2000bb-l(b)(2). “Government” is defined as “a branch, department, agency, instrumentality, and official (or other person acting under color of law) of the United States.” Id. § 2000bb-2(l). The petitioners maintain that the relocation of St. Johannes Cemetery would substantially burden the religious exercise of some of the petitioners by interfering with a sacred site of worship and with the physical resurrection of those buried there. In approving a new O’Hare ALP that would require the cemetery’s relocation as eligible for federal funding, the FAA violated RFRA, petitioners contend, because the FAA did not demonstrate that Alternative C (even as modified) is the least restrictive means of furthering the governmental interest in increasing capacity and reducing delay.
As the FAA is undeniably an “agency ... of the United States,” id. § 2000bb-2(l), it is prohibited from substantially burdening a person’s exercise of religion except when it can meet the compelling interest test. Whether that prohibition is implicated, however, depends on *61whether the FAA’s approval of the City’s ALP is properly characterized as the source of what the petitioners contend is a substantial burden placed on the free exercise of religion.2 The City, as intervenor, in addition to objecting that its plan does not burden petitioners under First Amendment precedent, contends that the relocation of the cemetery does not implicate RFRA because the City, not the FAA, is responsible for the imposition of the claimed burden on religious exercise. Although “[ijntervenors may only argue issues that have been raised by the principal parties,” and the petitioners object that the City’s contention that RFRA is not implicated by the FAA’s action is not properly before this court, the court retains “discretion [to] entertain arguments raised only by an intervenor on review if they have been ‘fully litigated in the agency proceedings and [are] potentially determinative of the outcome of judicial review.’ ” Nat’l Ass’n of Regulatory Util. Comm’rs v. ICC, 41 F.3d 721, 729-30 (D.C.Cir.1994) (quoting Synovus Fin. Corp. v. Bd. of Governors of the Fed. Reserve Sys., 952 F.2d 426, 433 (D.C.Cir.1991)). The court is particularly solicitous of intervenors’ arguments in cases in which the intervenor’s success before the agency forecloses it from petitioning for review and the issue raised logically precedes the issues in dispute between the principal parties. See id. at 730. The City, and the issue it raises, satisfy those criteria. The court, therefore, will consider the FAA’s responsibility for the burden on religious exercise.
To determine whether the burden on religious exercise can be properly attributed to the FAA, the petitioners suggest that the court look to the fact that the FAA’s role in approving the ALP constitutes a “major Federal action[ ]” for purposes of NEPA. 42 U.S.C. § 4332(C); see 40 C.F.R. § 1508.18(b)(4). From this perspective, because the FAA must answer for the effects of its decision on the environment, it must also answer for the effects on religious exercise. NEPA broadly applies to all “projects and programs entirely or partly financed, assisted, conducted, regulated, or approved by federal agencies.” 40 C.F.R. § 1508.18(a). “Major Federal action” is defined to “includ[e] actions with effects that may be major and which are potentially subject to Federal control and responsibility,” “including] the circumstance where the responsible officials fail to act.” Id, § 1508.18. Applying NEPA’s sweeping definitions would undoubtedly subject the FAA’s approval of the City’s ALP and its intention to provide partial funding for the O’Hare expansion to the strictures of RFRA.
This approach has some facial appeal. In aviation, as in many fields, the federal government plays a significant role. Cf. Northwest Airlines, Inc. v. Minnesota, 322 U.S. 292, 303, 64 S.Ct. 950, 88 L.Ed. 1283 (1944). The Illinois legislature noted as much when endorsing the O’Hare expansion plan in “recognizing] that ... the planning, construction, and use of the O’Hare Modernization Program will be subject to intensive regulatory scrutiny by the United States and that no purpose would be served by duplicative or redundant regulation of the safety and impacts of the airport or the O’Hare Modernization Program.” 620 ILL. COMP. STAT. 65/5(a)(8) (2003). Given the FAA’s author*62ity to guide airport development nationwide, one might be tempted to use NEPA to hold the FAA accountable for nearly every aspect of development at the airports that it regulates. But the Supreme Court has instructed that “courts must look to the underlying policies or legislative intent in order to draw a manageable line between those causal changes that may make an actor responsible for an effect and those that do not.” Metro. Edison Co. v. People Against Nuclear Energy, 460 U.S. 766, 774 n. 7, 103 S.Ct. 1556, 75 L.Ed.2d 534 (1983). Under that analysis, it becomes clear that NEPA’s broad application is ill-suited to RFRA’s statutory regime.
Although RFRA speaks broadly of “government,” with RFRA Congress intended to “restore” the standard by which federal government actions burdening religion were to be judged, see 42 U.S.C. § 2000bb(b)(1); City of Boerne, 521 U.S. at 532, 117 S.Ct. 2157, not to expand the class of actions to which the standard would be applied, see Hall v. Am. Nat. Red Cross, 86 F.3d 919, 921 (9th Cir.1996) (quoting S. Rep. NO. 103-111, at 12, as reprinted in 1993 U.S.C.C.A.N. at 1901). The Senate Judiciary Committee reported that “the purpose of [RFRA] is only to overturn the Supreme Court’s decision in Smith,” not to “unsettle other areas of the law.” S. Rep. NO. 103-111, at 12, as reprinted in 1993 U.S.C.C.A.N. at 1902. RFRA was not meant to “expand, contract or alter the ability of a claimant to obtain relief in a manner consistent with the Supreme Court’s free exercise jurisprudence under the compelling governmental interest test prior to Smith.” Id. To the extent RFRA was designed to restore a legal standard rather than to extend the compelling interest test to the far reaches of government activities, importing NEPA’s applicability into RFRA would give the statute far greater breadth than Congress ever intended. See Hall, 86 F.3d at 921. Moreover, to do so would fundamentally recast federal-state relations in a manner that Congress has yet to indicate that it intended.
