Joe G. Garcia, on Behalf of Himself and Others Similarly Situated v. San Antonio Metropolitan Transit Authority

E. GRADY JOLLY, Circuit Judge: Joe G. Garcia appeals the district court’s decision granting the motion of the San Antonio Metropolitan Transit Authority (“SAMTA”) for summary judgment and denying his own. For the following reasons, we affirm the judgment of the district court. I SAMTA is the governmental body that runs the public transit system in San Antonio, Texas. The plaintiffs (collectively, “Garcia”), employees of SAMTA, brought this action in 1979 for back pay and liquidated damages attributable to SAMTA’s *1412 violation of the Fair Labor Standards Act (“FLSA”). The historical context for the current appeal is unusually complex. In 1974, Congress amended the FLSA to repeal the exemption from the minimum wage and overtime provisions for all state and local government employees. In 1976, however, the Supreme Court curtailed the application of the FLSA to state and local governments, holding that principles of federalism prevented Congress under the commerce clause from regulating “integral operations in areas of traditional governmental functions.” National League of Cities v. Usery, 426 U.S. 833, 96 S.Ct. 2465, 2474, 49 L.Ed.2d 245 (1976). The Court did not attempt, however, to enumerate all the governmental activities that were exempt from federal regulation. Instead, the Court merely listed “fire prevention, police protection, sanitation, public health, and parks and recreation” as examples of protected functions. The Court did not address the status of government-owned mass transit systems. After the 1976 decision in National League of Cities, SAMTA notified its employees that it would no longer abide by the FLSA’s maximum hours and overtime provisions. The Department of Labor, which was charged with the enforcement of the FLSA, made no attempt to enforce the overtime provisions against any public transit system until September 17, 1979, when it informed SAMTA that it did not consider the operation of a public transit system a “traditional governmental function” protected under National League of Cities. Shortly afterward, SAMTA filed a declaratory judgment action against the Secretary of Labor, seeking reversal of the Department’s decision. The same day and in the same district court, Garcia brought the case now on appeal, an action for back pay and liquidated damages because of SAMTA’s claimed violation of the FLSA. 1 The district court stayed Garcia's case while the parties litigated SAMTA I. In 1981, the district court rendered summary judgment for SAMTA against the Secretary, holding that SAMTA’s operation of a public mass transit system was protected from federal regulation under National League of Cities. The Supreme Court, however, vacated the judgment and remanded the case for reconsideration in the light of the intervening decision of UTU v. Long Island Railroad, 455 U.S. 678, 102 S.Ct. 1349, 71 L.Ed.2d 547 (1982), which held that a railroad that carries passengers and freight is not protected by National League of Cities. In February 1983, however, the district court again ruled for SAMTA, adhering to its view that mass transit was a traditional governmental function and distinguishing Long Island Railroad. On appeal, the Supreme Court reversed, overruling National League of Cities and holding that Congress could constitutionally apply the FLSA to all state and local government employees. Garcia v. SAMTA , 469 U.S. 528, 105 S.Ct. 1005, 83 L.Ed.2d 1016 (1985). On remand to the district court, the parties agreed to dismiss SAMTA I and to litigate SAMTA II, which had been stayed during the entire litigation of SAMTA I. The dismissal allowed the Secretary of Labor to exit the dispute. With SAMTA I dismissed by agreement in the district court, both parties moved for summary judgment in Garcia’s original action. SAMTA argued that the Supreme Court’s decision in SAMTA I should not be applied retroactively under the facts of the case; therefore, the court should apply the old National League of Cities test to exempt SAMTA from liability. Garcia argued that SAMTA I should be applied, and that therefore SAMTA was liable for back pay and liquidated damages because of its violation of the FLSA. The district court granted SAMTA’s motion and denied Garcia’s. II In National League of Cities v. Usery, 426 U.S. 833, 96 S.Ct. 2465, 2474, 49 L.Ed. *1413 2d 245 (1976), the Supreme Court held that “insofar as the challenged [FLSA] amendments