Lunding: The Supreme Court Debates Discrimination
Introduction
Christopher Lunding took his case to the Supreme Court and won, coming away $3,724 richer (before expenses), and making some interesting constitutional law along the way. His case(1) established that New York's denial of an alimony deduction for nonresidents, while granting such a deduction for residents, unconstitutionally denied Mr. Lunding the privileges and immunities enjoyed by residents of New York. (For the full text of the High Court's ruling in Lunding, see Doc 98-3734 (24 pages).)
The Supreme Court's decision in Lunding offers an interesting window on one particular aspect of the Constitution's injunctions against interstate discrimination. The majority of the Court found that New York had not adequately justified its discriminatory treatment of nonresident payers of alimony. An equally articulate minority, however, argued forcefully for the constitutionality of New York's tax regime. Within this debate one finds expressions of many practical and theoretical considerations that have shaped the analysis of tax discrimination thus far, and will continue to affect the law of local boosterism.
To appreciate these influences, let's take ourselves back to school.
English
It must first be clarified that this paper discusses alimony. In this discussion it is necessary to refer to payers and payees, and sometimes to refer to their relative stations and sources of income. "For convenience [and to avoid linguistic inelegance, albeit at the expense of virtually every advance in gender equality achieved to date] the payee spouse will hereafter in this [paper] be referred to as the 'wife' and the spouse from whom she is divorced or separated [and who pays the alimony] as the 'husband.'" Treas. Reg. section 1.71-1(a).
Gym
Charles Barkley has been quoted for the proposition that "it is good to have the home court advantage because otherwise you have to win on the road. "I am told that insofar as the NBA playoffs are concerned, this apparently tautological statement actually has meaning. Insofar as the U.S. Constitution is concerned, however, home court advantage is a foul thing indeed. As a nation, we have agreed to provide a level playing field, and over the years we have developed a jurisprudence that bars states from choking off national commerce for the benefit of their locals.
Civics
An obvious and exceedingly important function of the Constitution is to knit the various states together into a whole within which people and commerce are treated equally and move freely, without regard to the state of origin or destination. In the area of state taxation, three different provisions of the U.S. Constitution are most frequently invoked to challenge and invalidate state or local taxes that grant favors to "locals" or impose burdens on "foreigners." These are the Privileges and Immunities Clause of Article IV section 2, the Commerce Clause of Article I section 8, and the Equal Protection Clause of the 14th Amendment. The Supreme Court's decision in Lunding was based on the Privileges and Immunities Clause, although below he had challenged New York's statute as violating all three clauses.
As evident from the fact that this entire paper deals with one case, it is impossible here (perhaps anywhere) to do justice to the jurisprudence of discriminatory state taxation. Outlining some of the principal theories and leading cases, however, provides a backdrop for the Lunding debate.
The Privileges and Immunities Clause provides:
The Citizens of each State shall be entitled to all Privileges and Immunities of Citizens in the several States.
As characterized by a leading commentator, this clause is the "principal constitutional provision protecting nonresidents against discriminatory State personal income taxation...." Hellerstein and H., State Taxation, para. 20.06. Since "citizens" have been defined as limited to natural persons, however (Western Turf Ass'n v. Greenberg, 204 U.S. 359 (1907)), this clause rarely appears outside the personal income tax and death tax areas.
The Equal Protection Clause, part of the 14th Amendment, provides that:
No State shall make or enforce any law which shall abridge the privileges or immunities of citizens of the United States; nor shall any State deprive any person of life, liberty or property without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws. (Emphasis added.)
Challenges under the Equal Protection Clause often lead to analyses similar to those articulated in Privileges and Immunities cases.
The equal protection line of authority requires first that a taxpayer identify some actual discrimination in taxation between itself and an identical class of favored taxpayers. This can be difficult. We know, for example, that butter is not the same as margarine, so one may be taxed more onerously than the other. A. Magnano Co. v. Hamilton, 292 U.S. 40 (1934). Similarly, a local regulated utility is not the same as an unregulated energy marketer, General Motors Corp. v. Tracy, 519 U.S. 278 (1997), and a developer of condominiums is not the same as a developer of single-family homes. Trump v. Chu, 65 N.Y. 2d 20 (1985).
In addition to the difficulty of identifying discrimination between groups, the Courts generally are quite deferential to legislatively drawn distinctions. As articulated in General Motors, "the hurdle facing GMC is a high one, since state tax classifications require only a rational basis to satisfy the Equal Protection Clause." 579 U.S. 278. (For the full text of this ruling, see Doc 97-4863 (22 pages).) And the standard of rationality seems to be that of a reasonably prudent legislator, giving new meaning to the term. For example, in Nordlinger v. Hahn, California succeeded in convincing the Supreme Court that the reassessment-on-sale regime of Proposition 13 did not unconstitutionally tax similar properties differently, because it promoted neighborhood stability and discouraged rapid turnover that leads to gentrification and the displacement of lower-income persons (exactly what the California voters were so worried about). In short, "if a legislature could have reasonably concluded that the challenged classification would promote a legitimate State purpose," the tax will stand. Exxon Corp. v. Egerton, 462 U.S. 176 (1983).
The Commerce Clause, by contrast, has a much lower tolerance for discrimination. The Commerce Clause provides simply that:
The Congress shall have Power...To regulate Commerce with Foreign Nations, and among the several States, and with the Indian Tribes.
In granting Congress the power to regulate, the Constitution has also, it has been held, created a dead zone -- a "'negative' aspect that denies the States the power unjustifiably to discriminate against or burden the interstate flow of articles of commerce." Oregon Water Sys. v. Department of Environmental Quality, 511 U.S. 93 (1994). "No State, consistent with the Commerce Clause, may impose a tax which discriminates against interstate commerce...by providing a direct commercial advantage to local business." Northwestern States Portland Cement Co. v. Minn., 358 U.S. 450 (1959). While certain narrowly targeted compensatory taxes may survive Commerce Clause scrutiny -- notably state use taxes that mirror their sales taxes, Henneford v. Silas Mason Co., 300 U.S. 577 (1937) -- facially discriminatory taxes have a very high death rate under the Commerce Clause.
