Brett Turner, Senior Attorney, Family Law, National Legal Research Group
This is a long-term series of posts on one of the hottest topics in modern American family law—same-sex marriage. It began in the aftermath of United States v. Windsor, 133 S. Ct. 2675 (2013). Windsor held that the federal government must defer to state law on the validity of same-sex marriage, and it struck down a federal statute providing that the federal government can never recognize a same-sex marriage. But if state law is now controlling, the immediate next question is, which state's law? This blog began as an attempt to collect case law answering that question.
But this purpose was quickly overtaken by events. Perhaps motivated in part by the difficulty of determining which state's law applies, the post-Windsor case law has held with near-uniformity that the states cannot constitutionally restrict same-sex marriage at all. If this trend continues, the federal courts will soon force all states not only to recognize out-of-state same-sex marriages but also to permit such marriages by their own citizens.
This blog therefore evolved into a more general attempt to collect post-Windsor case law on same-sex marriage. As the reader will see, the author has grown somewhat critical of the current trend to resolve the nationwide dispute over same-sex marriage by federal court decision. Same-sex marriage is almost certainly coming, and coming quickly, but it will come with more legitimacy and ultimately more support if it comes through the normal processes of state law, at least until the number of states recognizing same-sex marriage reaches 50%.
Please scroll down past the "Read More" to find articles discussing the various aspects of United States v. Windsor.
Brett Turner, Senior Attorney, National Legal Research Group
Two more decisions have held that the U.S. Constitution requires broader acceptance of same-sex marriage than state law presently permits.
In Bourke v. Beshear, No. 3:13-CV-750-H, 2014 WL 556729 (W.D. Ky. Feb. 12, 2014), the court held that Kentucky was required to recognize out-of-state same-sex marriages, and that Kentucky's constitutional and statutory provisions barring such recognition are unconstitutional. The court relied entirely upon a Fourteenth Amendment due process theory. Because no clear majority of Justices in Windsor applied a higher standard, Bourke used the rational-basis test. It nevertheless found no rational basis for refusing to recognize out-of-state same sex marriages. The effect of the Kentucky provisions was to identify a group of state-sanctioned marriages (same-sex marriages from other states) and subject them to unequal and demeaning treatment. The only justification alleged in support of the provisions was to preserve a traditional definition of marriage as a relationship between persons of opposite sexes. But the court held that tradition alone cannot justify discriminatory treatment.
Interestingly, Bourke did not expressly rely on a finding that Kentucky's provisions against same-sex marriage had been enacted with discriminatory intent. It is therefore a stronger opinion than that in Obergefell v. Kasich, which this blog criticized here.
But Bourke still ignores that portion of the Windsor opinion that held that recognition of marriages is uniquely a subject for state law. That portion clearly implies that states may choose either to recognize or not recognize same-sex marriage. If there is literally never any rational basis for recognizing out-of-state same-sex marriages—and that is what the Bourke court held—then Windsor would not have placed so much emphasis upon state law.
In this author's opinion, the federalism portion of the Windsor opinion was not intended to be meaningless. The validity of marriages is a subject over which state legislatures have particularly great power, and a subject over which the federal government has particularly little power. A strong majority of Kentucky citizens have chosen to place restrictions upon the recognition of same-sex marriage. The decision made by these citizens should not lightly be discarded as irrational.
There are especially strong practical reasons why federal courts should not dismiss all opposition to same-sex marriage as irrational. Recognition of same-sex marriage is a fundamental social change. In a democracy, fundamental social changes should result from elections, not from court decisions.
Only in 2012 did same-sex marriage begin to show strength at the ballot box. This showing was very important, for it offered a path to recognition of same-sex marriage that was relatively free of divisive controversy. If a majority of voters choose to reject restrictions upon same-sex marriage, that rejection will be a clear basis for accepting fundamental social change. If the voters reject same-sex marriage in a sufficiently large number of states, that may even provide a basis for applying constitutional theories to force recognition of same-sex marriage nationwide.
But it is important for the legitimacy of the process that additional elections be held on same-sex marriage. The strong wave of post-Windsor decisions striking down restrictions upon same-sex marriage raises a very real possibility that no more elections will be held and that the federal courts will declare the issue resolved. Such a declaration would substitute judicially imposed change, which is relatively less legitimate, for electorally or legislatively imposed change, which is relatively more legitimate. The dangers of judicially imposed change are especially strong when the decisions hold that all opposition to same-sex marriage, such as that expressed by a strong majority of Kentucky voters, is irrational—and that is what the Bourke court held.