operate to directly displace the States’ freedom to structure integral operations in areas of traditional governmental functions, they are not within the authority granted Congress” by the commerce clause. As examples of traditional state functions exempt from federal regulation, the Court listed “fire prevention, police protection, sanitation, public health, and parks and recreation.” Id. The Court noted, however, that the examples it suggested were “obviously not an exhaustive cat-alogue of the numerous line and support activities which are well within the area of traditional operations of state and local governments.” Id. n. 16. It soon became apparent, however, that courts would have difficulty in applying National League of Cities consistently in deciding which state functions were “integral operations in areas of traditional [state] governmental functions.” As the Court later noted, Just how troublesome the task has been is revealed by the results reached in other federal cases. Thus, courts have held that regulating ambulance services, licensing automobile drivers, operating a municipal airport; performing solid waste disposal, and operating a highway authority, are functions protected under National League of Cities. At the same time, courts have held that issuance of industrial development bonds, regulation of intrastate natural gas sales, regulation of traffic on public roads, regulation of air transportation, operation of a telephone system, leasing and sale of natural gas, operation of a mental health facility, and provision of in-house domestic services for the aged and handicapped, are not entitled to immunity. We find it difficult, if not impossible, to identify an organizing principle that places each of the cases in the first group on one side of a line and each of the cases in the second group on the other side. Garcia v. SAMTA, 469 U.S. 528, 105 S.Ct. 1005, 1011, 83 L.Ed.2d 1016 (1985). Based in part on the practical difficulty in applying National League of Cities, the court in SAMTA I overruled that decision, holding instead that the states were adequately protected against unreasonable federal interference by their substantial role in constituting the federal government. Id. 105 S.Ct. at 1016-20. Congress reacted to the overturning of National League of Cities by amending the FLSA to eliminate retroactive liability for functions already categorized as “traditional,” but did not take a position on retroactivity as to other government functions such as the transportation function presented in this case. Fair Labor Standards Amendments of 1985, Pub.L. No. 99-150, 99 Stat. 787 (1985), §§ 2(c), 7. See also S.Rep. 159, 99th Cong., 1st Sess. 15 (1985); H.R.Rep. No. 331, 99th Cong., 1st Sess. 27-28 (1985), U.S. Code Cong. & Admin.News 1985, p. 651. In the present case, the district court refused to apply SAMTA I retroactively. Instead, using the analysis of Chevron Oil Co. v. Huson, 404 U.S. 97, 92 S.Ct. 349, 30 L.Ed.2d 296 (1971), the court concluded that SAMTA I should be applied prospectively only. Chevron identified three factors which courts should consider in deciding whether to recognize an exception to the general rule that the applicable law is that in effect at the time of the decision: First, the decision to be applied nonretro-actively must establish a new principle of law, ... by overruling clear past precedent on which litigants may have re-lied____ Second, it has been stressed that “we must * * * weigh the merits and demerits in each case by looking to the prior history of the rule in question, its purpose and effect, and whether retrospective operation will further or retard its operation.” Finally, we have weighed the inequity imposed by retroactive application, for “[w]here a decision of this Court could produce substantial inequitable results if applied retroactively, there is ample basis in our cases for avoiding the ‘injustice or hardship’ by a holding of nonretroactivity.” Id. 92 S.Ct. at 355. In applying the Chevron analysis, the district court found that the first factor favored nonretroactivity, because SAMTA I *1414 had overruled a clear past precedent upon which SAMTA had “reasonably relied.” The second factor favored neither, because SAMTA I would be fully enforced in the future regardless of the outcome of the decision to apply it retroactively. The third factor strongly favored nonretroactivity, as SAMTA had “structured its conduct based on the reasonable belief that its employees performed a traditional government