Morgan Guaranty Trust Co. v. Tax Appeals Tribunal of the New York State Department of Taxation & Finance

OPINION OF THE COURT Kaye, J. This appeal calls upon us to determine whether a State tax of general application — in this case a tax on gains derived from real property transfers within the State — is preempted by the Employee Retirement Income Security Act (ERISA). Based on the structural, administrative and economic impact of the tax on the qualified employee benefit plan in issue in this case, we conclude that the tax relates to the employee benefit plan in more than a tenuous, remote or peripheral way, therefore requiring preemption. The relevant facts have been stipulated. Respondent, Morgan Guaranty Trust Company, served as trustee under the American Motors Corporation Union Retirement Income Plan —a qualified employee benefit plan — during 1984, the year in issue. In 1965, nine years prior to ERISA, the Plan purchased real property in Greenburgh, New York, from an American Motors affiliate for $723,014 and entered into a 25-year leaseback arrangement. In December 1983, counsel advised American Motors that the lease was a "prohibited transaction” under ERISA because the 1965 rent schedule lagged behind current market rates, and that the applicable transitional exemption would soon expire. Counsel suggested that the lease be revised, a further exemption requested, or the Plan’s interest in the property transferred. Failure to take any action would subject the Plan to tax penalties. The property was sold in June 1984 to a second American Motors affiliate for $2,775,640. *47 The New York tax is imposed on the gain derived from a property transfer (Tax Law § 1441; see, Trump v Chu, 65 NY2d 20, 23, appeal dismissed 474 US 915). "Gain” is defined as "the difference between the consideration for the transfer of real property and the original purchase price of such property, where the consideration exceeds the original purchase price” (Tax Law § 1440 [3]). The consideration received must exceed $1 million, but the tax is imposed on 10% of the entire gain (Tax Law §§ 1441, 1443). On this transaction, Morgan paid the State $205,262. In May 1985, Morgan filed a refund claim with the Department of Taxation and Finance on the ground that imposition of the gains tax was preempted by ERISA. The Audit Division denied the claim based on the Department’s position that ERISA preempted only those State laws designed to regulate employee benefit plans. Morgan then filed an administrative petition for refund with the Tax Appeals Bureau of the State Tax Commission. The Administrative Law Judge granted the petition, holding that "the transfer itself was of a Plan asset, was made by the Plan itself, was a course of action dictated by ERISA rules regarding investment requirements and resulted in a direct tax upon the Plan.” Accordingly, the ALJ concluded that it was "clear that in this limited instance, the gains tax directly relates to and affects the Plan and hence must be held preempted.” On appeal, the Tax Appeals Tribunal reversed the ALJ, holding that imposition of the gains tax on the Plan was "too tenuous, remote and peripheral to require preemption.” The two majority members of the Tribunal noted that options to sale of the parcel were available and that any administrative burden resulting from imposition of the tax was de minimis. The dissenting Tribunal member found that where "the tax is imposed directly on the earnings from the sale of a Plan’s assets, that in itself is sufficient to reach the conclusion that the tax 'relates to’ the Plan in an impermissible way and is preempted.” Morgan brought this proceeding in the Appellate Division to set aside the Tribunal’s determination. In annulling the determination, Justice Yesawich, writing for a unanimous court, reasoned that the gains tax, which was imposed directly on income derived from the sale of a Plan asset, was related to the Plan and preempted by ERISA. We now affirm. *48 ERISA Preemption Principles ERISA preemption of State law has been the subject of substantial recent litigation (see generally, Gregory, ERISA Law in the Rehnquist Court, 42 Syracuse L Rev 945 [1991]). Controversy has centered on preemption of State law causes of action that offer remedies for the violation of rights expressly guaranteed by ERISA (see, e.g., Pilot Life Ins. Co. v Dedeaux, 481 US 41; Tingey v Pixley-Richards W, 953 F2d 1124 [9th Cir]). The question before us is different. None of the cases has considered the issue presented here— preemption of a State tax law of general application — which, nevertheless, is determined by settled ERISA preemption principles. Analysis begins with the presumption that State law has not been preempted by Federal statute "in the absence of persuasive evidence to the contrary” (Sassa v Vachris, 66 NY2d 28, 33). The focus is on congressional intent: " '[t]he purpose of Congress is the ultimate touchstone.’ ” (Malone v White Motor Corp., 435 US 497, 504, quoting Retail Clerks v Schermerhom, 375 US 96, 103.) As the Supreme Court has noted, where Congress "has expressly included a broadly worded pre-emption provision in a comprehensive statute such as ERISA, our task of discerning congressional intent is considerably simplified.” (Ingersoll-Rand Co. v McClendon, 498 US 133,138.) ERISA’s preemption clause is indeed broad: all State laws are superseded "insofar as they may now or hereafter relate to any employee benefit plan” (29 USC § 1144 [a] [emphasis added]). The clause was designed to establish pension plan regulation as exclusively a Federal concern. (Ingersoll-Rand Co. v McClendon, 498 US, at 138.) Broad preemptive authority was intended to facilitate the "emergence of a comprehensive and pervasive Federal interest and the interests of uniformity with respect to interstate plans” (Statement of Senator Javits, 120 Cong Rec 29942 [1974]). That Federal interest was invoked, and ERISA promulgated, to address the concern that "the soundness and stability of plans with respect to adequate funds to pay promised benefits may be endangered.” (29 USC § 1001 [a].) The ERISA preemption clause is "conspicuous for its breadth” (FMC Corp. v Holliday, 498 US 52, 58), "virtually *49 unique” (Franchise Tax Bd. v Laborers Vacation Trust, 463 US 1, 24, n 26), "one of the broadest preemption clauses ever enacted.” (Evans v Safeco Life Ins. Co., 916 F2d 1437, 1439 [9th Cir]; see also, Gregory, The Scope of ERISA Preemption of State Law: A Study in Effective Federalism, 48 U Pitt L Rev 427, 431-432 [1987].) The scope of the preemption clause prompted one court to remark that in "considering Congress’ intent in the ERISA context, all roads lead to Rome.” (McCoy v Massachusetts Inst, of Technology, 950 F2d 13, 17 [1st Cir].) Moreover, it is evident from the legislative history that Congress intended exactly this broad effect. As described by the Supreme Court: "The bill that became ERISA originally contained a limited pre-emption clause, applicable only to state laws relating to the specific subjects covered by ERISA. The Conference Committee rejected these provisions in favor of the present language, and indicated that the section’s pre-emptive scope was as broad as its language.” (Shaw v Delta Air Lines, 463 US 85, 98 [citing HR Conf Rep No. 93-1280, at 383 (1974); S Conf Rep No. 93-1090, at 383 (1974)].) State laws specifically designed to affect employee benefit plans are viewed as preempted by ERISA (see, Mackey v Lanier Collection Agency & Serv., 486 US 825, 829). If, as with the tax law at issue in the present case, the State law is not specifically designed to affect employee benefit plans — if it is a law of general application — the next step is to determine whether the law "relate[s] to” employee benefit plans (Mackey v Lanier Collection Agency & Serv., supra, at 830-831). The phrase "relate[s] to” must be given a " 'broad commonsense meaning’ ” (Ingersoll-Rand Co. v McClendon, 498 US, at 139, supra). Does the law have "a connection with or reference to such a plan” (Shaw v Delta Air Lines, 463 US, at 97, supra)? Even in the absence of an express link to an employee benefit plan, State law is preempted "insofar as” the law applies to benefit plans in particular cases (Shaw v Delta Air Lines, supra, at 97, n 17). Thus, a State law may be preempted even though not specifically designed to affect such plans, and even though any effect is only indirect (Ingersoll-Rand Co. v McClendon, 498 US, at 138-139, supra). Two recent preemption cases, Ingersoll-Rand and FMC *50 Corp., have put to rest the contention that only laws relating to the terms, conditions or administration of a plan are preempted (see, Ingersoll-Rand Co. v McClendon, supra; FMC Corp. v Holliday, 498 US, at 58-60, supra [Congress did not intend to preempt only State laws dealing with subject matters covered by ERISA, such as reporting, disclosure and fiduciary duties]). Likewise, the argument that only State laws that "purport to regulate” an ERISA plan are preempted has been rejected (Ingersoll-Rand Co. v McClendon, 498 US, at 141-142; see also, Smith v Dunham-Bush, Inc., 959 F2d 6, 9, n 3 [2d Cir]). It is apparent that the Supreme Court has consistently expanded the scope of ERISA preemption while identifying certain limits "which allow for the independent operation of some narrowly prescribed state laws.” (Smith v Dunham-Bush, Inc., 959 F2d, at 9, supra.) Despite the breadth of the preemption clause, certain State laws "may affect employee benefit plans in too tenuous, remote, or peripheral a manner to warrant a finding that the law 'relates to’ the plan.” (Shaw v Delta Air Lines, 463 US, at 100, n 21.) Several Federal courts have referred to that language in upholding State laws of general application against challenges based on ERISA preemption (see, Matter of Dyke, 943 F2d 1435, 1447 [5th Cir]; see also, Gregory, op. cit., 48 U Pitt L Rev 427, 465-466 ["Unfortunately, the Court presented no principled formula by which to ascertain which state laws were so 'tenuous, remote, or peripheral’ as to avoid the broad scope of ERISA preemption.”]). Thus, in the present case we must determine whether this tax law of general application "relate[s] to” benefit plans in more than a tenuous, remote or peripheral way so as to warrant preemption. We conclude that it does. Preemption of the Gains Tax Even State tax laws of general application, representing a traditional exercise of State authority, may be preempted when they "relate to” ERISA plans (see, Firestone Tire & Rubber Co. v Neusser, 810 F2d 550, 556 [6th Cir]). While explicitly exempting other State laws of general application — State criminal law, for example — ERISA made no mention of tax law in the original preemption provision (see, 29 USC § 1144 [b] [2] [A]; [4]). In 1983, Congress amended the statute to exempt Hawaii’s Prepaid Health Care Act and at the same time affirmed that the preemption clause applied *51 to "any State tax law relating to employee benefit plans” (29 USC § 1144 [b] [5] [B] [i]; see, HR Conf Rep No. 97-984, 97th Cong, 2d Sess 