From all evidence visible to the author, the nation is moving rapidly toward widespread recognition of same-sex marriage. But it is important for the perceived legitimacy of the change that the change occur as democratically as possible. Democratic change is impeded, not faciliated, when federal courts insist that all opposition to same-sex marriage is irrational.
The questions remains, if opposition to same-sex marriage is rational, what is the precise rational basis for such opposition? The federalism portion of Windsor holds clearly that recognition of marriage is a subject on which states and state legislatures have particular power. In enacting restrictions upon same-sex marriage, state governments have been responding to the fear on the part of their constituents that recognition of same-sex marriage will pose substantial dangers to society. Widespread recognition of same-sex marriage is a recent development, historically speaking; there is reason to proceed with caution.
The federal court decisions to date have been inclined to take a hard look at the dangers of same-sex marriage, and conclude that the dangers are so imaginary that opposition to same-sex marriage is irrational. But state law is entitled to particular deference on matters involving the validity of marriages, and it is especially important to the legitimacy of fundamental social change that the change be ratified by as many elections as possible. Against this background, fears that widespread recognition of same-sex marriage will change society in harmful ways are not entirely irrational.
The author therefore respectfully submits that the federal courts should not hold that all opposition to same-sex marriage is without rational basis. The courts should take a less assertive approach to this issue, and allow the democratic process to work its will for another few years.
The above analysis assumes, however, that a rational-basis test applies. In Bostic v. Rainey, No. 2:13CV395, 2014 WL 561978 (E.D. Va. Feb. 13, 2014), the court held that the federal Constitution requires Virginia to allow same-sex marriages of its own citizens, even though both a constitutional and a statutory provision bar such marriages. Bostic held flatly that the right to marry is a fundamental right. It did not, therefore, apply rational-basis review. Rather, the court applied strict scrutiny. It then found that preserving traditional marriage is not a sufficiently powerful goal to survive strict scrutiny. Like the Utah opinion in Kitchen v. Herbert, the opinion was consciously based upon Loving v. Virginia, 388 U.S. 1 (1967), and indeed the opinion begins with a quotation from the wife in Loving.
Of all of the rationales yet expressed for mandating recognition of same-sex marriage, the Kitchen/Bostic rationale is clearly the strongest. If persons of the same sex have a fundamental right to marry one another, then tradition alone is not the sort of compelling reason the law has traditionally required to justify interference with such a right.
Yet one might legitimately ask why the right to marry a person of the same sex is deemed fundamental. As of this writing, 17 states have recognized same sex marriage by normal operation of state law. That means that 33 states do not recognize same-sex marriage—well over half of all American states. By contrast, when Loving was decided, 35 states allowed interracial marriage and only 15 states did not. In short, when Loving was decided, a large majority of American states allowed interracial marriage.
At present, a nearly equally large majority of American states do not allow same-sex marriage. This difference in public support suggests that the analogy to Loving should be used with caution. Bostic started its opinion by noting that "[a] spirited and controversial debate is underway regarding who may enjoy the right to marry in the United States of America," 2014 WL 561978, at *1, but the court then cut that debate off by asserting that it is resolved by the federal Constitution.
Supporters of same-sex marriage would probably argue that the right to marry a person of the same sex is objectively fundamental, regardless of what a majority of the voters think. The author is not a constitutional originalist, and believes that the definition of a fundamental right can change over time in response to changes in society itself. But surely the bar for recognition of a new fundamental right should be relatively high. When a right is not yet recognized by a majority of states, courts should be cautious about declaring the right fundamental. That is especially so when proponents of same-sex marriage are showing strength at the ballot box—raising the possibility of an electoral resolution of the debate over same-sex marriage that will have much more legitimacy than any court decision will.
At the same time, it might also be noted that there does not seem to be widespread outrage over post-Windsor case law striking down restrictions upon same-sex marriage. Republicans in Congress seem more interested in Obamacare and budget issues than in attacking same-sex marriage. Protests over recognition of same-sex marriage are not dominating the nightly news. Even outgoing Virginia Attorney General Ken Cuccinelli, a determined opponent of same-sex marriage, wrote only a short summary opinion rejecting the possibility, missing the chance to make more determined protest. If social conservatives seem disposed to accept the validity of the recent lower court decisions, perhaps legitimacy is a less important concern.
