dissenting.
This is a close case because it tests the outer limits of ERISA preemption. Although my decision is made more difficult because the New Jersey regulatory scheme (the “Act”) is admirable for its intended purpose and goals, I think Congress intended to preempt these kinds of statutes. The issue is whether a statute of purported general applicability “relates to,” in the ordinary and broad sense of that term, ERISA plans if it disproportionately impacts upon ERISA plans, presupposes the existence of ERISA plans, and depends on funds extracted from ERISA plans to implement its legislative mandate. Because I believe that the district court correctly concluded that such a statute is preempted, see United Wire, Metal & Machine Health and Welfare Fund v. Morristown Memorial Hosp., 793 F.Supp. 524, 531-37 (D.N.J.1992), I respectfully dissent.
I.
Congress enacted ERISA to subject employee benefit plans to a uniform system of federal laws governing disclosure, reporting, *1197standards of conduct, remedies, sanctions, and access to federal courts. Since uniformity cannot be achieved if ERISA plans are subject to varying state regulations, Congress preempted “any and all State laws insofar as they ... relate to any employee benefit plans.” ERISA § 514(a), 29 U.S.C. § 1144(a) (emphasis added).
Section 514(a) is deliberately expansive and “conspicuous for its breadth.” FMC Corp. v. Holliday, 498 U.S. 52, 58, 111 S.Ct. 403, 407, 112 L.Ed.2d 356 (1990). It is “virtually unique” among federal preemption statutes, Franchise Tax Bd. v. Construction Laborers Vacation Trust, 463 U.S. 1, 24 n. 26,103 S.Ct. 2841, 2854 n. 26, 77 L.Ed.2d 420 (1983), as it is “one of the broadest preemption clauses ever enacted by Congress.” Evans v. Safeco Life Ins. Co., 916 F.2d 1437, 1439 (9th Cir.1990).
The term “relate to” must be given a “broad common-sense meaning.” Pilot Life Ins. Co. v. Dedeaux, 481 U.S. 41, 47, 107 S.Ct. 1549, 1553, 95 L.Ed.2d 39 (1987). A state law relates to an ERISA plan “in the normal sense of the phrase, if it has a connection with or reference to such a plan.” Shaw v. Delta Air Lines, Inc., 463 U.S. 85, 97, 103 S.Ct. 2890, 2900, 77 L.Ed.2d 490 (1983). Any connection may trigger preemption, and preemption is not limited to laws relating to the specific subjects covered by ERISA Pilot Life, 481 U.S. at 47-48, 107 S.Ct. at 1553; Shaw, 463 U.S. at 97, 103 S.Ct. at 2900. That a state law may be “consistent with ERISA’s substantive requirements” or was enacted to “effectuate ERISA’s underlying purposes” does not save it from preemption. Metropolitan Life Ins. Co. v. Massachusetts, 471 U.S. 724, 739, 105 S.Ct. 2380, 2388-89, 85 L.Ed.2d 728 (1985); Mackey v. Lanier Collection Agency & Serv., 486 U.S. 825, 829, 108 S.Ct. 2182, 2185, 100 L.Ed.2d 836 (1988). A state law may relate to a benefit plan even if it is not specifically designed to affect such plan or its effect is only indirect. Pilot Life, 481 U.S. at 47, 107 S.Ct. at 1552-53; Shaw, 463 U.S. at 97, 103 S.Ct. at 2900; Alessi v. Raybestos-Manhattan, Inc., 451 U.S. 504, 524-25, 101 S.Ct. 1895, 1907, 68 L.Ed.2d 402 (1981); Ingersoll-Rand Co. v. McClendon, 498 U.S. 133, 139, 111 S.Ct. 478, 483, 112 L.Ed.2d 474 (1990).