18, reprinted in 1982 US Code Cong & Admin News 4598, 4603). The fact that a State tax law of general application is at issue is therefore not determinative of ERISA preemption. Applying the broad commonsense meaning of the statutory phrase "relate[s] to,” we conclude that this gains tax has more than a tenuous, remote or peripheral connection to employee benefit plans and is therefore preempted by ERISA. We reach that conclusion from analysis of the structural, administrative and economic impact of the tax on the Plan, viewed against the backdrop of the terms and objectives of ERISA (see, Mackey v Lanier Collection Agency & Serv., 486 US 825, supra; Firestone Tire & Rubber Co. v Neusser, supra; Aetna Life Ins. Co. v Borges, 869 F2d 142 [2d Cir]). The gains tax clearly impacts on the structure and administration of the Plan. As the dissenting Tribunal member noted, the gains tax would impose certain recordkeeping and reporting requirements on the Plan, mandating administrative procedures pertaining to asset disposition not required in other jurisdictions (see, Tax Law § 1447 [2]; § 1448 [2], [3]). Far more significant than this administrative burden, however, is the influence the gains tax will necessarily have on the Plan’s investment strategy. ERISA imposes Federal standards of conduct on the managers of benefit plans, and fiduciaries are required to tailor investment strategy to those guidelines (see, 29 USC §§ 1101-1114; 29 CFR 2509.75-5, 2550.404a-l). Although real estate transactions such as the one in issue are sanctioned by ERISA, New York fiduciaries will have to consider the State law that, by directly taxing gains on the sale of such assets, makes them less attractive investments. By the same token, an administrator taking the cost of the New York gains tax into account may be required to retain an asset that would otherwise have been liquidated. As the Supreme Court has made clear, it is undesirable to require plans and employers to tailor their conduct "to the peculiarities of the law of each jurisdiction” (Ingersoll-Rand Co. v McClendon, 498 US, at 142, supra). Indeed, such a result is "fundamentally at odds with the goal of uniformity that Congress sought to implement.” (Id.; see also, FMC Corp. v Holliday, 498 US, at 58-60, supra.) Moreover, this is not a "cost of doing business” law, as appellants argue but a tax *52 applied directly to the income derived from appreciation of a Plan asset. In the present case, the effect on the Plan’s structure and administration is even more direct: this transaction was a response to ERISA’s "prohibited transaction” rules, which required that some action be taken. The Tribunal noted that the sale was not the sole option available to the Plan. However, as Morgan points out, the other options — exacting a higher rent or requesting an extended exemption — were not feasible, forcing the Plan to pay either the gains tax on the profit realized or a penalty under ERISA. Either would affect the operation of the Plan. Finally, preemption in this case would be consistent with the favorable tax treatment given to benefit plans under the Internal Revenue Code. Congress designed ERISA "to make sure to the greatest extent possible that those who do participate in * * * plans actually receive benefits and do not lose their benefits as a result of * * * failure to accumulate and retain sufficient funds to meet its obligations.” (HR Rep No. 93-807, 93d Cong, 2d Sess 8, reprinted in 2 Sen Subcomm on Labor of Comm on Labor and Pub Welfare, Legislative History of ERISA, at 3129 [1976]; see also, Massachusetts v Morash, 490 US 107, 115.) In furtherance of this goal, earnings on a plan’s assets are exempt from Federal taxation (see, 26 USC § 501 [a]; § 401 [a]; see also, 26 USC §512 [b] [5]). This favorable tax treatment was a mainstay of prior pension legislation and its continuation considered essential to ERISA (see, Statement by Senator Humphrey, Cong Rec — Senate [Aug. 12, 1974], reprinted in 3 Legislative History of ERISA, op. cit., at 4778). Unlike other forms of general State regulation that may have only incidental effect on plan resources, the gains tax "directly depletes the funds otherwise available for providing benefits” and flies "in the face of ERISA’s goal of assuring the financial soundness of such plans.” (Birdsong v Olson, 708 F Supp 792, 801 [WD Tex].) As Morgan points out, imposition of the gains tax on ERISA plans withdraws plan assets Congress sought to protect, levying against the very funds Congress denied the Federal treasury. Appellants urge that this tax is indistinguishable from others that have been allowed, but analysis of appellants’ cases reveals material differences. For example, this tax is significantly different from one imposed upon employees re *53 gardless of whether their income was subsequently contributed into an ERISA plan (Firestone Tire & Rubber Co. v Neusser, 810 F2d 550, supra), or a tax levy or withholding procedure on payments-out from the plan to the beneficiaries (Retirement Fund Trust of Plumbing v Franchise Tax Bd., 909 F2d 1266). Unlike those examples, the gains tax here directly depletes Plan assets. As one court has noted, the gains tax is not like a stamp or documentary transfer tax — taxes generally imposed on the entire consideration paid, at a rate of less than 1%, irrespective of profit (995 Fifth Ave. Assocs. v New York State Dept, of Taxation & Fin., 963 F2d 503). Instead, this gains tax is contingent on the profitability of the underlying transaction: a 10% tax