But the fact still remains: In the United States, divisive public issues are resolved by elections. Same-sex marriage did well at the polls in 2012, but that was only one election, in which only four states voted on the issue. The debate over same-sex marriage would be resolved in a better way, a more legitimate way, a way more in keeping with the American tradition, if a broader base of states were permitted to vote.
And that is the fundamental problem with the recent wave of decisions holding that states are required to recognize same-sex marriage. It ends an important and necessary debate early, with an arbitrary decision by the courts that one side of the debate is taking an irrational position. The people whose position is being called irrational are likely to disagree with this reasoning, and to reject the legitimacy of these decisions. The result may well be greater social opposition to same-sex marriage, even after the purely legal issue is resolved.
The more prudent approach, therefore, is to let the normal political processes of state law run their course. If current trends continue, within only a few more years, 50% or more of the states will recognize same-sex marriage under state law. At that point, when support for same-sex marriage begins to approach the same levels as support for interracial marriage did in 1967, it would then be time to recognize a fundamental right to marry a person of the same sex.
This approach will admittedly delay widespread recognition of same-sex marriage for a few more years. But supporters of same-sex marriage will then be able to say to opponents, "We resolved this issue in the traditional American way, by holding a series of elections, and the people decided." That seems likely to produce a better world, one in which same-sex marriage is more accepted socially, than a world in which the courts declare the contest over after same-sex marriage won only a single election (2012) in which it was on the ballot in only four states.
* * * *
The author of this blog received a certain amount of criticism from opponents of same-sex marriage for writing that former Virginia Attorney General Ken Cuccinelli's opinion rejecting same-sex marriage was "[a] notable step backward." The author stands by this characterization. Cuccinelli's opinion is problematic even for opponents of same-sex marriage, because it is such a tremendous missed opportunity to write a reasoned defense of provisions restricting such marriage. Given the strong trend in federal court case law, opponents of same-sex marriage could use such a defense. Cuccinelli could have provided one—but mostly, he did not.
The present post may well attract criticism from the other side of the debate.
This blog remains committed to providing objective coverage of post-Windsor case law, but it is not possible to blog on so divisive an issue without developing an opinion. Discussing that opinion openly gives readers information they need to evaluate the information provided by the blog.
The author supports same-sex marriage, and would vote for it if the issue were to arise in an election. Increasingly, however, the author is concerned that the federal courts are declaring an end to the debate over same-sex marriage too quickly, without allowing sufficient time for the debate to be addressed through the normal electoral process. Because an electoral resolution of this divisive issue is greatly to be preferred to a judicial resolution, recent decisions have been too quick to hold that restrictions upon same-sex marriage are unconstitutional.
Brett Turner, Senior Attorney, National Legal Research Group
A federal district court in West Virginia has released an opinion addressing questions of abstention and standing in same-sex marriage litigation. McGee v. Cole, Civ. No. 3:13-24068, 2014 WL 321122 (S.D. W. Va. Jan. 29, 2014). The opinion does not reach the merits.
McGee is an action by gay couples living in West Virginia, challenging that state's statutory restrictions upon same-sex marriage. The plaintiffs argue both that the restrictions upon same- sex marriage are substantively unconstitutional and that the restriction against the recognition of out-of-state same-sex marriage is procedurally unconscionable.
The defendants moved to dismiss the complaint on the ground of Burford abstention. Burford abstention applies when a case presents difficult questions of state law, the importance of which transcends the case at hand, or where adjudication in a federal forum would disrupt a state’s attempt to establish coherent policy on a matter of public importance. First Penn-Pac. Life Ins. Co. v. Evans, 304 F.3d 345, 348 (4th Cir. 2002).
The court rejected the first basis for Burford abstention. McGee did not present a difficult question of state law. West Virginia state law is perfectly clear: Same-sex marriages are never recognized. The issue is whether West Virginia state law violates the federal Constitution. That is a question of federal law, not state law.
The court also rejected the second basis for Burford abstention. There have been sporadic attempts either to repeal the West Virginia statutes or to write them into the state constitution, and the defendants argued that these attempts should be respected and permitted to continue. "Defendants point to no cases, however, suggesting that legislative efforts to define domestic relations justify federal court abstention, and the Court has not found any." McGee, 2014 WL 321122, at *7.