RFRA was enacted to reestablish a constitutional test with the expectation that courts would look to constitutional precedent for guidance. See S. Rep. NO. 103-111, at 8, as reprinted in 1993 U.S.C.C.A.N. at 1898. The petitioners can point to no indication of a contrary expectation and we have found none. Given that background, it is not surprising then that the proper inquiry is one classically reserved for constitutional jurisprudence. Whether the federal government can be characterized as responsible for the relocation of St. Johannes Cemetery under RFRA requires the same analysis normally necessary to determine whether the FAA could be held responsible for an alleged infringement of constitutional rights. Cf. Rendell-Baker v. Kohn, 457 U.S. 830, 838, 102 S.Ct. 2764, 73 L.Ed.2d 418 (1982). Modified to apply to this inquiry, the question the court must decide is whether “there is a sufficiently close nexus between the [federal government] and the challenged action of [the City] so that the action of the latter may be fairly treated as that of the [federal government] itself.” Blum v. Yaretsky, 457 U.S. 991, 1004, 102 S.Ct. 2777, 73 L.Ed.2d 534 (1982) (quoting Jackson v. Metro. Edison Co., 419 U.S. 345, 351, 95 S.Ct. 449, 42 L.Ed.2d 477 (1974)). “The purpose of this requirement,” the Supreme Court explains, “is to assure that constitutional standards are invoked only when it can be said that the State [here, the FAA] is responsible for the specific conduct of which the plaintiff complains.” Id. Similarly, by conducting a state action inquiry here, the court can assure that RFRA’s heightened standard *63is only applied when it can be said that the federal government is responsible for the burden on religious exercise. See Sutton v. Providence St. Joseph Med. Ctr., 192 F.3d 826, 834-35 (9th Cir.1999); see also Hall, 86 F.3d at 921-22.3
This case presents an unusual state action question because the regulated party is a separate sovereign rather than a private entity. Despite the rarity of this situation (inasmuch as federal and state govern-mente are generally bound by the same constitutional standards), the analysis proceeds with the same “necessarily fact-bound inquiry,” Lugar v. Edmondson Oil Co., 457 U.S. 922, 939, 102 S.Ct. 2744, 73 L.Ed.2d 482 (1982), as if the federal government were regulating the decision of a private entity, with the City standing in the place of a private party. See, e.g., Kitchens v. Bowen, 825 F.2d 1337 (9th Cir.1987), cert. denied, 485 U.S. 934, 108 S.Ct. 1109, 99 L.Ed.2d 270 (1988).4
*64The Supreme Court has held that “[t]he mere fact that a business is subject to state regulation does not by itself convert its action into that of the State.” Jackson, 419 U.S. at 350, 95 S.Ct. 449. Furthermore, “a State normally can be held responsible for a private decision only when it has exercised coercive power or has provided such significant encouragement, either overt or covert, that the choice must in law be deemed to be that of the State.” Blum, 457 U.S. at 1004, 102 S.Ct. 2777. “Mere approval of or acquiescence in the initiatives of a private party is not sufficient to justify holding the State responsible for those initiatives .... ” Id. at 1004-05, 102 S.Ct. 2777; see also Am. Mfrs. Mutual Ins. Co. v. Sullivan, 526 U.S. 40, 52, 119 S.Ct. 977, 143 L.Ed.2d 130 (1999); Moose Lodge v. Irvis, 407 U.S. 163, 177, 92 S.Ct. 1965, 32 L.Ed.2d 627 (1972). The receipt of public funds, even of “virtually all” of an entity’s funding, is not sufficient to fairly attribute the entity’s actions to the government. See Rendell-Baker, 457 U.S. at 840-41, 102 S.Ct. 2764 (citing Blum, 457 U.S. at 1011, 102 S.Ct. 2777).
In analyzing whether the alleged burden on religious exercise is fairly attributable to the FAA, we “begin[] by identifying ‘the specific conduct of which the plaintiff complains.’ ” Am. Mfrs. Mut. Ins. Co., 526 U.S. at 51, 119 S.Ct. 977 (quoting Blum, 457 U.S. at 1004, 102 S.Ct. 2777). The specific conduct that the petitioners challenge is the seizure and relocation of St. Johannes Cemetery. Consequently, the court must decide whether the FAA’s role in the potential disinterment at St. Johannes is “[m]ere approval of or acquiescence in” the City’s plan or whether the FAA “has exercised coercive power or has provided such significant encouragement, either overt or covert, that the choice must in law be deemed to be that of the [FAA].” Blum, 457 U.S. at 1004, 102 S.Ct. 2777. Fairly characterizing the level of federal governmental involvement can be accomplished “[o]nly by sifting facts and weighing circumstances.” Moose Lodge, 407 U.S. at 172, 92 S.Ct. 1965. “[A]bsent government coercion or significant government encouragement of the measure under inspection,” Lunceford v. Dist. of Columbia Bd. of Educ., 745 F.2d 1577, 1581 (D.C.Cir.1984), the Supreme Court has held that the federal government may not be held responsible for a measure taken by a private actor.
In Citizens Against Burlington, Inc. v. Busey, 938 F.2d 190 (D.C.Cir.1991), this court quoted an EIS that described the FAA’s role in airport development:
In the present system of federalism, the FAA does not determine where to build and develop civilian airports, as an owner/operator. Rather, the FAA facilitates airport development by providing Federal financial assistance, and reviews and approves or disapproves revisions to Airport Layout Plans at Federally funded airports. Id. at 197.