is imposed directly on the gain — as defined by the provisions of the gains tax — accruing to the transferor (id.; compare, Tax Law §§ 1402, 1402-a [real estate transfer tax]). Thus the gains tax is not, as appellants argue, "akin to a sales tax,” and as such a "cost of doing business in New York.” It is a direct tax on Plan profits. The magnitude of the effect is underscored by the estimate that most of the asset increase projected for ERISA plans in the next decade will be attributable to the income earned from existing assets in established ERISA plans (see, Gregory, op. cit., 48 U Pitt L Rev, at 436). Depletion of those earnings, projected as the primary source of future funding necessarily increases the cost of the Plan — for the employer or the employees — or decreases the benefits (see, E-Systems, Inc. v Pogue, 929 F2d 1100, 1103, cert denied sub nom. Barnes v E-Systems, Inc., — US —, 112 S Ct 585). Appellants would ignore the direct impact and make the test one of comparison only: a generally applicable law should not be preempted if it is applied to a covered plan in the same way, and for the same reasons, as it would be applied to any other citizen, even though application of the law may burden the Plan (see, dissenting opn, at 56-57). That argument is unsupported and unsound. Appellants draw on language from the Ingersoll-Rand decision where the Supreme Court noted that it was not dealing "with a generally applicable statute that makes no reference to, or indeed functions irrespective of, the existence of an ERISA plan.” (498 US, at 139, supra.) Taken in context, however, it is plain that the Supreme Court was not making the sweeping declaration that all such laws are exempt from *54 preemption but rather referring to the specific laws at issue in Mackey (supra) and Fort Halifax Packing Co. v Coyne (482 US 1). Similarly, appellants’ reliance on Mackey is mistaken. The Supreme Court in Mackey did not exempt the Georgia garnishment scheme from preemption based on the failure of the statute to refer to ERISA plans, nor did it reject the notion that a law will be preempted because it burdens or affects a covered plan. 1 In Mackey, the Court reasoned that, because ERISA plans may "sue or be sued,” State law methods for collecting money judgments must, as a general matter, remain undisturbed by ERISA; otherwise, there would be no way to enforce such a judgment won against a plan. (486 US, at 834; see also, Retirement Fund Trust of Plumbing v Franchise Tax Bd., 909 F2d 1266, 1275, supra.) No similar reasoning underlies appellants’ test. Thus, there is no basis for determining preemption by whether a general law applies to covered plans in the same way, and for the same reasons, as it applies to other citizens. 2 It is this proposed test that is "pioneering” (dissenting opn, at 60), both for its widespread implications — especially in the area of State taxation — and for its inattention to the actual impact a State law of general application may have on covered plans. Analyzed from the correct perspective of statutory principles and Supreme Court precedents, it is clear that this gains tax is preempted because it affects the structure, administration and economics of a covered plan, and therefore "relate[s] to” it in more than a tenuous, remote or peripheral way. *55 Accordingly, the order of the Appellate Division should be affirmed, with costs. Titone, J. (dissenting). In 1984, in order to comply with ERISA requirements, petitioner, the trustee of a large pension fund governed by the Federal ERISA statute (see, 29 USC § 1001 et seq.), sold some real property that had previously been held under a leaseback arrangement. The issue in this appeal brought by respondent State Tax Appeals Tribunal is whether ERISA preempts the application of New York State’s real property transfer gains tax (see, Tax Law § 1441) to the gain from that sale. Relying on the generally broad preemption policy that is unquestionably reflected in the relevant ERISA provisions (29 USC § 1144 [a], [b] [5] [B] [i]), the majority holds that petitioner is exempt from the tax, essentially because the tax will have an economic impact on an ERISA covered plan. However, since all taxes inevitably have an economic impact, this criterion is clearly an inadequate one for determining whether a particular tax is preempted under ERISA. Furthermore, although the majority has attempted to supplement its economic-impact analysis with arguments about the tax’s effect on the plan’s structure and administration, its arguments are unconvincing. Finally, contrary to the majority’s conclusion, the relevant case law suggests that this generally applicable tax on gains was not the type of local enactment that was within Congress’s overall preemptive intent. Accordingly, I must, respectfully, dissent. L There are two relevant preemption provisions. The first of these States that "the provisions [of ERISA] shall supersede any and all State laws insofar as they may now or hereafter relate to any [covered] employee benefit plan” (29 USC § 1144 [a] ). The second states, more specifically, that State tax laws that "relate to” ERISA plans are preempted (29 USC § 1144 [b] [5] [B] [i]). As the majority stresses, the Supreme Court has consistently endorsed a broad construction of these preemption provisions (e.g., FMC Corp. v Holliday, 498 US 52; IngersollRand Co. v McClendon, 498 US 133, 139-140; Pilot Life Ins. Co. v Dedeaux, 481 US 41; Shaw v Delta Air Lines, 463 US 85, *56 98; Alessi v Raybestos-Manhattan, Inc., 451 