It would also seem that if some sort of deference to state legislative action is appropriate, it should be appropriate only where that action is imminent. For example, if the West Virginia legislature were about to vote on a bill to repeal the statutes at issue, it might in some cases make sense to refrain from addressing the constitutional issue until the vote is held. But West Virginia legislative activity on same-sex marriage has been very sporadic, and no changes are imminent. To abstain from hearing a federal constitutional challenge in that situation would greatly reduce the reach of the federal Constitution as the mere possibility of state legislative action, combined with some sporadic history of past legislation (some of which on the facts of McGee resulted from attempts to strengthen the provision being challenged) would paralyze the federal courts. Burford abstention should not be required by the sort of sporadic former legislative action that was present in McGee.
The defendants also argued that the plaintiffs sued only the court clerks of two counties. Any ruling in favor of the plaintiffs would apply only in those counties and would therefore frustrate West Virginia's attempts to have a uniform policy on same-sex marriage. The court agreed with this concern and directed the plaintiffs either (1) to add sufficient additional defendants so that the lawsuit would affect the entire state of West Virginia, or (2) to provide legal authority that the court's ruling would be binding statewide.
On the first point, the court noted that in Zablocki v. Redhail, 434 U.S. 374 (1978), which struck down a state statute requiring persons under a child support order not having custody of their children to obtain court permission before remarrying, the suit was a class action against all of the marriage clerks in the state. On the second point, the State of West Virginia had intervened to defend the statutes, and the court was uncertain whether the State's intervention would make an opinion in the plaintiffs' favor binding statewide. In any event, the court clearly wanted the plaintiffs to take all possible actions so that the result of the case would be binding statewide.
Baker v. Nelson
The defendants also argued that the result of the case should be controlled by Baker v. Nelson, 409 U.S. 810 (1972), which dismissed for want of a substantial federal question an argument that the Fourteenth Amendment required the state to recognize same-sex marriage. Relying expressly on the Utah decision in Kitchen v. Herbert and the Oklahoma decision in Bishop v. United States, the McGee court disagreed. Baker considered only sex discrimination, not discrimination against gay people. In addition, developments in Supreme Court jurisprudence since 1972—especially the recognition of "intermediate tier" scrutiny for sex discrimination and Windsor's failure to find itself bound by Baker—suggested that Baker is no longer binding.
Finally, the State moved to dismiss that portion of the plaintiffs' complaint alleging the unconstitutionality of West Virginia Code section 48-2-603, which provides that West Virginia will never recognize an out-of-state same-sex marriage. The fundamental problem was that none of the plaintiff couples had been married outside West Virginia. Since none of the plaintiff had out-of-state same-sex marriages, they were not injured by the statute they were attacking.
The plaintiffs argued that they did not need to obtain out-of-state marriages because the attempt would be futile; everyone agreed that West Virginia would not recognize the marriages. "The Court finds that even if a certain activity is futile, a plaintiff must nonetheless demonstrate willingness to engage in the activity were it not for a barrier in place that makes the activity futile." McGee, 2014 WL 321122, at *14. The plaintiffs did not even allege that they wanted to get married out of state. Moreover, there was no suggestion that the plaintiff would suffer any form of prejudice from obtaining an out-of-state same-sex marriage or even that obtaining such a marriage would place a material burden upon them.
The court therefore held that the plaintiffs lacked standing to challenge the statutory provision barring recognition of out-of-state same-sex marriages. The clear lesson is that future challenges to statutes preventing interstate recognition of same-sex marriages should be brought by plaintiffs who have actually entered into an out-of-state same-sex marriage that they seek to require the defendant state to recognize. This does not seem like an unreasonable requirement.
The plaintiffs' standing to challenge the statutory provisions preventing them from entering into a same-sex marriage in West Virginia
of course was not affected by the ruling. The action remains pending, and a decision on the merits should eventually be forthcoming.
Brett Turner, Senior Attorney, National Legal Research Group
As the courts struggle to implement the Windsor opinion, there will inevitably be steps forward and steps backward. A notable step backward is Virginia Attorney General Opinion 13-102, released by outgoing Virginia Attorney General Ken Cuccinelli on January 10, 2014—the very last day of his tenure as attorney general.