That the courts are tougher on discrimination under the Commerce Clause may be attributed to the fact that judicial findings of unconstitutional behavior under the Commerce Clause may be overridden by congressional action. By contrast, if a state statute is broken under the Equal Protection or Privileges and Immunities clauses, it stays broken.
Mr. Lunding clearly would have had an easier time invalidating New York's obviously discriminatory statute under the Commerce Clause. However, in keeping with its name, this clause requires the complainant to demonstrate an effect on interstate commerce. The Court of Appeals questioned whether Lunding's alimony involved interstate commerce; its recent decision in Tamagni suggests more strongly that, where personal issues are involved, the Commerce Clause may not apply. Tamagni v. Tax Appeals Tribunal of New York, 91 N.Y. 2d 530, (1998). (For the petitioner's brief in Tamagni, see State Tax Notes, Sep 14, 1998, p. 699.) A two-legged taxpayer with a personal problem, therefore, Mr. Lunding found himself mainly under the umbrella of the Privileges and Immunities Clause.
As summarized by the Supreme Court in 1870, "[b]eyond doubt [the Privileges and Immunities Clause is] of very comprehensive meaning, but it will be sufficient to say that the clause plainly and unmistakably secures and protects the right of a citizen of one State to pass in to any other State of the Union for the purpose of engaging in lawful commerce, trade, or business without molestation; to acquire personal property; to take and hold real estate; to maintain actions in the courts of the State; and be exempt from any higher taxes or excises than are imposed by the State upon its own citizens." Ward v. Maryland, 79 U.S. 418, 430 (Wall) (emphasis added). The objective is to ensure that nonresidents are not taxed more onerously than residents.
One difficult aspect of discrimination, however, and one that was important in Lunding, is identifying the element of state law that produces discrimination, and what it should be compared to. Was New York's denial of Lunding's deduction discriminatory because it allowed that deduction to residents? Or was it not discriminatory because, taken in conjunction with New York's failure to tax his ex-wife, it was part of an overall scheme of taxation that achieved "rough parity"?
In 1902, the Supreme Court considered Connecticut's system of assessing stock held by nonresidents at market value, while stock held by residents was assessed at market value less the value of taxable real estate. Travelers' Ins. Co. v. Conn., 185 U.S. 364 (1902). Connecticut's facially discriminatory statute was upheld because, "when the system of taxation prevailing in Connecticut is considered," id., at 367, the discrimination "disappears." Specifically, the Connecticut system, according to the Court, divided its overall tax burdens between residents and nonresidents by imposing local taxes on residents and a statewide tax on nonresidents. The Court believed that Connecticut's "Legislature...aimed...to apportion fairly the burden of taxes between the resident and the nonresident stockholder...." Id., at 369. Travelers therefore looked at the entire system to discern discrimination, for the apparently unfair taxation of nonresidents clearly was part of a larger taxing plan that was intended, overall, to reach a fair result. "Mere inequality in the results of a state tax law is not sufficient to invalidate it," at least where "no intentional discrimination has been made against nonresidents." Id., at 371.
A pair of 1920 decisions involving discrimination in the allowance of deductions figures very prominently in Lunding. Shaffer v. Carter, 252 U.S. 37, and Travis v. Yale & Towne Mfg. Co., 252 U.S. 60. Shaffer involved an Illinois taxpayer's challenge of an Oklahoma statute that permitted residents to deduct losses without regard to source, but limited nonresidents' deductions of what was apparently a business loss to losses incurred within the state. Residents and nonresidents were afforded the same exemptions, but nonresidents' deductions were curtailed. "The difference, however, is only such as arises naturally from the extent of the jurisdiction of the state...." Id., at 57. Because states can tax only the state-source income of nonresidents, it is reasonable that they not be required to allow deductions for losses incurred in other states.
Travis involved a variety of limitations New York imposed on nonresidents. New York permitted nonresidents to deduct only amounts "connected with income arising from sources with the state." Travis, at 73. It allowed exemptions only to residents, not to nonresidents. It afforded nonresidents a credit on their tax on New York-source income equal to the proportionate amount of their home-state tax paid on such income, but only if the home state afforded New Yorkers a "substantially similar" credit. It imposed withholding on salaries and wages paid to nonresidents, but waived withholding for persons filing a certificate of residence. Yale & Towne, as a Connecticut corporation employing Connecticut and New Jersey residents to work in New York, challenged New York's authority to require it to withhold, on the basis that the underlying statute was unconstitutional and withholding would subject it to claims by its employees.
Citing Shaffer, the Court affirmed New York's right to tax nonresidents on their New York-source income, and that "there is no unconstitutional discrimination against citizens of other states in confining the deduction of expenses, losses, etc., in the case of nonresident taxpayers, to such as are connected with income arising from sources within the taxing state...." Id., at 39-40. The Court also said that Yale & Towne could constitutionally be required to withhold New York tax on the portion of the salaries earned in New York.
They found, however, that New York's denial of an exemption to nonresidents violated the Privileges and Immunities Clause. Quoting from the decision below, the Court agreed that "[w]hether [nonresidents] must pay a tax upon the first $1,000 or $2,000 of income, while their associates and competitors who reside in New York do not, makes a substantial difference. Under the circumstances as disclosed, we are unable to find adequate ground for the discrimination, and are constrained to hold that it is an unwarranted denial to the citizens of Connecticut and New Jersey of the privileges and immunities enjoyed by citizens of New York. This is not a case of occasional or accidental inequality due to circumstances personal to the taxpayer...but a general rule, operating to the disadvantage of all nonresidents including those who are citizens of the neighboring states, and favoring all residents including those who are citizens of the taxing states." Id., at 80-81.