Since no law exists in a vacuum and arguably many laws could be held to “relate to” ERISA plans, without some limits Section 514(a) could become a legal blackhole with an attractive force no state law could resist. Hence, some laws are said to affect ERISA plans in “too tenuous, remote, or peripheral a manner to warrant a finding that [they] ‘relate to’ the plan.” Shaw, 463 U.S. at 100 n. 21, 103 S.Ct. at 2901 n. 21. See, e.g., Mackey, 486 U.S. at 840-41, 108 S.Ct. at 2191 (garnishment statute of general applicability is not preempted). ,The task then is to determine the precise relationship between the Act and ERISA plans. Mackey, 486 U.S. at 830, 108 S.Ct. at 2186.
II.
The declared public policy of the Act is to “contain the rising costs of health care services, and to ensure the financial solvency of hospitals.” N.J.S.A. § 26:2H-1. The linchpin in this scheme is the DRG rate, which is a statistical average of costs incurred by hospitals to perform a particular medical procedure. The DRG is the rate hospitals must bill for particular services. Thus, it functions as both a price cap and an incentive to reduce costs.
The plans do not challenge the validity of the DRG rate scheme, but only attack three surcharge components incorporated into the DRG. They are costs for uncompensated care and bad debts, Medicare cost shifts, and costs related to discounts given to certain payer groups. Only hospital patients, not the general public, pay these costs. The New Jersey scheme is; quite simply, a cost-shifting mechanism whereby those who do not have deep pockets of are favored by the law are given discounts or free services. Since these discounts and services cause state-mandated financial losses to hospitals, the Act remunerates hospitals by shifting these losses through the DRG, resulting in additional 19.7% and 7% surcharges in uncompensated and Medicare cost shifts to those with deep pockets — commercial insurers and ERISA plans.
*1198ERISA plans play such a critical role in this scheme that without them it would fall apart. They comprise a major segment of the paying hospital service users who subsidize those favored by the Act. The structure of the scheme indicates in not-so-subtle ways that the New Jersey legislature was well aware of the pivotal role ERISA plans would play in its scheme. Upon pain of not receiving reimbursements for their uncompensated care costs, the Act requires hospitals to screen patients to determine whether they are covered by a commercial insurer or a “union welfare plan.” N.J.S.A. § 26:2H-18.-31(c)(2). Also, under the appeal policy existing at the time, while individual patients as well as third-party payors could have appealed mistakes in the assignment of a particular DRG, only individual patients could have appealed an assignment of a DRG that was technically correct but may have been excessive. The third-party payer (commercial insurer, self-funded, union, etc.) was specifically excluded from appealing a correct but excessive charge. A vast majority of appeals, about 85%, were based on excessive charges, and only about 1% of them were denied. Thus, the goal is transparent and little explanation is needed to see what this appeal process was designed to do.
In October 1990, the New Jersey Governor’s Commission on Health Care Cost submitted a report that plainly concluded: “Only a portion of New Jersey businesses, those who purchase insurance for their workers, pay the lion’s share of caring for the uninsured.” It summarized the impact on employee benefit plans: “It was apparent that rapidly rising health insurance costs were a significant burden to both the business community and the labor force in this State with the potential to negatively affect New Jersey citizens, ... [and] that the Uncompensated Care Trust Fund while affording access to hospital services was unfairly financed on the backs of those who had health insurance .... ”
Last, the State concedes that the scheme is not “viable” without money from the plans as they “represent a major segment of the bill-paying public, to the point that it is impossible to devise a viable hospital rate-setting scheme if that segment is excluded or exempted from the scheme.” Br. 2 (emphasis added). This concession dispels any doubt that the effect of the Act is to reach into the deep pockets of ERISA funds.
The Act provides a conduit by which money is transferred from, among others, ERISA plans to hospitals. Rather than spend its own general funds, New Jersey implemented a money transfer scheme where ERISA plans subsidize the medical bills of those who are favored by law. One affidavit accurately concluded: “Indeed, if the State of New Jersey were to .assume the uncompensated care and the Medicare cost shift as social obligations from its general revenues, for example, the average hospital bill would decrease on a pro rata basis by this amount and these shifts would no longer be reflected in New Jersey hospital bills.”