The opinion answers an inquiry from a Virginia state senator, asking whether "public sector health plan sponsors" in Virginia must follow United States Department of Labor Technical Release 2013-04. That Release provides that administrators of benefit plans regulated by the federal Employee Retirement Income Security Act of 1974 (“ERISA”) must recognize same-sex marriages if those marriages are valid in the state where the marriage took place.
Most benefit plans regulated by ERISA are private plans, but federal law also regulates certain public sector health plans in states that accept federal funding under the Public Health Service Act. E.g., 42 U.S.C. § 300bb‑1. By accepting federal funds for public health, these states have voluntarily agreed that group health plans for public sector employees will be subject to federal requirements.
Mr. Cuccinelli opined that Virginia administrators of public sector health plans need not follow the position taken by the Department of Labor—the federal agency responsible for construing ERISA. He took the position that Windsor did not require any state to accept same-sex marriage and that section 2 of the Defense of Marriage Act (“DOMA”) remains valid. That statute provides that federal law shall not require any state to recognize same-sex marriage. He finally cited Virginia's constitutional provision against same-sex marriage, Va. Const. art. I, § 15-A, which prevents any recognition of same-sex marriage.
Given that federal law necessarily controls the operation of federally regulated benefit plans, Mr. Cuccinelli's insistence that state law nevertheless be followed brings to mind John C. Calhoun's insistence that states can nullify operation of any federal law within their borders—a doctrine that did not survive the Civil War.
As interesting as the position that Mr. Cuccinelli took is the position that he chose not to take. Federal regulation of public sector health plans appears to be rather narrow, focusing mainly upon rights to continued coverage after termination. 42 U.S.C. § 300bb‑1. In addition, all group health plans, whether public or private, are required to follow certain basic requirements, such as avoiding any exemption for preexisting conditions, see 42 U.S.C. § 400gg-3, as a result of the health-care reform law popularly known as Obamacare.
It is far from clear that federal regulation of public sector health plans is sufficiently comprehensive to require such plans to universally follow the federal definition of “spouse” in situations in which the state definition is more restrictive. Mr. Cuccinelli could have stressed the limited nature of federal regulation of public sector plans, or have even raised constitutional issues as to whether the federal government can properly insist that state public sector plans follow the federal definition of “spouse.” Instead, the opinion seemed to accept that the plans were federally regulated and he then asserted that the plans must nevertheless follow conflicting state law.
Mr. Cuccinelli's construction of Windsor is also strikingly inconsistent with most existing post-Windsor case law. Provisions similar to the Virginia provisions on which Mr. Cuccinelli relied have been held unconstitutional in Ohio and Oklahoma on the basis that they were enacted with the intent to discriminate against gay persons, and in Utah on the basis of a broad federal right to same-sex marriage. Mr. Cuccinelli did not cite these decisions or argue that they were wrongly decided. He also failed to cite at least one case that essentially adopts his position that state law restrictions upon same-sex marriage remain valid. Had the opinion contained more legal analysis, it might have been a useful contribution to public debate over the scope of Windsor and the future of same-sex marriage.
The timing of the opinion is also remarkable. Mr. Cuccinelli is Virginia's outgoing Attorney General. He lost that position when he ran, unsuccessfully, for Governor of Virginia. Virginia's new administration took office on "the Saturday after the second Wednesday in January," Va. Const. art. V. § 1—which was January 11, 2014. The opinion was issued on January 10, which was literally Mr. Cuccinelli's last day as Attorney General. Mr. Cuccinelli is an outspoken social conservative, and Opinion 13-102 appears to be something of a last shot against the recognition of same-sex marriage—which does not even contain enough analysis and argument to potentially move the public debate in Mr. Cuccinelli's direction.
Virginia's incoming Attorney General, Mark Herring, campaigned on a platform of marriage equality, and he has filed a brief arguing that Virginia's constitutional provision against same-sex marriage is unconstitutional. It therefore seems likely that future Virginia Attorney General opinions on the subject of same-sex marriage will be very different from those written by Mr. Cuccinelli.
Brett Turner, Senior Attorney, National Legal Research Group
A federal district court in the Northern District of Oklahoma has held that Oklahoma's constitutional provision barring same-sex marriage violates the federal Constitution. Bishop v. United States ex rel. Holder, No. 04-CV-848-TCK-TLW, 2014 WL 116013 (N.D. Okla. Jan. 14, 2014).