That quotation accurately depicts the FAA’s involvement here, where it reported that “[t]he FAA did not design the [ALP], nor can it compel the City to implement some or all of it.” If the owner or operator proposing to modify an airport’s layout *65wants to qualify for federal funding, certain requirements must be met. Under the AAIA, ALPs must “be in a form the [FAA] prescribes,” 49 U.S.C. § 47107(a)(16)(A), and the airport’s operator may “not make or allow any alteration in the airport or any of its facilities if the alteration does not comply with the plan the [FAA] approves,” id. § 47107(a)(16)(C); see also Communities Against Runway Expansion, 355 F.3d at 681. Therefore, if the City wishes to receive any federal funding, it cannot relocate the cemetery without the approval of the FAA.
But the FAA’s peripheral role in the City’s relocation of St. Johannes is not sufficient to hold the agency responsible for purposes of RFRA. Even under NEPA, “a ‘but for’ causal relationship is insufficient to make an agency responsible for a particular effect.” Dep’t of Transp. v. Pub. Citizen, 541 U.S. 752, 767, 124 S.Ct. 2204, 159 L.Ed.2d 60 (2004). The City — not the FAA — is the cause of any burden on religious exercise because of its role as inventor, organizer, patron, and builder of the O’Hare expansion. The City designed the ALP with its attendant impact on the cemeteries. The City submitted the plan to the FAA to retain O’Hare’s eligibility for federal funding. Before the FAA, the City fought for approval of its plan. The City will provide the lion’s share of the funding for the modernization project; the federal government will cover only twelve percent of the cost of Phase I. The City intends to provide all of the funding through other sources if the federal funds are not forthcoming. And at the end of the day, the City will carry out the seizure and physical relocation of St. Johannes Cemetery.
Our dissenting colleague would find that the FAA did more than merely approve the ALP because of its thorough consideration of alternatives pursuant to NEPA. See Dissenting Op. at 76-77. It is true that the FAA was careful in reviewing the harms to the environment and the benefits to the flying public of approving the City’s plan. But the measured approach the FAA took in approving the City’s ALP does not make the City’s plan an action of the federal government. The Supreme Court has never held that the government becomes responsible for the actions of a third party due to the length or intensity of its attention to the actions of the party before approval. Now that the FAA has approved the ALP, the FAA has no authority to demand that the City build the projects described therein. Earlier in the process, there is no indication that the FAA “exercised coercive power” or “provided ... significant encouragement,” Blum, 457 U.S. at 1004, 102 S.Ct. 2777, that provoked the City to choose a plan that would harm the cemetery. Indeed, the only significant modification to the ALP encouraged by the FAA was the rescue of Rest Haven Cemetery spurred by the FAA’s assumption that it was bound by RFRA. Had this course of events played out differently, with the FAA ordering the change responsible for the burden or playing some greater role in the design of the ALP, perhaps there might be a valid claim under RFRA. But that level of responsibility is not present here.5
*66To the contrary, the FAA’s role as regulator is similar to that in many cases where the Supreme Court has declined to find state action. See, e.g., Rendell-Baker, 457 U.S. at 841-42, 102 S.Ct. 2764. For instance, in Moose Lodge, 407 U.S. 163, 92 S.Ct. 1965, 32 L.Ed.2d 627, the Court held that a state’s issuance of a liquor license to a private club was insufficient to attribute the club’s refusal to serve an African-American to the state. See id. at 176-77, 92 S.Ct. 1965. And in Jackson v. Metropolitan Edison Company, the Court held that the state’s utilities commission was not sufficiently connected with a privately owned utility’s decision to terminate electric service to apply constitutional standards. See Jackson, 419 U.S. at 358-59, 95 S.Ct. 449. The Supreme Court observed that “governmental regulation of private utilities is such that a utility may frequently be required by the state regulatory scheme to obtain approval for practices a business regulated in less detail would be free to institute without any approval from a regulatory body,” and that approval “where the commission has not put its own weight on the side of the proposed practice by ordering it, does not transmute a practice initiated by the utility and approved by the commission into ‘state action.’ ” Id. at 357, 95 S.Ct. 449. “At the most,” the Court labeled the commission’s action a “failure to overturn this practice” that gave the utility the freedom to decide whether to employ it. Id. Similarly, despite the FAA’s broad regulatory power to approve ALPs as an incident to determining a development project’s eligibility for federal funding, the City’s “exercise of the choice allowed by [the FAA] where the [challenged] initiative comes from [the City] and not from the [FAA] does not make its action in doing so ‘state action.’ ” Id. (citation omitted).
In analyzing cases in which the Supreme Court found no state action in the choices of heavily regulated entities, this court found critical “the interposition of the independent judgment of a private party between the act that allegedly resulted in a constitutional deprivation and the decision of the state to accept that decision and continue funding the private activities.” Kolinske v. Lubbers, 712 F.2d 471, 480 (D.C.Cir.1983). Here, it was the “conduct of [the City] exercising independent judgment that yielded the contested result.” Id. Where the FAA “cannot be said to in any way foster or encourage,” Moose Lodge, 407 U.S. at 176-77, 92 S.Ct. 1965, the burden on religious exercise, “the simple device of characterizing the [FAA]’s inaction as ‘authorization’ or ‘encouragement,’ ” Flagg Bros., Inc. v. Brooks, 436 U.S. 149, 164-65, 98 S.Ct. 1729, 56 L.Ed.2d 185 (1979), is insufficient to justify imposition of RFRA’s compelling interest test. See also Am. Mfrs. Mut. Ins. Co., 526 U.S. at 53-54, 119 S.Ct. 977. The burden imposed on religious exercise by the City’s choices with the mere approval or acquiescence of the FAA does not require the FAA to demonstrate a compelling interest. See Blum, 457 U.S. at 1004, 102 S.Ct. 2777.