III.
The majority believes that the connection between the Act and ERISA plans is too tenuous and remote “[b]ecause we are here dealing with a statute of general applicability that is designed to establish the prices to be paid for hospital services, which does not single out ERISA plans for special treatment, and .which functions without regard to the existence of such plans....” Majority at 1192. The majority relies in part on Mackey v. Lanier Collection Agency & Serv., 486 U.S. 825, 108 S.Ct. 2182, 100 L.Ed.2d 836 (1988), to suggest that where there is no direct nexus, or where a statute does not directly affect the administration of ERISA plans, the statute is not preempted despite any “indirect ultimate effect of increasing plan costs.” Majority at 1193-94.
In Mackey the issue was whether a generally applicable Georgia garnishment statute was preempted by ERISA. The Court first recognized that a statute need not specifically single out, mention, or have a direct nexus to ERISA plans to be preempted. 486 U.S. at 831, 108 S.Ct. at 2186, citing Pilot Life, 481 U.S. at 46-47, 107 S.Ct. at 1552-53, and Shaw, 463 U.S. at 97, 103 S.Ct. at 2900. The Court held, however, that the statute was not preempted. In light of certain ERISA provisions, one of which provides that a plan may “sue or be sued” as an entity, 29 U.S.C. *1199§ 1132(d)(1), the Court reasoned that money judgments against ERISA plans must be collectable in some way and that garnishment is one permissible method. 486 U.S. at 834 & n. 9, 108 S.Ct. at 2188 & n. 9, citing FHA v. Burr, 309 U.S. 242, 246-47, 60 S.Ct. 488, 491, 84 L.Ed. 724 (1940) (where Congress provides that an entity may “sue or be sued,” all civil processes incident to legal proceedings including “garnishment and attachment” may apply). Thus, Mackey does not stand for the proposition that a generally applicable statute should not be preempted even though it may result in some “indirect economic impact.” It stands for the proposition that where Congress intended for state law to apply to ERISA, that law will not be preempted even though it may otherwise relate to ERISA plans by placing indirect financial or administrative burdens on them.1
The majority nowhere shows how any provision in or the structure of ERISA suggests that Congress did not intend these kinds of statutes to be preempted. See Malone v. White Motor Carp., 435 U.S. 497, 504, 98 S.Ct. 1185, 1190, 55 L.Ed.2d 443 (1978) (“purpose of Congress is the ultimate touchstone”). Where Congress adopts a broad preemption provision, the “task of discerning congressional intent is considerably simplified.” Ingersoll-Rand, 498 U.S. at 138, 111 S.Ct. at 482.
Although any connection may suffice, the task is made easier when the statute refers to or specially singles out ERISA plans. The Act requires hospitals to ask patients whether they belong to a “union welfare plan,” and it expressly excludes “self-funded union” plans from appealing hospital bills. The majority, however, brushes these references aside by concluding that they “can be excised without altering the effect of [the] statute in any way,” and therefore they “should be regarded as without legal consequence for § 514(a) purposes.” Majority at 1192 n. 6.
The majority underestimates the significance of these references. New Jersey has implicitly classified all hospital users into two groups: those who are able to pay and those who are not. This simple dichotomy then determines “whether the patient is eligible for participation in a public assistance program,” N.J.S.A § 26:2H-18.31(c)(3), whether one must pay the three surcharges, and whether and what one may appeal. A large segment of the small percentage of patients who are able to fully pay for medical services are plan beneficiaries, and hence New Jersey requires hospitals to ask whether a patient belongs to a “union welfare plan.” Thus, the conclusion that the references to ERISA plans in the Act “can be excised without altering the effect of [the] statute in any way” is without support.