New York's "ingenious argument" that its discrimination on the exemption front was counterbalanced by its exclusion of nonbusiness income from New York's tax base was not accepted because the income exclusion was "not so conditioned as probably to benefit nonresidents to a degree corresponding to the discrimination against them; it seems to have been designed rather...to preserve the preeminence of New York City as a financial center." Id. Similarly "it would be rash to assume that nonresidents taxable in New York under this law, as a class, are receiving additional income from outside sources equivalent to the amount of the exemptions that are accorded to citizens of New York...." Id. And finally, New York's belief that its discrimination would be offset by "the speedy adoption of an income tax by the adjoining states...is wholly speculative; ...besides...a discrimination by the state of New York against the citizens of adjoining states would not be cured were those states to establish like discriminations against citizens of the State of New York." Id., at 82. Thus, the denial of the exemption was intentional discrimination and, unlike Connecticut in Travelers, New York could not identify anything the Legislature might rationally assume to be a fair offset.
Austin v. New Hampshire, 420 U.S. 656 (1975) is, like Travelers, a case about what one compares. A New Hampshire tax on commuters equal to the tax imposed by their home state was invalidated because, when considered within the fabric of New Hampshire's income tax law, it clearly discriminated against nonresidents. On its face the statute applied to income earned in New Hampshire by nonresidents, and to income earned by residents outside New Hampshire. But the state then exempted its resident's incoming income from tax if the income was taxed in the source state, was exempted from tax in the source state, or was not taxed in the source state.
While difficult to parse, these rules are not difficult to understand -- the design and effect of New Hampshire's overall tax system was to tax only nonresidents. That is unconstitutional. That this tax simply shifted revenues from other states to New Hampshire was not the point: the taxation of nonresidents alone, with no offsetting tax on residents, was fundamentally irreconcilable with "the policy of comity to which the Privileges and Immunities Clause commits us." Id., at 666. Thus, over Justice Blackmun's dissent that the case was a "waste of time" because the commuters would pay the same amount of tax win or lose (but wouldn't you rather spend the money at home?), the Court struck down the commuter tax.
Several state decisions also are relevant to Lunding. In 1955 and 1956 the Appellate Division and Court of Appeals analyzed a New Jersey resident's challenge of New York's denial of various deductions. In Goodwin v. State Tax Commission, 296 A.D. 694, aff'd 1 N.Y. 2d 680, the New York statute provided that deductions are allowed to nonresidents "only if, and to the extent that, they are connected with income arising from sources within the State and taxable under this article to a nonresident taxpayer." Tax Law section 360(11). The courts considered both the constitutionality of the statute and its proper application to Mr. Goodwin's various expenses.
All of Mr. Goodwin's income was derived in New York. He had claimed, and was allowed, deductions for bar association dues, legal periodicals, business entertainment, car expense, and certain charitable contributions. His claim for New York deductions for his New Jersey real property taxes and home mortgage interest, medical expenses, and life insurance premiums, however, had been denied. The taxpayer conceded these expenditures had no connection to his New York income, but argued that he could not constitutionally be denied, purely on the basis of his nonresidency, deductions granted to residents. Citing Travis and Shaffer, it was held that New York could constitutionally limit deductions to those connected with the state, and these were not.
Interestingly, the Appellate Division inferred that Travis had specifically upheld New York's right to deny deductions for property taxes on nonresidents' homes. It noted that this exact example of discrimination had been illustrated in the briefs to the Supreme Court in Travis, and that the Supreme Court had upheld the source-based deduction statute knowing this to be its effect.
The "principle of the Supreme Court's decision holding that a distinction could validly be made between residents and nonresidents with respect to real estate taxes is equally applicable to [home mortgage interest, medical expenses, and life insurance costs.]" 286 A.D. 2d 694, 698. Limiting deductions to those connected with in-state income was a proper and necessary corollary to source-based income taxation. "To the extent...the taxpayer is allowed to deduct [personal or non-business expenses],...these deductions undoubtedly [have] their genesis in governmental policy...[for example] a governmental policy in favor of home ownership." Id., at 699.
"The question is whether the factor of residence has a legitimate connection with the allowance of these deductions so that a classification upon the basis of residence is justifiable." Id., at 700. The answer here was yes. All of the challenged expenditures "relate to the personal activities of the taxpayer and his personal activities must be deemed to take place in the State of his residence, the State in which his life is centered." Id., at 701. "[E]ach of the deductions embodies a governmental policy designed to serve a legitimate social end. The governmental policies are peculiarly related to the factor of residence within the State." Id., at 702. New York had no obligation to extend to nonresidents the governmental incentives it offered, for good policy reasons, to its residents.
Importantly in the context of Lunding, the Appellate Division in Goodwin distinguished these personal deductions from Travis' exemption. "A flat exemption, in connection with a graduated income tax, has a direct effect upon the rate of taxation. It excludes a part of the taxpayer's income from the tax base, and it lowers the tax bracket into which his remaining income falls. Obviously, it would be unconstitutional for New York State arbitrarily to tax the New York income of a nonresident at a higher rate than that applied to the income of residents." Id., at 702. That legal point seems obvious to everyone, and widely accepted. As argued herein, however, its application seems to have escaped notice in Lunding.
The Appellate Division opinion in Goodwin was affirmed by the Court of Appeals, with no opinion.
Subsequently, in Golden v. Tully, 88 A.D. 2d 1058, the Appellate Division again considered a nonresident's claim for personal deductions, and this time allowed it. Mr. Golden was a New Jersey resident who moved to New Mexico, and claimed his moving expenses as a deduction against his New York income. New York allowed residents to claim moving expense deductions, but claimed Mr. Golden, as a nonresident, was not entitled to the deduction.
The Appellate Division concluded that the Tax Department had applied New York's statute correctly, but that its application in this case violated the Privileges and Immunities Clause. "The modern rule, evolved in Toomer v. Witsell (334 U.S. 385 (1948)), does not bar disparity of treatment where perfectly valid reasons for it exist, but does prohibit discrimination against citizens of other States where there is no substantial reason for the discrimination beyond the mere fact that they are citizens of other states." Id.