Our dissenting colleague errs in concluding that mere approval is sufficient to hold the government responsible for the actions *67of a third party. It is plainly incorrect to state that “both the Supreme Court and this [c]ourt held, prior to Smith, that a federal agency’s approval can be the ‘source’ of a burden on religious exercise.” Dissenting Op. at 75. In the cases cited by the dissent—Lyng v. Northwest Indian Cemetery Protective Ass’n, 485 U.S. 439, 108 S.Ct. 1319, 99 L.Ed.2d 534 (1988), and Wilson v. Block, 708 F.2d 735 (D.C.Cir.1983)—the courts held that the government action did not constitute a burden on religious exercise within the meaning of the First Amendment. Lyng, 485 U.S. at 447, 108 S.Ct. 1319; Wilson, 708 F.2d at 745. Not even the quoted dicta from those decisions supports the notion that government acquiescence in another actor’s decision justifies finding that the government has violated the Constitution. In Lyng, the federal government proposed building a road and harvesting timber on federal land where Indians traditionally practiced their religion. See Lyng, 485 U.S. at 442-43, 108 S.Ct. 1319. The Supreme Court merely noted “that the Government’s proposed actions will have severe adverse effects on the practice of their religion.” Id. at 447, 108 S.Ct. 1319. The Court never addressed the question that the dissent claims that it did: whether “approval by a federal agency of third party action can be subject to a free exercise challenge.” Dissenting Op. at 76. And the Court had no reason to do so, because the ease had nothing to do with “approval by a federal agency of a third party action.” Lyng never mentions any party besides the government, because “the Forest Service [would] build a 6-mile paved segment” through the sacred area, id. at 442, 108 S.Ct. 1319, and “the Forest Service adopted a management plan allowing for the harvesting of significant amounts of timber,” id. at 443, 108 S.Ct. 1319. No third party was involved.
Similarly, Wilson involved a challenge to a federal governmental decision about what to do with federal land. There, the federal government proposed to allow private interests to develop ski facilities on federal land used for religious practice by several Indian tribes. See Wilson, 708 F.2d at 738. This court merely noted that the “construction approved by the [government] ... will cause the plaintiffs spiritual disquiet.” Id. at 742. Lyng and Wilson involved the government’s use of its own land rather than the government’s regulation of, a third party’s use of the third party’s land. Therefore, even if Lyng and Wilson had dealt with the question of who was responsible for the burden on religious exercise (which they did not), they would not apply here. The Supreme Court “has never held that a [government’s mere acquiescence in a private action cqnverts that action into that of the [government].” Flagg Bros., 436 U.S. at 164, 98 S.Ct. 1729. We decline our dissenting colleague’s invitation to do so today.
That the regulated party here is a government (i.e., a part of a sovereign State,) heightens our hesitancy to apply RFRA’s compelling interest test. To do so would subject the City’s airport-building plans to “the most demanding test known to constitutional law.” City of Boerne, 521 U.S. at 534, 117 S.Ct. 2157. In this case, state and local governments would again be hampered by RFRA’s “intrusion into the States’ traditional prerogatives and general authority to regulate for the health and welfare of their citizens.” Id. In City of Boerne, the Supreme Court rejected RFRA’s “intrusion at every level of government,” id. at 532, 117 S.Ct. 2157, observing that RFRA would exact substantial costs “both in practical terms of imposing a heavy litigation burden on the States and in terms of curtailing their traditional general regulatory power,” id. at 534, 117 S.Ct. 2157. Applying RFRA *68indirectly, by forcing the FAA rather than the courts to curtail the state’s traditional powers, would exact those costs just as surely as applying RFRA directly to the states. Whether Congress might exact those costs through its power to regulate commerce or place conditions on federal spending is a question the court need not decide because Congress has not attempted such a feat. See O’Bryan v. Bureau of Prisons, 349 F.3d 399, 401 (7th Cir.2003). If Congress seeks to alter so dramatically the balance of power between states and the federal government, it must state clearly its intention to do so. See Vermont Agency of Natural Res. v. United States ex rel. Stevens, 529 U.S. 765, 787, 120 S.Ct. 1858, 146 L.Ed.2d 836 (2000); Gregory v. Ashcroft, 501 U.S. 452, 460-61, 111 S.Ct. 2395, 115 L.Ed.2d 410 (1991). Because the relocation of St. Johannes Cemetery cannot be fairly attributed to the actions of the FAA, the petitioners’ RFRA claim fails.
III.
The petitioners also seek vacation of the Letter of Intent expressing the FAA’s intention to obligate federal funds to carry out the O’Hare expansion once the City submits grant applications for approval. The petitioners contend the FAA failed to make essential findings mandated by statute.
To begin, the court must determine whether it has jurisdiction to address a challenge to the LOI. See Citizens for the Abatement of Aircraft Noise v. Metro. Wash. Airports Auth., 917 F.2d 48, 53 (D.C.Cir.1990). Section 46110(a) of the AAIA provides that “a person disclosing a substantial interest in an order issued by the Secretary of Transportation ... in whole or in part under ... part B ... may apply for review of the order by filing a petition in [this court].” A LOI is issued under Part B. See id. § 47110(e). The question remains, however, whether a LOI is an “order” at all, much less a final order subject to judicial review.