function. Wage rates and benefit packages were negotiated based on this premise.” The complete Chevron test therefore led the district court to conclude that SAMTA I should not be applied retroactively. Garcia attacks the district court’s decision on three grounds. First, he argues that Chevron, which states a principle of mere federal common law, is inapplicable when Congress has by statute decided to apply the FLSA retroactively and that the FLSA, when read in conjunction with the Portal-to-Portal Act, 29 U.S.C. § 251 et seq., sets forth the only possible defenses to a FLSA violation, and that SAMTA has not fulfilled the requirements of any of these defenses. Second, he argues that Chevron is inapplicable because the parties in this case are the same as those in the Supreme Court’s SAMTA I decision, and Chevron cannot operate to deny to the victorious parties in SAMTA I the fruits of that victory. Last, he argues that even if Chevron is the proper standard, the district court erred in finding that Chevron requires nonretroactivity on the facts of this ease. Ill A. Garcia asserts that Chevron set forth a doctrine of federal common law concerning retroactivity of civil cases. As a doctrine of federal common law, it naturally must yield when Congress by statute has set a contradictory policy. See Milwaukee v. Illinois and Michigan, 451 U.S. 304, 101 S.Ct. 1784, 1790, 68 L.Ed.2d 114 (1981). And in the case of the FLSA, Garcia argues, Congress did set a contradictory policy. By its own terms, the FLSA has unquestionably applied to most state and local government employees since 1974, although from 1976 to 1984 National League of Cities prevented the full enforcement of the statute. Furthermore, this court has held that the Portal-to-Portal Act, 29 U.S.C. § 259, supplies the only defenses to liability for FLSA violations. Le-Compte v. Chrysler Credit Corp., 780 F.2d 1260, 1262 (5th Cir.1986). SAMTA does not assert any defense based on that act. A problem with Garcia’s argument is that there is little support for the proposition that Chevron, as a doctrine of federal common law, can never apply when a judicial decision changes the reach of a congressional statute. Garcia does cite one district court opinion, Joiner v. City of Macon, 627 F.Supp. 1532, 1535 (M.D.Ga.1986), rev’d in part on other grounds, 814 F.2d 1537 (11th Cir.1987), 2 which states that the Chevron factors “are a compendium of federal common law principles for use in the absence of federal statutory provisions.” Joiner’s statement, however, is dictum in view of the court’s additional conclusion that the Chevron factors compel retroactivity of SAMTA I. Id. at 1536. The proposition is further undermined by the fact that every other court to consider the issue of SAMTA 7’s retroactivity has employed the Chevron framework. See, e.g., Mineo v. Port Authority of New York and New Jersey, 779 F.2d 939, 944 (3d Cir.1985) (finding SAMTA I nonretroac-tive); Kukla v. Village of Antioch, 647 F.Supp. 799, 815-16 (N.D.Ill.1986) (same); *1415 Thurmond, v. City of Union City, 628 F.Supp. 146, 148 (W.D.Tenn.1986) (same). Garcia’s argument also suffers from problems of illogic. Even if Chevron is a doctrine of federal common law, there is no statute that sets a Congressional policy on retroactivity contrary to Chevron. Here, as in most cases when the Court decides to overrule a precedent, the Constitution and statutes fail to address the problem of re-troactivity. As a result, courts must employ doctrines such as Chevron to provide a solution. If Congress had addressed the retroactivity issue regarding the mass transit function, then Garcia’s argument that a statute should override the Chevron common law doctrine would have some force. 3 A further difficulty with Garcia’s argument is that it would never allow a decision holding a new rule of constitutional or statutory interpretation nonretroactive. The new rule, as the definitive expression of the meaning of either the Constitution or a statute, would naturally override any common law principles favoring nonretroactivity. The court would be limited under Chevron to considering the retroactivity of new rules of federal common law. But Supreme Court precedent belies that result. In some cases, the Court has applied Chevron to limit the retroactivity of cases announcing new principles of constitutional law. See, e.g., Lemon v. Kurtzman, 411 U.S. 192, 93 S.Ct. 1463, 36 L.Ed.2d 151 (1973) which refused to apply retroactively an earlier decision holding unconstitutional