Again, there had to be a state policy reason for the rule, and the discrimination against nonresidents had to have some rational relationship to the discrimination. "We are unable to find that such an independent reason exists here." Moving expenses, the court reasoned, related most clearly to the income that would be earned in the new location. In affording residents a deduction without regard to whether they moved within or out of New York, the state had delinked the deduction from its economic source, and simply decided to offer residents a tax benefit.
Given the lack of any policy relating to the factor of residence, New York could not justify denying the deduction to nonresidents.
This is a particularly interesting result, and probably correct. Golden's expenses clearly had no relationship whatsoever to the state. But neither would the moving expenses of a New Yorker who moved to New Mexico. Absent a state policy to encourage moving (perhaps to increase the economic activity that attends the sale of a New York home?) New York was simply doing its residents a favor. Mr. Golden could not constitutionally be denied that same favor.
Golden was affirmed by the Court of Appeals. 58 N.Y. 2d 1047 (1983). Significantly, given the Court's later opinion in Lunding, the state lost Golden in the Court of Appeals because the only rationale it offered for denying the moving deductions was his nonresidence.
Wood v. Department of Revenue, 10 O.T.R. 374 (1987), considered Oregon's denial of a pro rata alimony deduction to a Washington resident who earned all of his income in Oregon, had two Oregon ex-wives, each of whom paid Oregon tax on the income, and whose "present and future Oregon law practice income was a major factor in determining the amount of alimony to be paid." 1987 Ore. Tax Lexis 68, at 1. Tracing the development of the federal and Oregon treatment of alimony, the Oregon Tax Court concluded that, at the time of Mr. Wood's divorces, both the federal and the state income tax laws allowed him an itemized deduction for alimony, based on the relationship of his Oregon income to total income.
With the federal change (discussed in History, below) to an adjusted gross income deduction for alimony, Mr. Wood's Oregon deduction (like Mr. Friedsam's) was no longer clear. Under Oregon's statute, alimony would be deductible from Oregon AGI only if attributable to an Oregon trade or business. Believing that Mr. Wood's alimony was unrelated to his law practice, Oregon denied the deductions. The Tax Court concluded this was correct as a matter of statutory interpretation.
Considering the taxpayer's privileges-and-immunities claim, the Tax Court agreed that while it was true the alimony was now taxed twice, there was no double taxation of the petitioner himself. Furthermore, "the Constitution leaves the legislature broad discretion in determining what exemptions or deductions may be extended to nonresidents. Review of the statute leads the court to believe that the legislature has simply not been made aware of the problem." Id., at 9.
Thus, while the result seemed unfair and "it is hoped that the legislature will remedy the situation," id., the Tax Court disallowed the deductions. One might say they found no constitutional infirmity because whatever discrimination existed arose as a result of a federal change, and not out of an intention to discriminate against nonresidents.
One year later, the Supreme Court of Oregon reversed. 305 Ore. 23 (1988). The reversal was rendered en banc, but without opinion. Oregon thus held Mr. Wood was entitled to his alimony deductions, but did not say why. (Compare Friedsam, below.)
The last state court decision to add to the plot is Spencer v. S.C. Tax Commission, a 1984 decision regarding South Carolina's refusal to allow prorated nonbusiness deductions to nonresidents unless their home state afforded a reciprocal deduction to South Carolina residents. 281 S.C. 492. Citing the analysis of Toomer v. Witsell (1984), South Carolina's supreme court opined that discrimination against nonresidents is constitutional only when the state shows "that nonresidents are a peculiar source of the evil at which the statute is aimed." Id., at 496. The court rejected the state's explanation that the disallowance was not retaliatory but instead "designed to encourage sister states to enact legislation favoring South Carolina residents....[T]he goal of encouraging other states to enact reciprocal legislation does not bear a substantial relationship to the result of penalizing taxpayers like the Spencers who live in North Carolina and work in South Carolina. These taxpayers are not the source of the evil sought to be remedied by our legislature." Id.
After granting certiorari in Spencer, the U.S. Supreme Court found itself evenly divided - and in any event was apparently considering only an issue of the award of attorney's fees. See respondent's brief to the Supreme Court in Lunding, at note 9. The judgment of the South Carolina Supreme Court was therefore affirmed, without opinion. 471 U.S. 82 (1984). The decision stands, but without elucidation.
Mr. Lunding therefore enters a scene in which the constitutionality of denying nonresidents a variety of "personal" expenses has already been much litigated. And the law that emerges from these cases is that states cannot discriminate on rates and they cannot discriminate for no good reason, but they may be able to discriminate if they can articulate a reasonable policy behind it.
History
In Lunding, policy justifications New York State might otherwise have advanced as legitimizing its denial of alimony deductions may well have been compromised by its past schizophrenia regarding alimony deductions.
Mr. Lunding characterized New York's denial of the alimony deductions as stemming from New York's "desire to take the easy path of increasing revenue by imposing discriminatory taxes on nonresidents, who cannot vote in New York and thus have no voice in its legislative processes." Petitioner's brief at 14. That may not be entirely fair, given that the enactment of the alimony denial was just a small part of the general overhaul of New York's income tax laws effected in 1987. (See McIntyre and Pomp, "State Income Tax Treatment of Residents and Nonresidents Under the Privileges and Immunities Clause," State Tax Notes, Jul 28, 1997, p. 245.) However, in the face of "not one word of legislative history for this statute," petitioner's brief at 4, and the state's burden to offer some explanation for a provision that clearly discriminates against nonresidents, the statement could be true. In any event, the state's historical treatment of nonresidents' alimony is a fair line of inquiry.
Originally, New York allowed alimony deductions to residents and nonresidents if the recipient was subject to New York income tax. 1943 N.Y. Laws Ch. 245, section 3; 1944 N.Y. Laws Ch. 333, section 2. In 1961, the New York personal income tax was amended to adopt a system of conformity with federal law to provide simplicity, improve enforcement, and aid interpretation. 1961 N.Y. Laws Ch. 68, section 1. At that point, the matching of husband's deduction to wife's income was eliminated, and nonresidents were allowed a pro rata share of their itemized deductions.