Few courts have had the opportunity to address what constitutes an “order” under § 46110(a), but many, including this one, see City of Rochester v. Bond, 603 F.2d 927, 932-33 (D.C.Cir.1979), have interpreted the statutory section’s predecessor, 49 U.S.C. § 1486 (1976). See Aerosource, Inc. v. Slater, 142 F.3d 572, 577 (3d Cir.1998) (collecting cases). In doing so, the courts have concluded that an “order” must possess the quintessential feature of agency decisionmaking suitable for judicial review: finality. See Aerosource, 142 F.3d at 577-78; see also City of Rochester, 603 F.2d at 932-33; Air Cal. v. Dep’t of Transp., 654 F.2d 616, 622 (9th Cir.1981). The Supreme Court has explained:
As a general matter, two conditions must be satisfied for agency action to be “final”: First, the action must mark the “consummation” of the agency’s deci-sionmaking process, Chicago & Southern Air Lines, Inc. v. Waterman S.S. Corp., 333 U.S. 103, 113, 68 S.Ct. 431, 92 L.Ed. 568 (1948)—it must not be of a merely tentative or interlocutory nature. And second, the action must be one by which “rights or obligations have been determined,” or from which “legal consequences will flow,” Port of Boston Marine Terminal Assn. v. Rederiaktiebolaget Transatlantic, 400 U.S. 62, 71, 91 S.Ct. 203, 27 L.Ed.2d 203 (1970).
Bennett v. Spear, 520 U.S. 154, 177-178, 117 S.Ct. 1154, 137 L.Ed.2d 281 (1997). Therefore, the outcome of this inquiry depends upon the place of the LOI in the FAA’s decisionmaking process and upon its precise legal effect.
A LOI is an odd creature of statute that, unlike other more definitive agency decisions, merely “stat[es] an intention to obli*69gate from future budget authority an amount.” 49 U.S.C. § 47110(e)(1). “A letter of intent ... is not an obligation of the Government ... and the letter is not deemed to be an administrative commitment for financing.” Id. § 47110(e)(3). Instead, a LOI is a planning document that “establish[es] a schedule under which the [FAA] will reimburse the sponsor for the Government’s share of allowable project costs, as amounts become available.” Id. § 47110(e)(1). It does not complete the agency’s decisionmaking process, as the City is required to file a further grant application for approval before the FAA will be obligated to disburse the funds described in the LOI. See id. § 47105. Thus, not only is the funding decision contingent on congressional appropriation, it also will require further administrative process. It follows that the LOI is non-final because it “does not itself adversely affect [the petitioners] but only affects [their] rights adversely on the contingency of future administrative action.” DRG Funding Corp. v. Sec’y of Hous. and Urban Dev., 76 F.3d 1212, 1214 (D.C.Cir.1996) (quoting Rochester Tel. Corp. v. United States, 307 U.S. 125, 130, 59 S.Ct. 754, 83 L.Ed. 1147 (1939)).
Moreover, the LOI is non-final because it does not impose an obligation, deny a right, or otherwise fix some legal relationship. See Reliable Automatic Sprinkler Co. v. Consumer Prod. Safety Comm’n, 324 F.3d 726, 731 (D.C.Cir.2003). The statutory text is clear in this regard. When asked during oral argument what the LOI meant to the City, counsel for the City helpfully responded that it is a planning tool that enables the City to approach financial partners for private funding for the development plan. Although airports and their financiers may rely on LOIs as planning tools, this provides the court with no basis to conclude that the LOI establishes a right or obligation when the statute explicitly denies that the LOI is an “obligation” or a “commitment.” Id. § 47110(e)(3). “Finality resulting from the practical effect of an ostensibly nonbinding agency proclamation is a concept we have recognized in the past,” but “if the practical effect of the agency action is not a certain change in the legal obligations of a party, the action is non-final for the purpose of judicial review.” Nat’l Ass’n of Home Builders v. Norton, 415 F.3d 8, 15 (D.C.Cir.2005). The LOI has no effect absent two conditions precedent: FAA approval of a further grant application by the City and congressional appropriation of funds. The noncommittal language of § 47110 deprives a LOI of the force necessary to make it a final order that may be judicially reviewed because it has no “direct and appreciable legal consequences.” Bennett, 520 U.S. at 178, 117 S.Ct. 1154. Therefore, the court is without jurisdiction to review issuance of the LOI under 49 U.S.C. § 46110.
Even were the nature of the commitment evidenced by the LOI such that the court could find that it had jurisdiction, the petitioners’ injury is not redressable by a decision vacating the LOI because the O’Hare ALP would go forward without the LOI funds. “The redressability inquiry poses a simple question: ‘If plaintiffs secured the relief they sought, would it redress their injury’?” Wilderness Soc. v. Norton, 434 F.3d 584, 590 (D.C.Cir.2006) (quoting Mountain States Legal Found. v. Glickman, 92 F.3d 1228, 1233 (D.C.Cir.1996)) (alterations omitted). Thus, for purpose of determining the petitioners’ standing, the court must decide whether “the practical consequence of [vacating the LOI] would amount to a significant increase in the likelihood that [the petitioners] would obtain relief that directly redresses the injury suffered.” Utah v. *70Evans, 536 U.S. 452, 464, 122 S.Ct. 2191, 153 L.Ed.2d 453 (2002). Or in concrete terms, would the City develop O’Hare as planned without the $337 million in federal funding set forth in the LOI?
The petitioners maintain that the LOI funding is vital because although the major airlines at O’Hare have agreed to the Phase I business plan, the airlines’ final approval is contingent on the City receiving $300 million in AIP discretionary funds. Indeed, the FAA’s Inspector General reported, “If the AIP funds are not granted, the City will have to renegotiate approval of Phase 1 with the airlines.” That same report makes clear, however, that “[i]f any shortfalls in funding or increases in project costs materialize, the City has indicated it plans to make up the funding/cost difference by issuing additional bonds.” The FAA maintains that vacating the LOI would not affect the City’s ability to complete the project because practically, the LOI funds only represent a tenth of the funding of the Phase I Airfield, and legally, nothing would preclude the City from buying the property and asking for a federal reimbursement grant later. See 49 U.S.C. § 47110(c)(1); Respondent’s Br. at 72. Even if the City were permanently deprived of federal funds, it could issue more revenue bonds, which would increase its costs by a relatively insignificant margin. Based on a number of studies of the project’s financial feasibility, the FAA has determined that removing the LOI funds would not imperil the project. The City agrees, stating that other traditional sources of airport financing are sufficient to fund the project even if the LOI funds are withdrawn.