US 504, 525). Yet, the reach of ERISA’s preemption clause is not limitless (see, Shaw v Delta Air Lines, supra, at 100, and n 21). Both 29 USC § 1144 (a) and (b) (5) (B) (i), whose identical language requires "the same preemption analysis” (Retirement Fund Trust of Plumbing v Franchise Tax Bd., 909 F2d 1266, 1276), apply only to the extent that the challenged State law "relates to” a covered plan. Thus, notwithstanding the Supreme Court’s general statements about Congress’s broad preemptive intent, Congress chose to preempt something less than all local laws that affect or impact upon covered ERISA plans. Unfortunately, the Supreme Court’s discussions of what constitutes a law that "relates to” a covered plan have not, to date, produced a workable test for preemption. The Court has said that the term "relates to” should be given "its broad common-sense meaning” (Metropolitan Life Ins. Co. v Massachusetts, 471 US 724, 739) and that a State law "relates to” a covered plan " 'if it has a connection with or reference to such a plan’ ” (Ingersoll-Rand Co. v McClendon, 498 US, at 139, supra; accord, Shaw v Delta Air Lines, supra, at 96-97). The Court has also said that a State law may be preempted "even if [it] is not specifically designed to affect [covered] plans, or the effect is only indirect” (Ingersoll-Rand Co. v McClendon, 498 US, at 139, supra), but that there will be no preemption where the State law "affectfs] employee benefit plans in too tenuous, remote, or peripheral a manner” (Shaw v Delta Air Lines, 463 US 85, 100, n 21, supra). While these formulations are somewhat helpful, none is sufficiently specific to provide a definitive basis for resolving the issue in this case: whether Congress intended to exempt covered plans from liability for generally applicable State taxes. Thus, further analysis and an exploration of the specific contexts in which 29 USC § 1144 (a) has been applied are required. IL In our view, a simple reading of the relevant case law demonstrates that a generally applicable law does not "relate to” the plan upon which its burden is to be imposed unless the law either (1) specifically refers to covered plans or (2) does not operate, either in general or in the particular case, without reference to the plan’s existence. In other words, the Federal ERISA statute does not require preemption of a *57 generally applicable law if it is being applied to a covered plan in the same way, and for the same reasons, as it would be applied to any other entity, even though application of the law may, in fact, burden the plan. Initially, the Supreme Court’s opinion in Ingersoll-Rand Co. v McClendon (498 US, at 139-140, supra), on which the majority places much reliance, specifically appears to sanction this formula. Although the Ingersoll-Rand Court found the particular local law before it to be preempted by ERISA, the Court went on to suggest that the outcome would have been otherwise if it had been "dealing * * * with a generally applicable statute that makes no reference to, or indeed functions irrespective of, the existence of an ERISA plan” (id., at 139; emphasis supplied). Further, the Court bolstered the point by noting that a finding of preemption was warranted in the case before it because "the existence of a pension plan [was] a critical factor in establishing liability under the State’s * * * law” (id., at 139-140), suggesting once again that the outcome would have been different if liability were being imposed without regard to the existence of a covered plan. 1 A similar conclusion may be inferred from Mackey v Lanier Collection Agency & Serv. (486 US 825, 833, and n 8), in which the Supreme Court took note of the virtually universal assumption that ERISA plans may be subjected to liability for "run-of-the-mill state-law claims such as unpaid rent, failure to pay creditors, or even torts [they commit].” Further, many of the lower Federal courts that have considered the ERISA preemption issue have reached their conclusions using the analysis suggested in Ingersoll-Rand. In Lane v Goren (743 F2d 1337, 1340), for example, the Ninth Circuit Court of Appeals held that a generally applicable antidiscrimination law was not preempted by ERISA because it was being applied to a covered plan "only * * * in [the plan’s] capacity as an employer, and [only] in a way that all other employers are affected.” Significantly, the Lane court contrasted the State antidiscrimination statute before it with the antidiscrimination statute at issue in Shaw v Delta Air Lines (supra), which was specifically directed toward the structure of certain covered employee benefit plans (see, 463 US, at 96-97; see also, 743 F2d 1340-1341 [distinguishing Alessi v Rayhestos-Manhat *58 tan, Inc. (supra) on the ground that it involved a State statute "directed at or affect(ing) the internal operations of the trust”]). Moreover, the Ninth Circuit squarely rejected the argument that the statute at issue in Lane should be held preempted because its application would increase the plan’s cost of doing business. In the court’s view, that argument was not persuasive because the same could be said about virtually any local law that regulates zoning, health and safety (id., at 1340). A similar view was adopted by the Sixth Circuit in Firestone Tire & Rubber Co. v Neusser (810 F2d 550, 556), in which the court acknowledged the broad reach of ERISA’s preemption provisions but nevertheless upheld a municipal income tax applied to wages contributed directly to a covered ERISA plan because