Specifically, the new tax included a provision entitling nonresidents to "substantial equality" with residents. Tax Law section 635(c)(1). The statute permitted New York residents to take the same deductions from New York AGI as were permitted federally as itemized deductions. Tax Law section 615. At the time, alimony was an itemized deduction for federal income tax purposes, and thus deductible by New York residents. Under the "substantial equality" rule, nonresidents were permitted New York deductions for a proportionate share of their federal itemized deductions, including alimony. From 1961 through 1976, therefore, New York followed a policy "that nonresidents be allowed the same nonbusiness deductions as residents, but that such deductions be allowed to nonresidents in the proportion of their New York income to income from all sources." Friedsam v. State Tax Comm. 64 N.Y. 2d 76, at 81 (1984), quoting from Murphy and Petite, "Taxation of Nonresidents by New York State," 12 Syracuse L. Rev. 147, 161-62, and referring also to a Memorandum of Governor, 1961 N.Y. Laws Ch. 68.
The trouble started in 1976 when the U.S. Congress, intending no harm, made alimony an above-the-line deduction. Tax Reform Act of 1976, section 502, enacting code section 62(a)(13), now section 62(a)(10). The stated purpose of this federal change was to provide alimony deductions even to those taxpayers who did not itemize. The problem it created in New York (and Oregon) was that, rather than falling under the rubric of Tax Law section 615 and the itemized deductions to which nonresidents were automatically entitled, alimony was now deductible under Tax Law section 632, which defined the New York adjusted gross income (AGI) of a nonresident. To be deductible in computing New York AGI, an expenditure had to be "derived from or connected with New York sources [which meant] items attributable to...business, trade, profession or occupation carried on in this state." Tax Law section 632(b)(1)(B).
Following the federal change, New York adopted the administrative position that alimony was no longer deductible by nonresidents, because it was not related or connected to New York income. In re Lance J. Friedsam, TSB-H-82-(37)-I. Mr. Friedsam was a Connecticut resident paying alimony out of his New York income, and he challenged New York's denial of the alimony deduction (1) under the Privileges and Immunities Clause, and (2) as violative of New York's statute.
In an opinion relied upon heavily in its 1996 decision in Lunding, the Appellate Division determined that New York's denial of the alimony deduction to Mr. Friedsam violated the Privileges and Immunities Clause. 98 A.D. 2d 26 (1983). In so holding, the court noted that the Privileges and Immunities Clause proscribes discriminatory treatment of nonresidents "[w]here there is no substantial reason for the discrimination beyond the mere fact that they are citizens of other States." Id., at 28, quoting from Golden, supra.
The state's attempted denial of Mr. Friedsam's deduction should, one would think, occupy considerably shakier constitutional ground than even Lunding, given that Friedsam's denial was based solely on administrative practice, and grew out of a technical federal computational change never commented on or addressed by the New York Legislature. (Compare Wood.) In light of New York's prior allowance of proportional alimony deductions when itemized, it is difficult indeed to articulate a principled basis for denying such deductions administratively solely because a federal change moved them above the line -- all the more so in the face of legislative silence.
Even the Court of Appeals had trouble with the department's position in Friedsam. It did not reach the constitutional issue, but instead focused on the mechanics and policy of New York's statute. "We hold that the Commission's determination supporting a disparate tax classification between resident and nonresident taxpayer is contrary to the statute and tax policy of this State....The salutary policy of substantial equality embodied in Section 635(c)(1) of the Tax Law serves to invalidate the challenged determination...." 64 N.Y. 2d 76, at 81-82. "There was no State legislative response to [the federal change] and the Legislature gave no indication whatsoever that either New York's historical policy of following Federal tax policy, or the long-standing treatment of alimony vis--vis nonresidents was to be changed in any respect." Id., at 81.
The Legislature got the hint. Although we have no official legislative history as to its source or policy justifications, in 1987, the New York State Legislature approved section 631(b)(6) and the absolute denial of nonresidents' alimony deductions entered the Tax Law.
In a 1990 Advisory Opinion, New York State's Department of Taxation and Finance explained that "alimony paid by a nonresident to a resident is not considered a deduction derived from New York sources." Ronald S. Rosenblatt, TSB-A-90(3)I. Even in a circumstance where the wife paid tax on the alimony received from (yet another) Connecticut resident, the nonresident husband was denied a deduction based on the characterization of alimony as unrelated to New York. The 1987 legislation "specifically reversed" Friedsam; "the effect of the allowance of the deduction in the base and the denominator and disallowance in the numerator [explained below] is that Petitioner cannot get the benefit of a proportional deduction of the alimony payments...." Id.
As is clear from this history, New York affirmatively changed its treatment of alimony, from proportionally allowable to totally disallowed. Significantly, it made this obvious turnabout, and enacted a facially discriminatory statute, without any attempt to explain itself, or to justify its change of heart toward nonresidents.
Math
Several mathematical aspects of the arguments and results of Lunding merit elucidation. New York's computation of the tax owed by nonresidents is derived by formula, and the effect of Lunding on that formula is interesting.
The New York Tax Law arrives at the tax liability of a nonresident through a series of steps. Tax Law section 601(e). The first step involves the computation of the taxpayer's tax liability "as if" a resident of New York. To perform this task, one must compute the individual's taxable income under the provisions applicable to New York residents (Part II of Article 22). This calculation of "as if" income will include the taxpayer's worldwide income (based on federal AGI -- Tax Law sections 611(a), 612(a)); will make the modifications required of residents (e.g., no deduction for state income taxes deducted in computing federal AGI -- Tax Law section 612(b)(3)(A)); and will, by virtue of its deductibility in computing federal AGI (IRC section 62(a)(10)) allow a deduction for alimony. The "as if" income will also reflect the New York standard or itemized deductions (Tax Law section section 613-615) and the New York exemptions (Tax Law section 616) afforded to New York residents. "As if" income is therefore as advertised - the amount of income on which a New York resident would pay tax, taking into account all of the deductions allowed to residents.