Although the City’s agreement with the major airlines at O’Hare might require further negotiations were the LOI vacated, renegotiations with the airlines do not create “a significant increase in the likelihood,” Utah v. Evans, 536 U.S. at 464, 122 S.Ct. 2191, that the project would be scuttled altogether rather than merely delayed. The relatively minor role of the LOI dollars in funding Phase I of the O’Hare expansion, the fact that the City could return to the FAA for a grant in a new application, and the existence of alternative sources of funding means that vacating the LOI is unlikely to redress the petitioners’ injury. Therefore, because the petitioners do not satisfy the redressability requirement of Article III standing, the court cannot reach the merits of them challenge to the LOI.
IV.
Finally, the petitioners present a variety of administrative law challenges to the FAA’s decisionmaking process. They contend that the FAA (1) used stale and unreliable data in a manner arbitrary, capricious, and contrary to law under the Administrative Procedure Act, 5 U.S.C. §§ 701-706, and NEPA, and (2) violated the Due Process Clause of the Fifth Amendment by denying them fair deci-sionmaking procedures. Neither contention has merit.
A.
“A party seeking to have a court declare an agency action to be arbitrary and capricious carries ‘a heavy burden indeed.’ ” Wisconsin Valley Improvement v. FERC, 236 F.3d 738, 745 (D.C.Cir.2001)(quoting Transmission Access Policy Study Group v. FERC, 225 F.3d 667, 714 (D.C.Cir.2000)). It must show that the agency has failed to consider relevant factors, see Citizens to Preserve Overton Park, Inc. v. Volpe, 401 U.S. 402, 416, 91 S.Ct. 814, 28 L.Ed.2d 136 (1971), has made a clear error in judgment, see id., or has failed to “articulate a satisfactory explanation for its action including a *71‘rational connection between the facts found and the choice made,’ ” Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983) (quoting Burlington Truck Lines v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 9 L.Ed.2d 207 (1962)).
The petitioners contend that the FAA erred in using an unreasonably short period of evaluation to gauge delay savings benefits by ending its evaluation in 2018, just five years after the project’s planned completion. They maintain that a five-year time horizon is at odds with FAA Order 1050.1E, which states that the FAA “usually selectfs]” a timeframe lasting until “5 to 10 years after implementation.” Needless to say, a “build out plus five year” timeframe is consistent with an order recommending timeframes “5 to 10 years after implementation.” But the petitioners also contend quite reasonably that a longer timeframe is desirable for a project of this size and note that longer time horizons have been used to assess other airport development projects. They fail to note, however, that the O’Hare modernization plan will take many years to complete, naturally pushing any useful timeframe far into the future. As it will take eleven years to complete the O’Hare project, the FAA’s projection extends sixteen years into the future. The FAA determined that predictions any further along would be of questionable reliability, which would defy the FAA’s NEPA obligation to determine “reasonably foreseeable” impacts. 40 C.F.R. §§ 1508.25(c); 1508.8. The petitioners present no grounds on which the court could question that judgment. The time-frame used was in keeping with FAA precedent and reasonable under the circumstances.
The petitioners also contend that the FAA should have used the 2003 rather than the 2002 Terminal Area Forecast (“TAF”) in its computer modeling. Using the more recent forecast, according to the petitioners, would have demonstrated that the ALP was ineffective in preventing delays from returning to O’Hare. The 2002 forecast was the most recent available at the time the FAA began its work. In the ROD, the FAA provided three reasons for relying on the 2002 TAF: its belief in the accuracy of the representation provided by the 2002 forecast, the administrative necessity of cutting off new data at some point, and its ability to take account of the 2003 and 2004 TAF projections by other means. The record demonstrates that the FAA conducted analyses to determine how significantly variations in the 2003 and 2004 TAFs would affect its modeling and that it determined the variations would not affect its conclusions. However desirable it may be for agencies to use the most current and comprehensive data available when making decisions, the FAA has expressed its professional judgment that the later data would not alter its conclusions in the EIS or the approval of Alternative C, and it is reasonably concerned that an unyielding avalanche of information might overwhelm an agency’s ability to reach a final decision. Cf. W. Coal Traffic League v. ICC, 735 F.2d 1408, 1411 (D.C.Cir.1984). The method that the FAA chose, creating its models with the best information available when it began its analysis and then checking the assumptions of those models as new information became available, was a reasonable means of balancing those competing considerations, particularly given the many months required to conduct full modeling with new data.
Similarly, there is no merit to the petitioners’ claims that the FAA improperly relied on other outdated information. The petitioners contend that the FAA used outdated airport operating procedures to *72construct its base case, rendering its comparison of alternatives invalid. But as the FAA explains, these operational procedures remain in place. The FAA’s assumptions regarding airline behavior in response to the restrictions are typical of the predictive judgments to which courts defer and the petitioners fail to show that they were not reasonable. See Public Utils. Comm’n of Cal. v. FERC, 24 F.3d 275, 281 (D.C.Cir.1994). The petitioners also contend that the FAA should have altered its baseline forecast to account for delay improvements caused by the FAA’s 2004 scheduling order limiting the number of flights at O’Hare. The FAA reasonably explains that it did not use those numbers in the base case because they did not appear until after the FAA began its analysis and because it would be inappropriate to choose as the baseline a set of conditions only achieved by imposing limits that Congress has deemed to be detrimental to the public interest. See 49 U.S.C. § 47101(a)(9). Again, these judgments regarding the development of the baseline against which alternatives would be assessed are the sorts of expert analytical judgments to which courts typically defer. See Public Citizen, Inc. v. FAA, 988 F.2d 186, 196-97 (D.C.Cir.1993). The petitioners have not provided any ground to doubt the reasonableness of those judgments in this instance.