The rationale is based heavily upon the second part of the Windsor opinion, which held that DOMA section 3 was invalid because it was enacted with intent to discriminate against gay couples. The court cited strong evidence that the purpose of the Oklahoma provision was to express moral disapproval of same-sex marriage. The court continued:
The Court recognizes that moral disapproval often stems from deeply held religious convictions. See Lawrence [v. Texas], 539 U.S. , 571 [2008)] (explaining that moral disapproval of homosexual conduct was shaped by "religious beliefs, conceptions of right and acceptable behavior, and respect for the traditional family"). However, moral disapproval of homosexuals as a class, or same-sex marriage as a practice, is not a permissible justification for a law. See Lawrence, 539 U.S. at 577 ("'[T]he fact that the governing majority in a State has traditionally viewed a particular practice as immoral is not a sufficient reason for upholding a law prohibiting the practice.'") (quoting and adopting Justice Stevens' dissent in Bowers v. Hardwick, 478 U.S. 186, 216, 106 S.Ct. 2841, 92 L.Ed.2d 140 (1986)). Preclusion of "moral disapproval" as a permissible basis for laws aimed at homosexual conduct or homosexuals represents a victory for same-sex marriage advocates, and it forces states to demonstrate that their laws rationally further goals other than promotion of one moral view of marriage. Therefore, although Part A rationally promotes the State's interest in upholding one particular moral definition of marriage, this is not a permissible justification.
Id. at *27 (footnote omitted) (some citations omitted).
The court's direct holding that legislation cannot be based upon moral disapproval of same-sex marriage is extremely broad. Since moral disapproval is the primary basis for all restrictions upon same-sex marriage, the court's reasoning would strike down all such restrictions, creating in effect a broad federal right to same-sex marriage. But the court's interpretation of Windsor is open to question. Windsor discussed at considerable length the long tradition of treating marriage requirements as a subject for state law. If Bishop was correctly reasoned, then a federal right to same-sex marriage essentially exists, and the entire discussion of federalism in Windsor was irrelevant. In addition, Windsor itself could have created a broad federal right to same-sex marriage, and it did not. Thus, Bishop stretches the Windsor rationale to reach a result that the Windsor Court itself failed to reach.
Interestingly, Bishop recognized earlier in its opinion that "[a] state [restriction upon same-sex marriage] must be approached differently, and with more caution, than the Supreme Court approached DOMA" in Windsor. Id. at *19. But the court's very broad holding that state legislatures cannot constitutionally base legislation upon moral disapproval of same-sex marriage shows very little caution. If the federalism language in Windsor carries any weight whatsoever, the reasoning of Bishop is highly questionable.
In other words, Windsor recognizes that states have the power to decide who can and cannot be married. This determination has always been made with consideration of moral issues. For example, provisions barring bigamy and polygamy have traditionally been based upon moral disapproval of bigamists and polygamists. Moral disapproval is part of the rationale for state laws against incest. Morality has often been one relevant factor in deciding who can and cannot be married. If the state legislatures cannot consider morality when determining who can and cannot be married, the effect might be to overturn many provisions other than those barring same-sex marriage.
The author continues to believe that the preferable rationale for creating a federal right to same-sex marriage is Kitchen v. Herbert, 2:13-CV-217, 2013 WL 6697874 (D. Utah Dec. 20, 2013). Kitchen refused to hold that Utah's provision barring same-sex marriage was unconstitutional due to antigay prejudice, finding that the restriction was based upon a complex mixture of good and bad intentions. Instead, Kitchen held directly that same-sex couples have a fundamental constitutional right to marry. The holding was based consciously upon Loving v. Virginia, 388 U.S. 1 (1967), which recognized a similar right for mixed-race couples. Extending Loving to same-sex couples is the cleaner and more logical route for creating a broad federal right to same-sex marriage.
In sum, the fundamental problem with Bishop is that the reasoning goes too far and inflicts too much collateral damage. Some morality-based marriage restrictions, such as provisions that bar polygamy, are generally accepted. The fundamental problem with same-sex marriage is not that the legislature considered morality, but rather that it considered morality in a way that violated the fundamental right of same-sex persons to marry. The proper path for recognition of same-sex marriage is Kitchen, not Bishop.