"As if" income is then multiplied by the relevant tax rate to derive the "tax base" -- generally the tax the nonresident individual would owe if he failed his residency exam. Tax Law section 601(e)(2).
The tax base is then multiplied by the "New York source fraction." The numerator of this fraction is the individual's "New York source income" determined under the rules for nonresidents (Tax Law Art. 22, Part III). New York-source income (NYSI) is items of income, gain, loss, and deduction "derived from or connected with New York sources." Tax Law section 631(a). The offending section 631(b)(6) now comes into play, specifically providing that alimony "shall not constitute a deduction derived from New York sources," meaningthat (pre-Lunding) income sourced in New York entered the numerator unreduced by alimony deductions. Hence, Mr. Lunding's problem.
The last component of the formula is the denominator of the fraction, which is the taxpayer's New York adjusted gross income. Tax Law section 601(e)(3). New York AGI (NYAGI) is, like the "as if" income, computed under the Part II resident rules. It allows a deduction for alimony, and otherwise is the same as federal AGI, with certain specified modifications.
The New York tax liability of a nonresident individual is therefore equal to:(2)
"As If" Income x Tax Rate x |
NYSI NYAGI |
At this point some examples are helpful to elucidate what was at issue in Lunding. Assume a nonresident husband derives $90 of income from a New York business and $90 of income from an out-of-state business, and pays $70 of alimony and $10 of interest and taxes on his out-of-state home.
His "as if" income is $180 minus $70 minus $10, or $100. His (1990) tax rate is 7.875 percent. His New York-source income, pre-Lunding, is $90, representing $90 of income from the New York business, with no deduction for the alimony (Tax Law section 631(b)(6)), and no deduction for the $10 of interest and taxes (see Goodwin, supra). His NYAGI is $110, reflecting income of $180 and above-the-line alimony of $70, but without reduction for the itemized deductions. His tax liability therefore is:
$100 x 7.875% x 90/110 = $6.44
On his $90 of New York income, his effective tax rate is 7.16 percent. If one thinks of the alimony as chargeable equally against all income sources, however, his effective rate of tax on $55 of "net" New York income is 11.71 percent.
Mr. Lunding believed that New York was wrong to disallow alimony in computing the NYSI numerator. He argued that a nonresident taxpayer should be able to deduct a "pro rata portion" of alimony in computing NYSI. On these (oversimplified) facts, Mr. Lunding would add a preliminary proration computation, under which the $70 alimony would be sourced to New York in proportion to the New York source of all gross income. Thus:
$70 x 90/180 is deducted in computing NYSI.
Post-Lunding our formula now looks like this:
$100 x 7.875% x 55/110 = $ 3.94
Effectively, because his income was equally divided between two sources, and alimony was the only item deducted in computing NYAGI, the effect of the prorated alimony deduction is to reduce his New York-source fraction from 82 percent to 50 percent. Half of his income comes from New York, so half of his alimony is deducted here. The $3.94 tax paid on his $55 net represents an effective rate of tax of 7.16 percent.
The Supreme Court did not do the math, but the foregoing approach post-Lunding is supported by the Supreme Court's repeated references to "pro rata" deductions for alimony. And the Supreme Court did seem right in taking "little comfort" from an argument advanced by the state- and embraced by the Court of Appeals -- that the effect of all this formula business was to give nonresidents some deduction for alimony.
Specifically, in upholding New York's denial of the alimony deduction, the Court of Appeals noted that Lunding had not been denied "all benefit of the deduction," because alimony is deducted in computing the "as if" income. Without a deduction for the $70 alimony in the "as if" component, however, the State's formula (pre-Lunding) would be as follows:
$170 x 7.875% x 90/110 = $10.95
That tax far exceeds the maximum rate of tax on the gross income from New York, something that clearly makes no sense. Rather than stemming from a desire to grant some partial deduction for alimony, therefore, the allowance of the alimony deduction in the "as if" portion of the formula seems mathematically necessary to maintain some logical correlation to the NYAGI denominator. This is an unworthy basis for upholding the denial of an alimony deduction in computing NYSI.
Interestingly, notwithstanding Goodwin and the emphasis placed on it by the Court of Appeals, the operation of New York's current formula does actually allow nonresidents some deduction for itemized deductions. Assume in our example that there was no alimony deduction, just $10 of itemized deductions. Our taxpayer's formula would be:
"As If" income = $170 x 7.875% x 90/180 = $6.69
If all the nonresident's income were derived in New York, his formula would be:
"As If" income = $170 x 7.875% x 180/180 = $13.38
Effectively, the itemized deductions that are allowed in computing "as if" income survive the formula. They appear in neither the New York-source income numerator (because they are unrelated to New York sources) nor the New York AGI denominator (because they are below the line). But their allowance in computing "as if" income imports them into the computation of the nonresident's tax, and nothing takes them out again. In this respect, New York's treatment of alimony was even more restrictive than its treatment of "purely personal expenses."
A very simple example illustrates this, and distills the problem. Assume there is only $90 of income, all New York sourced, and $10 of either (1) itemized deductions or (2) alimony:
(i) "As if" Income = $80 x 7.875% x 90/90 = $6.30
(ii) "As if" Income = $80 x 7.875% x 90/80 = $7.09
In (i) the taxpayer pays tax on $80; in (ii) he pays tax on $90. Mathematically, this difference arises because alimony reduces the NYAGI denominator, increasing the portion of the tax base owed; itemized deductions, by contrast, do not reduce NYAGI.
To achieve parity with the treatment of itemized deductions, the formula should add alimony back to the denominator. What Lunding and the court seem to do, however, is to subtract a portion of the alimony (100 percent here, because all the income is New York sourced) from the numerator.