The petitioners further contend that the FAA established an unprecedented acceptable delay level of fifteen minutes that made Alternative C appear better than it is at reducing delays. To begin, the petitioners fail to point out where the FAA determines that fifteen minute delays are acceptable. Perhaps the petitioners are referring to the fact that, in comparing the “No Action” alternative to other alternatives, the “No Action” alternative was constrained so as to produce between fifteen and eighteen minutes of delay. The FAA explained that such constraints were necessary to generate comparative operational and environmental impacts. The petitioners present nothing to call this judgment into doubt, and given the deference accorded the FAA in forecasting air transportation demand and capacity, see City of Olmsted Falls v. FAA, 292 F.3d 261, 272 (D.C.Cir.2002), the court concludes that the FAA was neither arbitrary nor capricious in establishing this baseline.
This litany of arbitrary and capricious challenges is thoroughly rebutted by the FAA, which appears to have acted with great care in conducting its analyses for the EIS and ROD. Without more from the petitioners, the court has no basis to conclude that the FAA’s methods were less than rational.
B.
Invoking the Fifth Amendment’s Due Process Clause, the petitioners assert that the FAA denied it a right to a fair hearing in several ways: by creating financial incentives that drive FAA employees and officials to approve runway projects, by employing individuals who formerly worked for the City or its consultants, and by withholding thousands of documents that would help them establish this procedural misconduct. Like the FAA, the court has difficulty responding to these assertions because they are vague and conclusory, and the petitioners’ failure to provide any greater detail in their reply brief suggests the weakness of these claims. Clearly, “administrative decisions made by adjudicators with a pecuniary interest in the results of the proceeding may suffer reversal,” Jonal Corp. v. District of Columbia, 533 F.2d 1192, 1197 (D.C.Cir.1976) (citing Gibson v. Berryhill, 411 U.S. 564, 579, 93 S.Ct. 1689, 36 L.Ed.2d 488 (1973), and Tumey v. Ohio, *73273 U.S. 510, 47 S.Ct. 437, 71 L.Ed. 749 (1927)), but what the petitioners describe appears to be a fairly standard performance-based compensation system and the FAA has advised that its bonuses are not directly linked to individual performance or runway approvals. See Respondent’s Br. at 61-62. As to the claim that some FAA employees formerly worked for the City or its consultants, the petitioners do not identify a single employee who might have such a conflict of interest from prior employment, so the court cannot determine whether their role in FAA decision-making was central enough to question the integrity of the process. Both of these claims of employee bias “fall[ ] far short of demonstrating that the [FAA] had ‘a fixed opinion — a closed mind on the merits of the case.’ ” Pharaon v. Bd. of Governors of Fed. Reserve Sys., 135 F.3d 148, 155 (D.C.Cir.1998) (quoting Throckmorton v. NTSB, 963 F.2d 441, 445 (D.C.Cir.1992)) (internal quotation marks omitted). Regardless, “[c]laims of bias must ‘be raised as soon as practicable after a party has reasonable cause to believe that grounds for disqualification exist,’ ” id. (quoting Marcus v. Dir., Office of Workers’ Comp. Programs, 548 F.2d 1044, 1051 (D.C.Cir.1976)), and it does not appear that petitioners raised these claims before the agency, thus waiving them here. Finally, as to the petitioners’ assertion that the FAA has withheld thousands of documents, they point to nothing in the record to justify this claim. In addition, the voluminous administrative record, much of which includes specific responses to points raised by the petitioners, their lawyers, and their consultants, belies the claim that they have been denied “a reasonable opportunity to know the claims of the opposing party and to meet them.” Morgan v. United States, 304 U.S. 1, 18, 58 S.Ct. 773, 82 L.Ed. 1129 (1938).
Accordingly, we deny the petitions for review.
. In response to City of Boeme, Congress passed the Religious Land Use and Institutionalized Persons Act ("RLUIPA”), see Pub.L. No. 106-274, 114 Stat. 803 (2000) (codified at 42 U.S.C. § 2000cc et seq.), which again applies the compelling interest standard to action by the states, but only as to the limited categories of regulations affecting land use or institutionalized persons. Congress grounded RLIUPA in its Commerce Clause and Spending Clause authority. See 42 U.S.C. §§ 2000cc(a)(2)(A)-(B), 2000cc-l(b)(l)-(2). Some of its applications have been upheld against constitutional challenge. See Cutter v. Wilkinson, 544 U.S. 709, 125 S.Ct. 2113, 161 L.Ed.2d 1020 (2005). In RLIUPA, Congress also amended RFRA to remove references to state and local governments while preserving the law's application to the federal government. See 114 Stat. at 806; see also Cutter, 544 U.S. at 715 n. 2, 125 S.Ct. 2113. A RLIU-PA claim against the City for its plan to acquire St. Johannes Cemetery was dismissed by the District Court for the Northern District of Illinois and is now pending before the Seventh Circuit Court of Appeals. See St. John's United Church of Christ v. City of Chicago, 401 F.Supp.2d 887 (N.D.Ill.2005); St. John's United Church of Christ v. City of Chicago, No. 05-4418, 2005 WL 3749817 (7th Cir. filed Dec. 30, 2005).