it was a "neutral * * * tax of general application” "affecting * * * employees in their capacity as employees, without regard to their status as participants in an ERISA plan.” Like the Lane court, the Firestone court rejected the contention that a State statute that increases the cost of doing business should be deemed preempted for that reason alone (id., at 555; accord, Rebaldo v Cuomo, 749 F2d 133, 139 [2d Cir], cert denied 472 US 1008 ["mere fact that the statute has some economic impact on the plan does not require that the statute be invalidated”]). The position that generally applicable State laws are not preempted if they burden covered ERISA plans only in the same way that all other similarly situated citizens would be burdened was also adopted in Retirement Fund Trust of Plumbing v Franchise Tax Bd. (supra) and In re Seolas (140 Bankr 266 [ED Cal]). 2 In light of this formidable list of cases, it is difficult to understand the *59 majority’s unqualified assertion that "there is no basis for determining preemption by whether a general law applies to covered plans in the same way, and for the same reasons, as it applies to other citizens” (majority opn, at 54). Finally, a clear pattern emerges from a review of the facts in virtually all of the leading Federal ERISA preemption cases, and that pattern supports the foregoing analysis. The courts tend to find preemption either where a State law refers specifically to covered ERISA plans (FMC Corp. v Holliday, supra; Mackey v Lanier Collection Agency & Serv., supra; Shaw v Delta Air Lines, supra; Alessi v Raybestos-Manhattan, Inc., 451 US 504, supra; E-Systems, Inc. v Pogue, 929 F2d 1100, cert denied sub nom. Barnes v E-Systems, Inc., — US —, 112 S Ct 585; General Motors Corp. v California State Bd. of Equalization, 815 F2d 1305, cert denied sub nom. General Motors Corp. v Bennett, 485 US 941; In re Seolas, supra; Birdsong v Olson, 708 F Supp 792) or where the particular application of a "generally applicable” State law depends on the existence of a covered plan (Ingersoll-Rand Co. v McClendon, supra; Pilot Life Ins. Co. v Dedeaux, 481 US 41, supra; Cefalu v B.F. Goodrich Co., 871 F2d 1290). 3 In contrast, where the challenged law is a "generally applicable” one and is being applied to a covered plan in a way that is no different than it would be applied to all other citizens, the courts tend to hold that its application is not preempted (Mackey v Lanier Collection Agency & Serv., supra; Retirement Fund Trust of Plumbing v Franchise Tax Bd., supra; Firestone Tire & Rubber Co. v Neusser, supra; Lane v Goren, supra; Aetna Life Ins. Co. v Borges, 869 F2d 142, cert denied 493 US 811; In re Seolas, supra). Thus, there exists a substantial body of case law that strongly suggests the validity of respondent’s contention that generally applicable State laws are not preempted when they "[make] no reference to, [and] indeed function[J irrespective of, the existence of an ERISA plan” (Ingersoll-Rand Co. v Mc-Clendon, 498 US, at 139, supra [emphasis added]). Even more *60 importantly, with one possible exception of dubious precedential worth, 4 none of the leading Federal cases goes so far as to find preemption — as the majority does here — where a State law of general application is being applied to a covered plan in the same way as it would be applied to any other entity doing business in the State. Since nothing in the majority’s opinion provides a persuasive rationale for its pioneering position, we conclude that that position is analytically unsound. ¡IL The precise basis of the majority’s holding remains unclear. The majority begins with a brief reference to the legislative history of 29 USC § 1144 (b) (5) (B) (i) (majority opn, at 50-51), the specific ERISA provision dealing with preemption of local tax laws, but it draws no useful conclusions from the cited history. This restraint seems well advised, since the legislative history establishes no more than that Congress did not wish to give State tax laws favored status vis-á-vis the preemption rule. It would require a great leap of logic to infer from this history that Congress intended to accord State tax laws a disfavored status and to mandate their preemption in all circumstances. The majority also seems to adopt petitioner’s argument that State tax laws are preempted under ERISA because, by their nature, they deplete plan assets and thereby undermine one of the central purposes of the Federal legislation: to ensure "to the greatest extent possible that those who do participate in * * * plans * * * do not lose their benefits as a result of * * * failure to accumulate and retain sufficient funds to meet [plan] obligations” (HR Rep No. 93-807, 93d Cong, 2d Sess 8, reprinted in 2 Sen Subcomm on Labor of Comm on Labor and Pub Welfare, Legislative History of ERISA, at 3129 [1976]). However, the argument has little support in the case law. In Lane v Goren (supra, at 1340), the court specifically rejected a preemption argument based on the fact that the *61 local law increased the cost of doing business, thereby reducing plan assets. In Arkansas Blue Cross & Blue Shield v St. Mary’s Hosp. (947 F2d 1341,1348, cert denied — US —, 112 S Ct 2305), the court stated that, with regard to the preemption analysis, economic impact may be relevant, but it should not be dis-positive. The problem of fund depletion was expressly considered in E-Systems, Inc. v Pogue (supra) and Birdsong v Olson (supra, at 801), but even in those cases it was not the dispositive factor, since