Even assuming, arguendo, that the Act does not expressly single out ERISA plans for special treatment, it does so as applied. One cannot ignore the practical consequences produced by this statute, a consequence that according to the report commissioned by the State resulted in ERISA plans having paid “the lion’s share of caring for the uninsured.” See Mackey, 486 U.S. at 829, 108 S.Ct. at 2185 (it is “virtually taken ... for granted that state laws which are ‘specifically designed to affect employee benefit plans’ are pre-empted under § 514(a)”).
Moreover, the assumption that the Act “functions without regard to the existence of such plans” is puzzling in light of the State’s concession that the Act is not “viable” without ERISA plans. The Act was designed with ERISA funds in mind. In this sense, it has been eminently successful; the financial drain on ERISA funds has been enormous. The surcharges are paid not by the general public at large, but by the less than 25% of the population who use hospital services. Of those 25%, about 75% receive the uncompensated care assessments challenged here. ERISA plan participants comprise only *1200about 15% of the hospital patients, but pay about 40% of the more than $1.1 billion shortfall generated by the state-mandated cost shifts.
In Ingersoll-Rand, Co. v. McClendon, 498 U.S. 133, 111 S.Ct. 478, 112 L.Ed.2d 474 (1990), an employee sued in Texas state court, alleging that he was fired because his employer did not want to contribute to his pension fund. He sought damages under various tort and contract theories, but did not plead an ERISA cause of action. The Texas Supreme Court recognized a cause of action for wrongful discharge based upon “the employer’s desire to avoid contributing to or paying benefits under the employee’s pension fund.” The Supreme Court reversed and held that the cause of action was preempted because it was premised on the existence of a pension plan. It reasoned that “to prevail, a plaintiff must plead, and the court must find, that an ERISA plan exists and the employer had a pension-defeating motive in terminating the employment. Because the court’s inquiry must be directed to the plan, this judicially created cause of action ‘relates to’ an ERISA plan.” Id. at 140, 111 S.Ct. at 483.
Similar to the Texas cause of action, the New Jersey statute is predicated on the existence of ERISA plans and its mandates cannot be carried out without ERISA funds. See id. at 140, 111 S.Ct. at 484 (“there simply is no cause of action if there is no plan”) (emphasis in original). Also, the costs associated with the New Jersey regulatory scheme are not qualitatively the same as those in either Mackey or Ingersollr-Rand. In Mack-ey, the four dissenting Justices opined that the administrative burdens of allowing ERISA benefits to be garnished were so “significant” as to warrant preemption of a generally applicable garnishment statute. 486 U.S. at 842, 108 S.Ct. at 2192. In Ingersolh-Rand, a unanimous Supreme Court noted that the litigation costs of defending against a cause of action predicated upon the existence of an ERISA plan were substantial. 498 U.S. at 139-40, 111 S.Ct. at 483. No one will dispute that these costs pale in comparison to the hundreds of millions of dollars extracted from ERISA funds under the Act. And while the majority may classify these costs as “the limited nature of the statute’s impact on such plans,” I do not.
One may argue that nothing in the Act establishes the level of benefits or structures plan benefits to include these shifted costs and that each plan is free to cover all, some or none of these costs. This argument is unpersuasive. In General Elec. Co. v. New York Dep’t of Labor, 891 F.2d 25 (2d Cir. 1989), a New York labor law of general application provided that wages and “supplements” (nonwage benefits which included ERISA plans) on a public works contract must at least equal the prevailing rate and benefits paid in the locality. Where the cost of a benefit provided by an ex-locality contractor did not match those of a similar prevailing local benefit, the statute required the contractor either to bring the benefit into conformity with the local benefit, or to make up the difference through cash payments to its employees. Nothing in the statute, however, required a contractor to alter the benefits to match those in the locality, and the contractor was free to pay cash instead. But the Court of Appeals for the Second Circuit held that ERISA preempted the labor law. It reasoned that the law clearly related to ERISA plans in that to conform to the statute contractors had to maintain “schedules of supplements and wages and to make its books and records pertaining to wages, supplements and hours of labor available for inspection.” Id. at 29-30.