Two further peculiarities of the math in Lunding bear mention. First, in the course of Lunding it was noted that, in certain cases, the New York source fraction can exceed 100 percent. This is true, but the constitutional significance of this fact is not particularly clear. If in our example the taxpayer had $200 of New York income but sustained $20 of out-of-state business losses, his formula under Lunding would be as follows:
"As If" income = |
$200 (20) loss (70) alimony (10) interest & taxes $100 |
NYSI = |
$200 (70) alimony $130 |
NYAGI = |
$200 (20) loss (70) allocated alimony $110 |
$100 x 7.875% x 130/110 = $9.31 |
It appears the nonresident in this example is taxed more heavily than a resident -- a resident with $100 of income would pay only 7.875 percent in tax. The culprit here, however, is the $20 out-of-state loss, which is not taken into account in computing the NYSI numerator, but is reflected in NYAGI. States have no obligation to allow deductions for business losses associated with other states, however, so the result here is constitutional.
In his brief to the Supreme Court, Mr. Lunding invoked the +100 percent phenomenon through an example of a nonresident husband all of whose income was derived in New York, and 20 percent of which was paid in alimony. The effect of New York's denial of an alimony deduction to this hypothetical taxpayer was to tax 125 percent of his "as if" income. It looks like the +100 percent fraction is faulty, but in reality this is just another way of saying he was taxed on his income without any alimony deduction, which would be valid if alimony were comparable to out-of-state business losses.
The Supreme Court corrected the alimony-based "problem" without finding fault with the fraction per se. The possibility of a fraction greater than 100 percent should therefore be tolerable where this stems from the numerator's ignorance of expenses a state is not required to allow to nonresidents. That said, it still seems possible that, on some set of facts, the application of New York's formula may indeed be shown to lead to the taxation of income otherwise barred to the state. (See, e.g., British Land v. Tax Appeals Tribunal, 85 N.Y. 2d 139 (1995), which overrode New York's three-factor formula where the effect was an unreasonable taxation of extraterritorial gain.)
Finally, the workings of Lunding's proration of alimony in the case of a joint return are unclear. Tax Law section 651(e)(2) and (4), effective July 31, 1992, generally mandate joint New York returns for couples filing jointly for federal purposes, including nonresident couples. Couples with one resident and one nonresident spouse (the key to many successful unions) may file separately if they wish, but where both spouses are nonresidents and file a joint federal return, the Tax Law mandates a joint New York return.
This raises the interesting possibility that the income, expenses, etc. of the new spouse might have a mathematical effect on the deductibility of the alimony due to a prior spouse. At its extreme, the alimony paid by a nonresident husband to his nonresident former spouse might find itself more heavily allocated to New York by virtue of the New York earnings of the second spouse. This seems inappropriate where (as seems likely) an alimony award does not consider and is not paid out of the earnings of the second wife; but we need more definition to unravel the rules post-Lunding.
Science
Brother Mendel's box for analyzing the inherited traits of peas is useful also in analyzing the pre- and post-Lunding tax treatment of different combinations of exes. There are four possible combinations, as follows:
|
Husband |
||
Resident |
Nonresident |
||
Wife |
Resident |
1 HRWR |
2 HNWR |
Nonresident |
3 HRWN |
4 HNWN |
Under New York's taxing scheme pre-Lunding, HR's were permitted to deduct alimony without regard to the wife's residence; and WR's were required to include alimony received in taxable income, without regard to the deductibility of such payments by H. Thus, under the state's statutory scheme, the net amount of New York State income given $100 of alimony (and ignoring all else) is as follows:
|
Husband |
||
Resident |
Nonresident |
||
Wife |
Resident |
1 0 |
2 100 |
Nonresident |
3 (100) |
4 0 |
By contrast, under Lunding the net amount of New York income given $100 of alimony is as follows:
|
Husband |
||
Resident |
Nonresident |
||
Wife |
Resident |
1 0 |
2 0 |
Nonresident |
3 (100) |
4 (100) |
The state's statutory scheme, illustrated in the first boxed set of numbers, achieved parity in treatment of resident and nonresident couples (compare boxes 1 and 4). However, New York treated its resident husbands more favorably than nonresident husbands (compare Box 1 with Box 2, and Box 3 with Box 4). It also happened that New York treated its resident wives less favorably than nonresident wives (compare boxes 1 and 3, and boxes 2 and 4). By contrast, under Lunding, equality of overall tax burden now correlates with the residence of the wife, not the residence of the couple. And New York either offsets the husband's alimony deduction with the wife's income (if she is a resident), or grants the husband (resident or not) a deduction and foregoes taxation of the offsetting income of the nonresident wife.
Mr. and the former Mrs. Lunding occupied Box 4 -- neither spouse was a resident of New York. The dissent characterized Lunding as arguing for a comparison of his situation to Box 3, where the resident wife pays tax even though the nonresident husband has no deduction. He also argued, "more engagingly," that his box should be compared to Box 2, where the resident husband is allowed an alimony deduction even though the wife is a nonresident. As criticized by the dissent, "New York's choice, according to Lunding, is to deny the alimony deduction to the New Yorker whose former spouse resides out of state [Box 2], or else extend the deduction to him. The Court apparently agrees."
The dissent, however, believed New York "could legitimately assume that in most cases, as in the Lundings' case, payer and recipient will reside in the same State." Moreover, "in cases in which the State's system is overly generous [Box 3] or insufficiently generous [Box 2], there is no systematic discrimination discretely against nonresidents for the pairs of former spouses in both cases include a resident and a nonresident."
The dissent's argument in Lunding interweaves the inquiry into whether discrimination exists with the inquiry into the state's justification for discriminating; its focus, however, was clearly on the components in each box -- both H and W. In the view of the dissent, there was no "systematic discrimination discretely against nonresidents," because the unequally treated couples include both residents and nonresidents.
The majority, however, seemed to focus only on comparing H's, not couples, and on the state's inability to articulate an adequate justification for treating HR more favorably than HN. In analyzing the constitutionality of distinctions made among the boxes, therefore, the question of how to compare them may be outcome-determinative.