Brett Turner, Senior Attorney, National Legal Research Group
In IRS Notice 2014-1, 2014‑2 I.R.B. 270 (Jan. 6, 2014), the IRS dealt with various consequences of the Windsor decision on same-sex couples who participate in benefit plans. The notice is based upon Revenue Ruling 2013-17, 2013‑38 I.R.B. 201, which generally provides that a same-sex marriage is valid under federal law if it is valid in the state in which it was celebrated, regardless of where the parties reside thereafter.
Two particular issues discussed in the Notice deserve special attention. First, the Notice suggests that for purposes of certain benefit plans allowing beneficiaries to change benefit options only under limited circumstances, such as a change in marital status, the Windsor decision will be treated as a change in marital status. For example, if a couple had a same-sex marriage which was valid under state law, that marriage was not recognized by federal law before Windsor was decided. When Windsor required the federal government to recognize the marriage, the parties' marital status changed for purposes of federal law, allowing them to select new benefit options, even though the parties' marital status under state law remained
Second, many taxpayers are finding significant financial advantages in high-deductible health insurance plans and their associated Health Savings Accounts (“HSAs”). Taxpayers make regular tax-contributions to HSAs, and the amount of these contributions is generally not taxable income. But the total annual amount of the contributions is limited in amount, and when either spouse elects family coverage, there is a single cap applying to contributions made by the entire family.
Assume that before Windsor, a same-sex couple enters into a valid marriage under state law. Both spouses are employed and both elect to create HSAs. Because they have children, at least one spouse elects family coverage. They then plan their deductions so that each spouse will have the maximum permitted amount withheld from his or her salary and deposited in the HSA. Since the parties' marriage is not recognized by federal law, each spouse makes the maximum contribution permitted for an unmarried person.
Then, Windsor is decided, and the parties' marriage is recognized by federal law, partway through the tax year. Because at least one spouse elected family coverage, there is a single combined maximum deduction which applies to the entire family. But this amount is materially less than twice the maximum deduction for unmarried persons, so that if the 2013 deductions occur as originally planned, the total deduction for both spouses in tax year 2013 will exceed the maximum allowed for family coverage.
Prudent taxpayers may have seen this problem coming and reduced their post-Windsor HSA deductions to fall within the lower maximum. But many taxpayers do not consider tax issues until the time comes to prepare and file returns.
IRS Notice 2014-1 confirms that if a same-sex couple in this situation exceeds the maximum allowed deduction for family coverage under an HSA, they must pay the excise tax charged on excess contributions. I.R.C. § 4973. It notes, however, that under I.R.C. § 223(f)(3), excess contributions can be distributed out of an HSA back to the owners at any time before the date for filing that's year's tax return. These distributions are treated as taxable income, but they are not subject to the excise tax.
Thus, if a same-sex couple contributes too much to an HSA because the recognition of their marriage by Windsor reduced the allowable maximum annual HSA contributions, the remedy is to distribute the excess contributions (and any income earned on those contributions) before the due date of the couple's 2013 tax return.
Brett Turner, Senior Attorney, National Legal Research Group
As previously discussed, a federal district court in Utah ruled that federal law requires Utah to recognize same-sex marriage. Kitchen v. Herbert, No. 2:13-CV-217, 2013 WL 6697874 (D. Utah Dec. 20, 2013). The court refused to stay its ruling pending appeal, as did the Tenth Circuit. But the U.S. Supreme Court intervened and ordered a stay. Herbert v. Kitchen, No. 13A687 (Jan. 6, 2013). There was no recorded dissent.
Unlike some, the present author sees the Supreme Court ruling as a caution sign directed at the lower federal courts. In recent months, at least two major federal decisions have construed Windsor to require that state law restrictions on same-sex marriage be held unconstitutional. In addition to Kitchen, see Obergefell v. Kasich, No. 1:13‑CV‑501, 2013 WL 3814262 (S.D. Ohio July 22, 2013). The effect of these rulings, and in the case of Kitchen the stated intent, was to create a broad federal right to same-sex marriage—that is, to extend to gay couples the reasoning of Loving v. Virginia, 388 U.S. 1 (1967).
But the blunt fact is, if Windsor had intended to apply Loving to gay couples, it could easily have done so. Instead of doing so, Windsor spoke at length about the long tradition of treating the validity of marriages as a subject for state law. That discussion is very inconsistent with any intent on the Court's part to hold that federal law requires recognition of same-sex marriage.