The Act is analogous to the New York labor statute because it requires ERISA plans either to pay the surcharges or to restructure their benefits to avoid them. Since no one likes to pay for benefits or services one did not receive or benefit from, it is reasonable to expect that some plans may change their terms to exclude these costs, just as it was reasonable to expect that some New York contractors would restructure ERISA plans instead of paying cash benefits. Implicit, then, is the assumption that an ERISA plan, if it does not want to pay for the direct cost of services provided to nonbeneficiaries, “structures all its benefit payments in accordance with New Jersey [hospital rates], or adopts different payment *1201formulae for employees inside and outside of the State.” See Fort Halifax, 482 U.S. at 10, 107 S.Ct. at 2217; Alessi, 451 U.S. at 523-25, 101 S.Ct. at 1907. This is the precise reason why Congress enacted such a broad preemption provision: so that employee benefit plans would not be subject to the vagaries of state regulations. Ingersoll-Rand, 498 U.S. at 141-42, 111 S.Ct. at 484; Fort Halifax, 482 U.S. at 10-11, 107 S.Ct. at 2217.
Because New Jersey has asked ERISA plans to carry the brunt of the burden of keeping hospitals financially afloat and, in fact, to largely subsidize its venture into hospital price regulation, and because the State admits that result before this court, I agree with the district court’s conclusion that there is a definite, palpable connection between the Act and the plan. See United Wire, 793 F.Supp. at 531-37.
IV.
Much attention in this appeal has been focused on Rebaldo v. Cuomo, 749 F.2d 133 (2d Cir.1984). The New York statute provided that the state must establish for each hospital an “inpatient revenue cap,” which was a price cap with additional allowances made for bad debts and charity care. The statute allowed certain programs, like Blue Cross, to get discounts, but certain ERISA plans did not qualify. The Court of Appeals for the Second Circuit first reasoned that because Section 514(c)(2) defines “State” as any agency, instrumentality or political subdivision that “purports to regulate, directly or indirectly, the terms and conditions of employee benefit plans,” 29 U.S.C. § 1144(c)(2), this restriction modified Section 514(a)’s broad preemption provision. 749 F.2d at 137. The court held that where “a State statute of general application does not affect the structure, the administration, or the type of benefits provided by an ERISA plan, the mere fact that the statute has some economic impact on the plan does not require that the statute be invalidated.” Id. at 139. This holding is now infirm as the Supreme Court has since rejected the view that the definition of “State” restricts the scope of Section 514(a). Ingersoll-Rand, 498 U.S. at 141-42, 111 S.Ct. at 484.2
The Rebaldo court justified its holding with an economic/policy argument. It reasoned that the regulation of hospital rates are analogous to any state regulations that increase the cost of doing business for hospitals. 749 F.2d at 138. Thus, for example, the court opined that ERISA plans could not contend that they are exempt from price increases as a result of increased labor, utility, or rent costs, or from such minor costs as the bridge and tunnel tolls that are charged to plan officers or employees. Id. Moreover, it opined that since ERISA plans can be subject to nationally uniform supervision despite dissimilarities in hospital prices, any “indirect economic impact” would be consistent with ERISA’s aim of national uniformity in plan regulation. Id. at 139.