Social Studies
The nature of alimony is a central question in the Lunding debate. As illustrated above, in considering the constitutionality of state law disallowances of exemptions or deductions to nonresident individuals, the law had heretofore made distinctions among personal exemptions, personal expenses, in-state business expenses or losses, and out-of-state business expenses. Discrimination by disallowing personal exemptions translated into a different rate structure for nonresidents, and was forbidden. Discrimination by disallowing deductions of a "purely personal" nature was upheld, particularly where the expense was clearly related to a different state -- the chief example being Mr. Goodwin's interest and taxes on his New Jersey home. Business expenses and losses unrelated to in-state business or income likewise could properly be denied to nonresidents, as a necessary and reasonable correlation to the state's inability to tax the nonresidents' income and gains from other state sources. Expenses "connected" with an in-state business were, however, required to be allowed to offset the nonresident's taxable in-state income.
While not clearly its stated rationale, the majority's decision in Lunding seems influenced by the notion that the alimony paid by Mr. Lunding was related to or connected with the income he derived from his New York law firm. The dissent characterized the majority's approach as a "'personal expense deduction' in lieu of 'income attribution' categorization of alimony..." (at 60), but clearly the majority attached some significance to the fact that alimony has a "correlation" to income.
This then raises the issue of how best to categorize alimony, and its relationship to a given state. Certainly as a practical matter alimony relates in many cases directly to the payer's earnings, and usually is paid primarily out of those earnings. The simple analysis of alimony as a division (some might say siphoning off) of earnings between individuals has a great deal of commonsense appeal. On closer scrutiny, however, the analysis is not so simple.
Clues to the nature of alimony are found both in the state laws to which the couple's divorce is subject, and in the federal income tax treatment of alimony. In Connecticut, where the Lundings resided, as in most other jurisdictions, settlement of the marital rights of husband and wife following the dissolution of a marriage can involve both a property settlement, under which the accumulated assets of the couple are divided, and an award of alimony and maintenance. The professional credentials of one party to the marriage are not property subject to division, Simmons v. Simmons, 244 Conn. 158 (1997). However, the future earnings capacity of that professional is something legitimately considered in awarding alimony.
Indeed, the decision in Simmons sounds very much like alimony is the wife's share of the husband's earnings, to be adjusted as his earnings change. "The primary aim of property distribution is to recognize that marriage is, among other things, 'a shared enterprise or joint undertaking in the nature of a partnership to which both spouses contribute -- directly and indirectly, financially and nonfinancially -- the fruits of which are distributable at divorce.'" Krafick v. Krafick, 234 Conn. 783 at 795 (1995). Alimony, by contrast, "is a proper means of sharing in the future earnings of a spouse..." Simmons, at 182, and "an appropriate method to take into account future earning capacity because it can be modified whenever there is a change in the circumstances of the parties that justifies the modification....The [husband] may become disabled, die or fail his medical boards and be precluded from the practice of medicine.He may choose an alternative career either within medicine or in an unrelated field or a career as a medical missionary, earning only a subsistence income. An award of alimony will allow a court to consider these changes if and when they occur." Id., at 184.
This articulation of alimony dovetails with the majority's treatment of alimony as having some relationship to the state in which the husband's earnings are sourced. There are, however, other factors to consider in analyzing the connection between alimony and a given state.
In his brief, Mr. Lunding noted that New York's statute "makes no distinction in its application between nonresidents who were married and divorced in New York or whose sole source of personal income is in New York and other nonresidents who were married or divorced elsewhere or whose personal income was generated from more than one State." Petitioner's brief at 11-12. If the appropriate tax analysis is to search for a legitimate state interest in advancing a particular social or governmental policy, then it may be correct to question what justification there is for a provision that disallows alimony based solely on the residence of the husband at the time it is paid.
In Lunding, this point did not advance too far, given that both the marriage and the divorce occurred entirely in Connecticut, and his former wife remained resident in Connecticut. As argued by New York State, Lunding's attempt to assert the Full Faith and Credit Clause as importing his Connecticut divorce into New York does seem a completely inappropriate basis for establishing a relationship between the alimony and the state. "[S]uch link is too attenuated to be constitutionally significant....Although the amount of alimony payments may have been based on Mr. Lunding's income from New York (and elsewhere), the origin of the claim giving rise to the alimony obligations was entirely unrelated to Mr. Lunding's income producing activities in New York." Respondent's brief at 30.
The facts of Mr. Lunding's family life may be irrelevant to the analysis of New York's statute, inasmuch as the statute took the sledgehammer approach and denied all nonresidents everything, without considering possible New York connections. Writing on a blank slate, however, a state might fare better if it were to link its denial (or allowance) of alimony more specifically to the couple's relationship to the state. The origin of a wife's claim might easily be in New York even if the husband has since moved out.
Absent such fine-tuning, and taking into account the factors courts consider in awarding alimony, alimony can indeed look like a wholly personal obligation best "sourced" to the husband's residence. Under Connecticut law, the amount of the wife's entitlement may depend on the husband's ability to pay, but clearly alimony is not her "partnership" share of the husband's earnings. Moreover, in fixing the amount of alimony the statute directs that the courts consider "the length of the marriage, the causes for the [divorce], the age, health, station, occupation, amount and sources of income, vocational skills, employability, estate and needs of each of the parties, and the [property settlement]." [Conn. Stat. section 46b-82]. If more alimony is paid because the husband has parted with less (out-of-state) property in the property settlement, does that not mean that the alimony relates not only to his in-state earnings but also to the out-of-state assets retained? If more alimony is paid as a result of the "causes" for the divorce, is that expense legitimately chargeable to the offender's in-state business activities? And what does it mean when, as in Simmons, more alimony is ordered because the wife is 56 and the medical degree earned by her 36-year-old husband was her "retirement plan"? Simmons, at 180.
Alimony is, therefore, almost as complex as its causes. As a result, while it is tempting to think of Lunding as "right" because alimony looks a lot like an offset to the husband's earnings, that is not necessarily the true nature of the payment.