the challenged tax statute in both instances made specific and direct reference to ERISA plans. The primary weakness of the fund-depletion argument is that it is based on a skewed view of Congress’s purpose in enacting ERISA. The goal of this legislation was to safeguard the soundness of employee benefit funds, not to maximize their profitability (In re Seolas, supra). This goal was to be achieved by providing a uniform national system of substantive and procedural rules to govern the administration of these plans — a system that was not to be disrupted or undermined by conflicting (or even harmonizing) local legislation. It does not follow from Congress’s stated goal of protecting plan assets, however, that Congress also intended ERISA’s preemption provisions to protect plans from the ordinary costs of doing business in various localities and thereby assure plan participants the maximum possible benefits. 5 In an apparent effort to reconcile its argument with Federal cases in which local taxes affecting covered ERISA plans have been upheld (e.g., Firestone Tire & Rubber Co. v Neusser, supra), the majority contends that the gains tax challenged here is different because it directly depletes plan assets (see also, Birdsong v Olson, supra, at 801). However, the argument does not withstand analysis. First, it is inconsistent with the *62 majority’s own suggestion that some taxes that directly deplete plan assets, such as sales, stamp and documentary transfer taxes, are permissible (majority opn, at 53). Second, it cannot logically be reconciled with the majority’s central asset-depletion theory, which presumably should apply without regard to whether the depletion occurs through direct or indirect means. Indeed, taken to its logical conclusion, the majority’s asset-depletion theory would seem to require line-drawing based not on the direct or indirect nature of the tax, but rather on the amount of money involved. Of course, that is an unwieldy and inherently implausible means of determining whether preemption is required. Apparently recognizing these serious flaws in its asset-depletian theory, the majority also relies on the tax’s impact on the plan’s "structure” and "administration” as a basis for finding preemption. However, the majority’s analysis on these points is equally flawed. The majority concludes that the tax law at issue here affects the "structure and administration” of the plan because it imposes "certain recordkeeping and reporting requirements * * * mandating administrative procedures * * * not required in other jurisdictions” (majority opn, at 51). Yet, the record-keeping and reporting requirements imposed by New York’s gains tax law are certainly no more burdensome than were the recordkeeping and reporting requirements established by the garnishment laws that were held not to be preempted in Mackey v Lanier Collection Agency & Serv. (supra) or the levy and withholding procedures that were held not preempted in Retirement Fund Trust of Plumbing v Franchise Tax Bd. (supra). Furthermore, "[wjhat triggers ERISA preemption is not just any indirect effect on administrative procedures but rather an effect on the primary administrative functions of benefit plans, such as determining * * * eligibility for a benefit and the amount of that benefit” (Aetna Life Ins. Co. v Borges, supra, at 146-147). Clearly, requiring a plan to keep some additional records for tax purposes does not fall within this category. As for the majority’s repeated appeals to the gains tax’s impact on plan "structure,” it is significant that no specific effect is ever identified. A plan’s "investment strategies,” a consideration stressed by the majority, are plainly not part of the plan’s "structure” — a term that denotes internal relationships, rules and methods of operation. Moreover, however it is *63 characterized, the fact that a particular State law may influence a plan’s investment strategy is an unsatisfactory ground for finding preemption because the same could be said about a host of local laws, including zoning laws, that inevitably have an affect upon citizens’ economic choices. Indeed, a similar argument was recently rejected in In re Seolas (supra), where the court declined to find a local usury law preempted despite the fact that the law incidentally influenced an ERISA plan’s investment choices. Finally, the majority’s reliance on the plan’s exemption from certain Federal taxes is misplaced. If Congress had wished to insulate covered ERISA plans from local income and gains taxes, it could easily have specifically provided as much, as it, in fact, did in the case of Federal income and certain real property gains taxes (see, 26 USC § 401 [a]; § 501 [a]; § 512 [b] [5]). Instead, it provided for a rule of preemption that applies only to the extent that the local tax "relates to” covered ERISA plans. Thus, it is evident that no blanket prohibition against State taxation was intended. Further, all that can really be inferred from the enactment of the Federal taxation exemption is that, as a matter of national policy, Congress elected to use its broad taxing authority to promote the growth of ERISA plans. Nothing in that legislation, or in the relevant provisions of the ERISA statute itself, suggests the existence of an entirely different congressional intention to override local tax laws that do not affirmatively promote such growth in the same manner. CONCLUSION A review of the case law, as well as the language and purpose of the ERISA preemption provision, reveals no