Recently, even this economic rationale, for whatever worth it had, has been eroded. Although the Rebaldo court suggested that “State labor laws that govern working conditions and labor costs ... that [have] some economic impact on the plan” would not be preempted, id. at 138-39, the Court of Appeals for the Second Circuit has since retreated from this position and has preempted a generally applicable statute governing labor costs that imposed upon ERISA plans indirect financial and administrative burdens. General Electric, 891 F.2d at 29-30.3
*1202Despite the Supreme Court’s rejection of the holding in Rebaldo and the Court of Appeals for the Second Circuit’s implicit rejection of its rationale, the majority nonetheless relies heavily on it. Faced with Section 514(a)’s sweeping preemption provision and the cause-and-effect financial connection between the Act and ERISA funds, the majority justifies its decision by reasoning that the Act is analogous to any number of state laws that may indirectly increase the cost of doing business, such as regulations governing labor, utility or rent costs. Majority at 1193. It suggests as an example that regulations concerning the disposal of medical waste would not be preempted though their implementation may net higher medical bills as a result of the increased cost of doing business. Id. at 1195.
I agree with the majority insofar as ERISA plans do not lead a “charmed existence.” Rebaldo, 749 F.2d at 139. They are, of course, not exempt from paying rent, tolls, or even the many overhead costs associated with hospital management. The DRG takes into account both direct costs, such as hospital employees’ salaries and benefits, and indirect costs, such as institutional overhead expenses for management, research, education and maintenance. The plans, however, do not argue that the entire DRG system is incompatible with ERISA; they do not argue that they must only pay the “actual costs” for medical services; they do not argue that any laws that increase the cost of doing business for hospitals do not apply to them; nor do they argue that they are entitled to pay to the penny only their pro rata share of overhead and incidental costs. They argue instead that the three surcharges differ from ordinary overhead costs in two important ways: first, plan participants derive no benefit from these surcharges; second, these surcharges are not equitably distributed to the general population, or to the State, or to even hospital users, but are placed squarely on the shoulders of commercial insurers and ERISA plans. As a result, they contend that they are forced to subsidize nonparticipant patients with ERISA funds.
The majority largely ignores these distinctions, choosing instead to sweep all state-mandated costs under the general penumbra of “overhead” costs. When states regulate, for example, utility, labor, education or medical sanitation, they increase the cost of doing business. To remain solvent and be able to provide services, businesses must be able to incorporate, these increases into these prices. In a market-based- system, each consumer will pay for costs attributable to running the business, so overhead is part of the indirect costs associated with the purchase of goods and services. Since it is administratively impossible to isolate the precise actual costs attributable to any particular patient, many of the direct and indirect costs incorporated into the DRG such as overhead for management and maintenance are proper. The plans do not contend otherwise.
The surcharges are a different story, however. They are not indirect costs associated with any hospital services to plan beneficiaries. They are the direct cost of hospital services rendered to other patients, which have then been shifted to ERISA plans. One can argue, as the appellants do, that even businesses incorporate losses involved in stolen merchandises and bad debts into their prices. But the issue is not whether ERISA protects plans from the imposition of these kinds of -costs because ERISA preempts only “state law” and not private action, or whether ERISA plans are entitled to nationally uniform hospital prices because that is impossible. The issue is whether state regulations have interfered with the operation of ERISA plans to the point where the plans have suffered large financial losses.
The argument that plans are affected the same way without regulations is based on questionable applications of economic assumptions. In a free market system, as well as most regulated systems, no business would knowingly sell to one who cannot in *1203fall or in part pay, whereas here New Jersey requires hospitals to provide services regardless of one’s ability to pay and then places the losses squarely on a small class of patients. Without the Act, the financial calculus and its effects on ERISA funds would change drastically. Since each hospital will have different rates depending in part upon their total losses, with urban hospitals, for example, incurring more losses from indigent care and bad debts, the plans would be free to select hospitals with the lowest prices. And even if hospitals distribute their losses by overcharging ERISA plans, the plans would have an alternative: they would normally be free to negotiate with various hospitals for group rates just as some groups under the Act are able to- negotiate lower rates (the difference of which ERISA funds have been forced to defray). Under the Act, however, the plans are in effect precluded from negotiating group rates because, as the State concedes, the Act takes away any incentive to negotiate when it prohibits hospitals from passing on to other patients any revenue shortfall caused by a digression from the DRG rates. But see Majority typescript at 29-30 (“Nor does [the Act] deprive ERISA plans of any alternative they would otherwise have in these areas.”).