. The City maintains that there is no burden on petitioners under relevant First Amendment precedent. See, e.g., Lyng v. Northwest Indian Cemetery Protective Ass’n, 485 U.S. 439, 450, 108 S.Ct. 1319, 99 L.Ed.2d 534 (1988). We need not decide that question, but will assume for purposes of our analysis that the relocation of St. Johannes Cemetery imposes a substantial burden.
. Our dissenting colleague, although agreeing the court must look to pr e-Smith Free Exercise Clause cases in determining RFRA’s bounds, misreads those cases, stretching them beyond reason to eliminate virtually any limit on RFRA’s application. Relying primarily on pr e-Smith cases barring “indirect” burdens on free exercise, see Dissenting Op. at 75, the dissent fails to acknowledge that the word "indirect” in those cases referred to the nature of the burden imposed on religious exercise, not to the identity of the entity imposing the burden. In Sherbert, the Supreme Court held that the government's denial of benefits to an individual because of conduct demanded by her religious beliefs constituted á burden despite the fact that the burden was only "an indirect result” as "no criminal sanctions directly compelled]” her to violate her religious beliefs. 374 U.S. at 403, 83 S.Ct. 1790. "[T]he fact that no direct restraint or punishment [was] imposed,” id. at 404 n. 5, 83 S.Ct. 1790, was immaterial because "the pressure upon her to forego [her religious] practice [was] unmistakable,” id. at 404, 83 S.Ct. 1790. This view of what constitutes an "indirect” burden was confirmed in Thomas v. Review Bd. of the Indiana Employment Security Division, 450 U.S. 707, 101 S.Ct. 1425, 67 L.Ed.2d 624 (1981), in which the Court explained, "Where the state conditions receipt of an important benefit upon conduct proscribed by a religious faith, ... thereby putting substantial pressure on an adherent to modify his behavior and to violate his beliefs, a burden upon religion exists. While the compulsion may be indirect, the infringement upon free exercise is nonetheless substantial.” Id. at 717-18, 101 S.Ct. 1425; accord Hobbie v. Unemployment Appeals Comm’n of Fla., 480 U.S. 136, 141, 107 S.Ct. 1046, 94 L.Ed.2d 190 (1987). These cases tell us that before Smith the government was not free to burden relí-gious exercise through less direct restraints than criminal sanctions or finés. They tell us nothing about the constitutional values in play when such a restraint is imposed by a third party who is, in turn, regulated by the government. In the cited cases, there were only two actors: the government and the individual protesting a burden on religious liberty. Therefore, pr e-Smith free exercise cases do not support creative exploitation of the ambiguity of the word "indirect” to make the government responsible for a burden imposed by another that the government does not lift.
. Our colleague's conclusion that the state action doctrine is not useful in cases where the government is directly sued, see Dissenting Op. at 73-74, is undermined by the Supreme Court’s use of the state action inquiry not only to determine whether a third party’s actions. should be held to government standards, see, e.g., Moose Lodge v. Irvis, 407 U.S. 163, 92 S.Ct. 1965, 32 L.Ed.2d 627 (1972), but also to determine whether the government, when challenged directly, should be held responsible for the actions of a third party, see, e.g., Burton v. Wilmington Parking Authority, 365 U.S. 715, 81 S.Ct. 856, 6 L.Ed.2d 45 (1961); Public Utils. Comm'n v. Pollak, 343 U.S. 451, 461-63, 72 S.Ct. 813, 96 L.Ed. 1068 (1952); see also Kitchens v. Bowen, 825 F.2d 1337 (9th Cir.1987). The dissent provides no alternative limiting principle, concluding only that "indirect” burdens on religious exercise are sufficient, see Dissenting Op. at 74, and that "approval” of a third party’s actions constitutes an indirect burden, see id. at 75-76. In a system of pervasive licensing and regulation by the federal government, our colleague's unbounded approach would mean that as individuals whose *64religious exercise was burdened by private parties grasped for a federal decision somewhere in the vicinity they could deem to be the "indirect” cause of the burden, every federal licensee would become the state for purposes of RFRA, as the government through its inaction granted "approval” to their decisions. The specter of endless application of strict scrutiny to private actions will not be illusory if those potential plaintiffs are as willing as the dissent to label government involvement as "extensive!],” id. at 73, or "intense!],” id. at 76-77, even when the private party invents, designs, advocates, and implements the action that imposes the burden on free exercise.
. In focusing on the fact that RFRA extends without limit because it "applies to all federal law, and the implementation of that law,” 42 U.S.C. § 2000bb-3(a), and that the FAA's approval of the City’s ALP as eligible for federal funding was "implementation of” a federal statute, the AAIA, the dissent begs the question whether the FAA's approval can be fairly characterized as responsible for the burden designed and imposed by the City. At no point does the dissent grapple with the question "whether government has placed a substantial burden on the observations of a central religious belief or practice,” Hernandez v. *66CIR, 490 U.S. 680, 699, 109 S.Ct. 2136, 104 L.Ed.2d 766 (1989), or whether some other actor should bear the blame for that burden. Rather than "try[ing] to plot a line between state action subject to ... scrutiny and private conduct (however exceptional) that is not,” Brentwood Academy v. Tenn. Secondary School Athletic Ass’n, 531 U.S. 288, 295, 121 S.Ct. 924, 148 L.Ed.2d 807 (2001), the dissent has eviscerated that line in toto. Fortunately, the Supreme Court has applied the state action inquiry in a number of contexts, noting that “examples may be the best teachers,” id. at 296, 121 S.Ct. 924, providing this court with guidance with which to address this question.