state reimbursement of religious schools for some expenses. Other cases have denied retroactivity in statutory construction cases. See, e.g., Arizona Governing Committee v. Norris, 463 U.S. 1073, 1075, 103 S.Ct. 3492, 3494, 77 L.Ed.2d 1236 (1983) (Title VII of the 1964 Civil Rights Act); Los Angeles Dep’t of Water & Power v. Manhart, 435 U.S. 702, 719-23, 98 S.Ct. 1370, 1380-83, 55 L.Ed.2d 657 (1978) (same). These decisions indicate that “all questions of civil retroactivity [are] governed by the standard enunciated in Chevron Oil,” United States v. Johnson, 457 U.S. 537, 102 S.Ct. 2579, 2594, 73 L.Ed.2d 202 (1982), whether they concern statutes, the common law, or even the Constitution. We therefore hold that Chevron provides the standards we must apply in this case. B. Garcia argues that, because the parties in the present case are the same as those in SAMTA I, he should not be denied the fruits of his victory in the Supreme Court by a decision not to apply that decision retroactively. He cites two cases to support his position. In Noble v. Drexel, Burnham, Lambert, Inc., 823 F.2d 849 (5th Cir.1987), this court applied the Chevron test to a recent Supreme Court decision concerning arbitration of securities disputes, Shearson/American Express, Inc. v. McMahon, — U.S. -, 107 S.Ct. 2332, 96 L.Ed.2d 185 (1987). We held the decision retroactive. But before doing so, we noted a “strong irony” in retroactivity analysis: Unavoidably, the losing party in a case which establishes new law sustains a full blast of “retroactivity,” otherwise the change in law is merely precatory and without full precedential force. The second loser, however, gets the potential benefits of the analysis suggested by Chevron Oil. The McMahons lost first and without hope of escape. We must *1416 determine whether Mr. Noble is to suffer the same fate. 823 F.2d at 850. In Simpson v. Union Oil Co., 369 U.S. 13, 90 S.Ct. 30, 24 L.Ed.2d 13 (1969), the Supreme Court expressed a similar view. There, the Court in an earlier case had established a new antitrust rule concerning price fixing by the “consignment” device. Simpson v. Union Oil Co., 377 U.S. 1051, 84 S.Ct. 1051, 12 L.Ed.2d 98 (1964). In the earlier case, the Court had reserved the question of whether the equities warranted only prospective application of its new rule. On remand, the district court interpreted this comment as authorizing it, after a consideration of the equities, to deny the benefits of the new rule to the litigant before the Supreme Court, Union Oil. After the appeal from the remand, the Court ruled: The question we reserved was not an invitation to deny the fruits of successful litigation to this petitioner____ Formulation of a rule of law in an Article III case or controversy which is prospective as to the parties involved in the immediate litigation would be most unusual, especially where the rule announced was not innovative. The Supreme Court’s holding in Simpson and the Fifth Circuit’s dictum in Noble suggested that retroactivity analysis does not apply to the parties to the case that establishes the new rule. Simpson, however, was decided before the Court’s decision in Chevron established a consistent framework for retroactivity analysis. In addition, the court’s comments in Noble were dicta, as that case did not concern the parties to the case that established the new rule. Furthermore, the tone of the cited passage indicates that this court found it inequitable that only one party would suffer from the new rule of law. Most important, since Chevron, other cases have applied the Chevron framework even to the parties to the case that established the new rule. In Lemon v. Kurtzman, 403 U.S. 602, 91 S.Ct. 2105, 29 L.Ed.2d 745 (1971), for example, the Court invalidated Pennsylvania’s program of reimbursement of religious schools for providing secular educational services. On remand to the district court after the Supreme Court’s decision, however, the court allowed the state to reimburse the schools for services provided before the date of the Lemon decision, enjoining only further payments for services performed after that date. The Supreme Court affirmed, though without an opinion for the Court. Chief Justice Burger, announcing the judgment of the Court and the opinion of a four-justice plurality, cited Chevron in his conclusion that equitable factors warranted the nonretroac-tive application of the original Lemon ruling, even to the very parties in that case. 411 U.S. 192, 93 S.Ct. 1463, 1467-72, 36 L.Ed.2d 