sound reason to hold a generally applicable tax on gains from real property sales to be preempted solely because the entity to be taxed happens to be the operator of a covered ERISA plan. Although Congress intended to protect ERISA plans from State laws that would regulate or "relate to” their operations as covered pension and welfare programs, it clearly did not intend to immunize such plans against all local legislation that might burden or otherwise affect them. The test for preemption that respondent advances — which would reject preemption when a generally applicable State statute does not refer to covered plans and, in fact, functions irrespective of such plans’ existence — would seem sufficient to protect both *64 ERISA’s goals and the legitimate, non-plan-related goals of the States. Accordingly, I dissent from the majority’s position and vote to reverse the Appellate Division order precluding imposition of the tax. Chief Judge Wachtler and Judges Simons and Bellacosa concur with Judge Kaye; Judge Titone dissents and votes to reverse in a separate opinion in which Judge Hancock, Jr., concurs. Order affirmed, with costs.. Footnotes [1] Appellants’ reliance on Mackey’s statement that run-of-the-mill State law claims — such as unpaid rent and failure to pay creditors — may be brought against covered plans is even more puzzling (see, dissenting opn, at 57). The Court mentioned such claims only as they supported the conclusion that some garnishment or enforcement proceeding was required to render any recovery meaningful (486 US, at 833). No similar need for the gains tax is advanced by appellants. [2] The dissent’s heavy reliance on In re Seolas (140 Bankr 266 [ED Cal]), a decision of a Federal District Court Judge in California, is misplaced. At issue in Seolas was preemption of the California usury law. Like the examples of State garnishment and antidiscrimination laws on which the dissent rests, the usury example provides no analogy to a tax imposed on gain realized by an ERISA plan from the sale or exchange of plan assets. Moreover, preemption of the California usury law would entail both civil and criminal sanctions, and criminal laws are expressly exempt from ERISA preemption. [1] The majority’s effort to minimize the significance of the above-quoted language in Ingersoll-Rand (majority opn, at 53-54) is simply not supported by a plain reading of the decision. [2] The majority’s effort to distinguish In re Seolas (supra) (majority opn, at 54, n 2) is wholly unpersuasive. The majority’s first assertion that the usury law at issue in that case "provides no analogy to [the] tax imposed [here]” is too conclusory to have any analytical value. The majority’s alternative argument that applying the preemption rule in Seolas would have "entailfed] both civil and criminal sanctions” even though the latter are expressly exempt is simply inaccurate. The only issue in Seolas was the right of a covered plan under California’s civil law to enforce a usurious loan contract. The enforceability of the State’s criminal usury laws was not, in fact, implicated. Finally, notwithstanding the majority’s creative effort to demonstrate that Seolas does not really support respondent’s position, the fact remains that the Seolas court’s rationale — i.e., that the challenged State law was "a provision of general application dealing with all nonexempt lenders in a uniform manner” (140 Bankr, at 272) — is precisely the rationale that respondent urges here. [3] In Ingersoll-Rand Co. v McClendon (supra), for example, the State lawsuit that was held to be preempted was based on generally applicable common-law "wrongful discharge” principles, but its underlying theory— that the discharge had been "wrongful” because it was motivated by a desire to circumvent the employer’s ERISA obligations — plainly required "reference to” a covered plan. [4] In Northwest Airlines v Roemer (603 F Supp 7), the court found a generally applicable State law requiring withholding as a means of recovering delinquent taxes to be preempted to the extent that the State sought to apply it to covered ERISA plan benefits. This decision differs from the others in that it extends preemption to a situation in which the State law neither referred to ERISA plans nor required the existence of an ERISA covered plan in order to operate. However, the Supreme Court’s subsequent decision in Mackey v Lanier Collection Agency & Serv. (supra) upholding application of a general State garnishment law to a covered plan brings the continued validity of Northwest Airlines into serious question. [5] The majority asserts that the particular tax at issue here, a tax imposed on the gain from the sale of real property, is not properly characterized as a "cost of doing business,” because it is "not like a stamp or documentary transfer tax” and is also not "akin to a sales tax” (majority opn, at 53). However, the majority never explains why the class of expenses that may be characterized as "costs of doing business” should be limited to the three-enumerated types of tax. Nor does it explain why the manner in which the tax is measured or the fact that the tax is higher than 1% are relevant to the preemption inquiry. In fact, since buying and selling assets— and realizing a gain on their assets’ appreciation — are, unquestionably, primary "business” goals of employee benefit plans, a tax imposed on that gain is a "cost of doing” business to a benefit plan in precisely the same way that a sales tax imposed on the goods its purchases would be.