Moreover, in both market-based and regulated systems, businesses generally pass losses to all customers. Here, the object of the Act is to pass hospital losses inequitably to a small segment of hospital consumers. The surcharges are neither paid by the State nor the public at large; they are not even paid by all hospital users. The State created these surcharges with the specific intention that, in essence, only a small group of the hospital users — a large majority of whom comprise ERISA beneficiaries — would pay for them. While the plans do not derive any benefit, real or abstract, from the surcharges, the Act takes a substantial chunk of money belonging to ERISA beneficiaries, and all because New Jersey does not want to expend its general funds to pay for the health care costs of others who are less fortunate. As one affidavit summarized: “In effect, New Jersey requires the hospitals to give the service but will pay for it using other people’s money, i.e., other users of the hospitals rather than a broader base revenue source such as general taxation.”
Thus, the surcharges here are anything but “ordinary” overhead costs that indirectly increase the cost of doing business, and the argument that the Act simply sets prices does not sufficiently credit the direct financial impact on ERISA plans. The majority is correct insofar as states may regulate, for example, the disposal of medical waste and allow hospitals to pass these costs off to patients; but if states require hospitals to pass off these costs only or largely to “commercial insurers and self-funded union plans,” such regulations must be preempted.
V.
I fear that the majority gives States free reign to spend and experiment with ERISA funds, held in trust for the many workers who have labored long for their security, in the noble pursuit of health care reforms so long as States exercise a minimum degree of imagination by couching their statutes in “generally applicable” terms. When I consider the financial impact and other causal effect on ERISA funds in a common sense manner, the connection between ERISA plans and the Act is not too tenuous, remote or peripheral at all. The Act refers specifically to ERISA plans, divests enormous sums of money from ERISA plans, and is predicated on the existence of ERISA plans. In my opinion the Act “relates to” ERISA plans when that term is construed in its ordinary and broad meaning, and I dissent.
. Without the congressional intent derived specifically from the structure of ERISA, Mackey would probably have been decided differently. Both the majority and the dissent in Mackey observed that the garnishment law subjected the plans to "substantial” and "significant administrative burdens and costs.” See 486 U.S. at 831, 842, 108 S.Ct. at 2186, 2192. Because "plans face the repetitious and costly burden of monitoring controversies involving hundreds of beneficiaries and participants in various States,” Id. at 844, 108 S.Ct. at 2193, Justices Kennedy, Black-mun, O’Conner, and Scalia dissented.
. Although the majority recognizes that Inger-soll-Rand overruled Rebaldo, at least in part, it distinguishes Ingersoll-Rand on the basis that the Rebaldo court dealt with "a generally applicable statute that makes no reference to, [and] functions irrespective of, the existence of an ERISA plan.” Majority at 1194 (quoting Ingersoll-Rand, 498 U.S. at 139, 111 S.Ct. at 483). Of course, that is the question here.
. Indeed, although Judge Van Graafeiland authored both Rebaldo and General Electric, the dissent in General Electric considered Rebaldo controlling:
The majority does not attempt to distinguish Rebaldo and Aetna Life [Ins. Co. v. Borges, 869 F.2d 142] nor, in my view, could it sensibly do so. Like the statutes we upheld in those decisions, New York's prevailing wage statute is a law of general application whose tangential effects on employee benefits plans are negligible and wholly incidental to the law’s primary purpose.
*1202891 F.2d at 31.
The rejection of Rebaldo is understandable. It is not so much attributable to inconsistent deci-sionmaking as to the court’s recognition that ERISA preemption is as broad as the statute would suggest. The Court of Appeals for the Second Circuit decided Rebaldo in 1984, without the benefit of, among other cases, Holliday, Pilot Life, Metropolitan Life, Mackey, and Ingersoll-Rand.