151 (1973). In Northern Pipeline Construction Co. v. Marathon Pipe Line Co., 458 U.S. 50, 102 S.Ct. 2858, 2880, 73 L.Ed.2d 598 (1982), the Supreme Court invalidated the broad grant of jurisdiction to the bankruptcy courts, but, under the Chevron framework, applied the judgment only prospectively to the parties in the case. Allen v. State Bd. of Elections, 393 U.S. 544, 89 S.Ct. 817, 835, 22 L.Ed.2d 1 (1969), held changes in Mississippi election law invalid, but refused to set aside elections already held under the law. Zambuto v. AT & T, 544 F.2d 1333, 1336 (5th Cir.1977), held that Chevron equitable factors compelled making its ruling prospective only from date of decision plus 90 days, even as to the parties in the case. Thus, we conclude that whether a decision should be made retroactively as to the parties in the original case to guarantee the prevailing party the “fruits of its successful litigation” is a factor to be considered under the third prong of the Chevron test: the weighing of the equities. IV Although in civil litigation, “[t]he usual rule is that federal cases should be decided in accordance with the law existing at the time of decision,” Goodman v. Lukens Steel Co., — U.S. -, 107 S.Ct. 2617, 2621, 96 L.Ed.2d 572 (1987), Chevron set up a system to evaluate when nonretroac-tivity is proper, and included three factors *1417 to be considered in making that decision. The district court in this case concluded that the first Chevron factor — whether the new rule overruled “clear past precedent on which litigants may have relied” — favored nonretroactivity. The second factor —whether retroactive application will further or retard the operation of the new rule — the district court found to be neutral. Finally, the district court found that the third factor — whether retroactive application could produce substantial inequitable results — strongly favored nonretroactivity. Garcia has attacked each of these conclusions. A. There is no question that SAMTA I overruled a clear past precedent: “National League of Cities v. Usery is overruled.” [Citations omitted], 105 S.Ct. at 1021. The real issue presented is whether SAMTA was entitled to rely on the overruled case in its decision not to follow the requirements of FLSA. See Goodman v. Lukens Steel Co., 107 S.Ct. at 2621, which held that the first Chevron factor will favor nonre-troactivity if the case overruled “clear ... precedent on which the complaining party was entitled to rely.” We feel that the facts of this particular case prevent our making a straightforward assessment of this factor; however, all of the Chevron factors rely on equitable considerations which allow us to extrapolate from the principles underlying them and to apply them to the unwieldy facts here. A straightforward assessment of this factor would be possible if we could look at National League of Cities and decide whether SAMTA had been entitled to rely on that case in refusing to follow the FLSA because that case had provided that mass transit was exempt from the FLSA. National League of Cities did not mention mass transit among the governmental functions which were exempt from FLSA requirements, however. Instead, National League of Cities listed a few examples, such as police and fire protection, and then stated that traditional governmental functions such as these were exempt from the FLSA while nontraditional state functions were not exempt. Thus, whether SAMTA was entitled to rely on National League of Cities in failing to follow the FLSA depends upon whether public mass transit is a traditional state function or a nontraditional one. Several appellate decisions which reached this issue have held that National League of Cities did not protect mass transit, because that function was not traditional. 4 Additionally, the Labor Department passed a regulation effective December 21, 1979, which determined that local public mass transit system is not a “traditional” public function, and thus is not exempted from the FLSA. 29 C.F.R. § 775.2(b)(3). It was this regulation that prompted SAMTA to file this suit for a declaratory judgment. SAMTA did have, however, numerous persuasive arguments why mass transit should have been categorized as a traditional government function and thus exempted from the FLSA. It had enough arguments, in fact, to convince the district court twice that it was performing a “traditional” function. Finally, it convinced four members of the Supreme Court in SAMTA I, (105 S.Ct. at 1021, Powell, J. dissenting). The majority opinion was silent on this question. Garcia argues that having only a credible position under a former precedent is not sufficient to fulfil the first prong of Chevron. Rather, Garcia argues that to find that SAMTA had fulfilled this Chevron factor in the usual manner, we would have to decide that mass transit in San Antonio was a traditional function under the National League of Cities doctrine. It is impossible for us to make such a determination one way or the other, however, because the Supreme Court has explained that the National League of Cities doctrine was “unsound in principle and unworkable in practice.” 105 S.Ct. at 1016. In that opinion, the Court spends pages showing how cases under the National *1418 League of Cities doctrine had reached inconsistent results and that there was no way “to identify an organizing principle that places each of the cases in the first group on one side of a line and each of the cases in the second group on the other side.” 105 S.Ct. at 1011. It would therefore be highly ironic for us to sit as a court of appeals following SAMTA I and attempt to determine whether mass transit falls on one side or the other of a line that the Supreme Court has declared does not exist and, indeed, has never existed. We therefore think it inappropriate after SAMTA I to try to determine whether SAMTA in particular, and mass transit in general, fulfills a “traditional” public function and was therefore exempt from FLSA prior to SAMTA I. With respect to the first prong of the Chevron test we can only say that clear precedent was overruled, and the test structure of that precedent, upon which SAMTA relied, was overruled and declared incapable of serving as a basis for reliance. This is the incongruous situation that makes our applying the Chevron test more difficult than in many preceding cases. We believe, however, that the equitable nature of the Chevron test gives it enough flexibility to allow us to reach a conclusion as to whether SAMTA 1 should be applied retroactively. Again, there is no doubt that prior to SAMTA I, SAMTA had a persuasive argument that it was exempt from the FLSA. SAMTA was relying on the general framework provided by National League of Cities, as it undoubtedly had a right to do. SAMTA also had a right to pursue its claim that it was exempt from the FLSA through the courts, and to the Supreme Court if necessary, to obtain a final resolution. Once it had obtained a district court decision in its favor, SAMTA definitely had a right to rely on that decision unless it was overturned. To be sure, the district court’s decision that SAMTA served a traditional function of government was never overturned. Although SAMTA pursued its right to litigate the question of its status as “traditional” or “nontraditional,” it never did obtain a final answer because, before it was able to do so, the Supreme Court swept away the framework upon which the question was to be decided. The special fact that distinguishes our case from other retroactivity cases is that not only did the Supreme Court sweep away the framework, but it also declared that framework illogical and unworkable. Thus, since the framework was illogical and unworkable from the start, we cannot go back and routinely determine under the first prong of Chevron whether mass transit would have been exempted from the FLSA under the old doctrine. We believe that this radical elimination of the entire test framework upon which SAMTA’s status would have been determined fulfills the first prong of Chevron — that the case to be applied nonretroactively “must establish a new principle of law ... by overruling clear past precedent on which litigants may have relied____” Chevron, 404 U.S. 97, 92 S.Ct. 349, 355, 30 L.Ed.2d 296 (1971). B. The second prong of the Chevron test requires examination of whether retroactivity will “ ‘further or retard [the] operation’ of the holding in question.” Northern Pipeline Construction Co. v. Marathon Pipe Line Co., 458 U.S. 50, 88, 102 S.Ct. 2858, 2880, 73 L.Ed.2d 598 (1982). The district court found that this prong of the test was neutral as far as deciding whether retroactive application should govern. The district court found that backpay awards were unnecessary to compel future FLSA compliance. We agree that since all state and local governments are now aware that National League of Cities has been overruled, and that they must comply with the FLSA in the future, “[t]here is no reason to believe that the threat of a backpay award is needed to cause [state and local governments] to amend their practices to conform to [the Supreme Court’s decision].” Los Angeles Dep’t of Water & Power v. Manhart, 435 U.S. 702, 720-21, 98 S.Ct. 1370, 1381-82, 55 L.Ed.2d 657 (1978). The Third Circuit, which denied retroactivity of SAM-TA I ms. case involving the Port Authority, reached a similar conclusion: Now that the Supreme Court has ruled explicitly that no state employees are *1419 exempt from FLSA, the Port Authority maintains, there is no reason to believe that state and municipalities will not comply in the future. We agree. The [SAMTA /] decision makes the law clear that, in the future, states must comply with FLSA. Regardless of how we decide this case, there is no reason to suspect that states would refuse to be bound by FLSA. This situation leaves us free to decide the instant case on its facts and equitable principles without concern for furthering or retarding the operation of [SAMTA /]. The second Chevron factor neither favors nor opposes the retroactive application of the [SAMTA 7] decision. Mineo v. Port Authority of New York & New Jersey, 779 F.2d 939 (3d Cir.1985). Indeed, SAMTA also argues that retroactive FLSA liability might in fact frustrate the SAMTA I decision by placing such a burden on municipal and state governments that such governmental functions, and therefore jobs, might have to be eliminated. We think, however, that this argument is best dealt with under the third prong of the Chevron test, that is, the equities of retroactivity versus nonretroac-tivity. C. The district court in this case found that the third Chevron factor compelled nonre-troactivity: [T]he third factor which requires consideration of the inequities caused by retroactive application of [SAMTA 7], compels nonretroactivity of the new rule of law and FLSA coverage____ SAMTA structured its conduct based on the reasonable belief that its employees performed a traditional government function. Wage rates and benefit packages were negotiated based on this premise____ Having negotiated this package, additional back pay liability would represent a windfall. As the Third Circuit observed in Mineo v. Port Authority of New York & New Jersey, 779 F.2d at 946, the courts cannot fairly punish those who are forced to make decisions without immediate guidance from the judiciary and who do so reasonably. This is particularly so where the consistent opinions of this district court found this employment relationship outside FLSA protection. The court need not expressly adopt the cries of “doom and gloom” to taxpayers, riders and recently hired public employees to agree that the third Chevron factor mandates nonretroactivity. While we recognize that SAMTA employees have not been compensated for overtime work in compliance with the FLSA, we find compelling the fact that SAMTA has developed its structure of wages and benefits through meetings with the employees’ union representative and that this structure included six types of overtime pay, and took into account the nonapplicability of the FLSA’s overtime requirements. The Third Circuit explained in Mineo why this background makes retroactive overtime pay inequitable: Any public or private organization must manage its revenues to most efficiently provide services at the lowest cost. When involved in labor negotiations, the organization poses estimates of how many hours it thinks the employees will work and how much money it has to compensate them. Within those parameters, the organization may opt for various pay structures. For example, some employees may be paid more than others; some compensation may be deferred; or employees may get a higher base pay in return for reduced overtime pay. It appears that the last situation was present in the instant case. Reasonably believing itself to be unshackled from the restrictions of FLSA, the Port Authority offered an attractive base pay that was balanced by lower overtime compensation. The Detectives agreed to this arrangement. The retroactive application of [SAMTA 7] to this situation would give the Detectives increased overtime pay without any reduction in base pay. To allow the Detectives to get a pay raise premised on retroactive application of an unforeseen decision that was made almost two years after the end of the contract period would be inequitable to *1420 the Port Authority and would constitute a windfall to the Detectives. Mineo, 779 F.2d at 946. See also Morrison, Inc. v. Donovan, 700 F.2d 1374, 1376 (11th Cir.1983); Ramey v. Harber, 589 F.2d 753, 760 (4th Cir.1978), cert. denied, 442 U.S. 910, 99 S.Ct. 2823, 61 L.Ed.2d 275 (1979). Thus, we find that the equities favor nonretroactivity as regards SAMTA for two basic reasons. First, huge back pay awards may very likely cripple the always closely budgeted municipal funds. There is a great difference between, on the one hand, knowing you are subject to a regulation, planning for it and taxing and charging the public for it all along, and, on the other hand, suddenly finding out that, contrary to precedent of the Supreme Court and decisions of the district court, you have been subject to a regulation for several years and are required to compensate under that regulation for previous years for which tax monies have already been otherwise allocated and spent. i Of course, the municipalities and the taxpayers have in some ways benefitted from their exemption from the FLSA. However, the equities do not justify saddling them with huge back-pay liabilities when they acted in good faith during all those years, believing for substantial precedential reasons that they were exempt from the FLSA, especially in the light of the next considered equity. The second equity compelling nonretroac-tivity of the SAMTA I decision is that what SAMTA employees have not received in overtime compensation over the past several years, they have received at least partially in other forms. SAMTA must compete for labor in the marketplace against employers who are subject to the FLSA, and so it could not simply subtract the benefits required by the FLSA from its own programs. It would be impossible at this point to calculate how many other benefits SAMTA employees have received instead of these overtime benefits and to subtract the amount from a backpay liability. V Thus, after consideration of all three prongs of the Chevron test, we find that the retroactive application of SAMTA I should be denied. We therefore affirm the judgment of the district court. AFFIRMED. Footnotes [1] The district court allowed Garcia, the plaintiff in the second case, to intervene in SAMTA’s action against the Secretary. Thus, both cases confusingly bear the name Garcia v. SAMTA. For that reason, we call the case that SAMTA brought against the Secretary of Labor “SAMTA /” and the case that Garcia brought, “SAMTA II, 557 F.Supp. 445." [2] Another source of possible support for this argument is the opinion of Judge Gibbons dissenting from the denial of rehearing en banc. Mineo v. Port Authority of New York and New Jersey, 783 F.2d 42, 42 (3d Cir.1986). A Third Circuit panel had decided originally that Chevron prevented SAMTA I from being applied retroactively in certain circumstances. 779 F.2d 939 (3d Cir.1985). Judge Gibbons would have granted rehearing en banc because "[t]he majority relies on federal common law, of which Chevron Oil Co. v. Huson is a part, for its rule of decision. Federal common law, however, never can be applied in contravention of the will of Congress expressed in a duly enacted statute." 783 F.2d at 43. The en banc court, of course, did not adopt Judge Gibbons’ position. [3] Garcia does argue that the 1985 amendments to the FLSA set a Congressional policy on re-troactivity. Those amendments did eliminate retroactive liability for the state functions that the Department of Labor's regulations had deemed "traditional” and thus exempt from the FLSA under National League of Cities. But section 7 of the amendments specifically states that the 1985 amendments "shall not affect whether a public agency ... is liable under ... [FLSA for a violation] occurring before April 15, 1986 [the date of SAMTA /], with respect to any employee of such public agency who would have been covered by such Act under the Secretary of Labor’s special enforcement policy on January 1, 1985____” Both the House and Senate Committee reports on the 1985 amendments also contain statements that Congress’ intent was not to take a position on the retroactivity of SAMTA I, except to eliminate liability for state agencies that were exempt from the FLSA under the Labor Department regulations. S.Rep. 159, 99th Cong., 1st Sess. 15 (1985); H.R.Rep. 331, 99th Cong., 1st Sess. 27-28 (1985), U.S. Code Cong. & Admin.News 1985, p. 651. [4] Dove v. Chattanooga Area Regional Transportation Authority, 701 F.2d 50 (6th Cir.1983); Alewine v. City Council, 699 F.2d 1060 (11th Cir.1983); Kramer v. Newcastle Area Transit Authority, 677 F.2d 308 (3d Cir.1982); Francis v. Tallahassee, 424 So.2d 61 (Fla.App.1982).