As this blog has previously noted, there are two parts to the Windsor opinion. The first part stresses federalism; the second part holds that section 3 of DOMA was unconstitutional because it was intended to discriminate against gay couples. The second part of the opinion is very broad, because essentially all restrictions on same-sex marriage were enacted with discriminatory intent on the part of at least some of their supporters. If not checked by the first part, the second part quickly leads to the position that all restrictions on same-sex marriage are invalid—which is tantamount to recognizing a federal right to same-sex marriage.
If the Supreme Court agrees that only the second part of the Windsor opinion matters, then Windsor essentially forces nationwide recognition of same-sex marriage in the very near future. If that was the majority's intent, it would seem logical to deny a stay in Kitchen.
By granting a stay, therefore, the Court suggests that the second part of Windsor does not automatically supersede the first part. The second part was the controlling rationale on the facts of Windsor, but that was true in part because the validity of marriages is a subject on which Congress has not traditionally legislated. State legislatures, by contrast, deal regularly and often with the validity of marriages. The validity of state law restrictions on same-sex marriage is therefore a different and somewhat closer question than the validity of federal law restrictions. It is not certain that the rationale of Windsor will apply to state law restrictions on same-sex marriage—and that is why the Court granted a stay.
The granting of a stay also does not mean, of course, that state law restrictions upon same-sex marriage will be upheld. The key question is whether the discussion of federalism in the Windsor opinion was intended as a material limitation on the holding. It is possible that the answer might be no—or that the answer might become no by the time the Supreme Court considers the question again.
But the granting of a stay in Kitchen does at least suggest the presence of a real and substantial question. If the discussion of federalism in Windsor was intended to have no weight, then the second part of Windsor probably would require nationwide recognition of same-sex marriage, and there would be no reason to grant a stay in Kitchen. By granting a stay, the Court suggested that the discussion of federalism in Windsor could potentially be a limitation on the remainder of the holding, and that the lower courts should not apply the second part without at least some consideration of federalism concerns.
Given the willingness of Obergefell in particular to apply the second part of the Windsor opinion without much consideration of the first part, the author reads granting of the stay as a caution sign. Windsor does not quickly and obviously lead to a federal right to same-sex marriage; the federalism portion of the Windsor opinion should be not summarily written off.
The ultimate answer to these questions must of course await the next Supreme Court same-sex marriage case—which could well be an appeal from the forthcoming Tenth Circuit ruling in Kitchen.
Brett R. Turner, Senior Attorney, National Legal Research Group
A previous entry in this series of posts discussed Obergefell v. Kasich, No. 1:13‑CV‑501, 2013 WL 3814262 (S.D. Ohio July 22, 2013). Obergefell granted a preliminary injunction, holding that Ohio was required to recognize a Maryland same-sex marriage. The parties to the marriage had almost no connection with Maryland. They flew there in an airplane, were married literally while the airplane sat on the tarmac, and then immediately flew back. The court reasoned, first, that Ohio's constitutional and statutory restrictions upon same-sex marriage were motivated by antigay prejudice, and second, that there is no rational basis for refusing to recognize gay marriage.
The final decision in Obergefell has now been handed down. Obergefell v. Wymyslo, No. 1:13-CV-501, 2013 WL 6726688 (S.D. Ohio Dec. 23, 2013). Not surprisingly, it reaches the same end result as the initial decision did. The decision expressly holds that when a couple is legally married in a state that permits same-sex marriage, the couple therefore acquires a constitutional right to remain married in other states. "The right to remain married is therefore properly recognized as one that is a fundamental liberty interest appropriately protected by the Due Process Clause of the United States Constitution. " Id. at *6 (emphasis added). Because the court viewed the plaintiffs as already married, it held that intermediate scrutiny should apply in an equal protection attack upon legislation restricting the right to remain married. It then held that the Ohio provisions failed both the intermediate scrutiny test and the rational basis test.
The opinion claims that "[t]he Court's ruling today is a limited one," id. at *1, but this is plainly not so. The plaintiffs in Obergefell had utterly no jurisdictional connection to Maryland. They were not domiciled there and did not live there. They went there only for the purpose of being married, and they left as soon as their purpose was accomplished. If Obergefell is broadly accepted, any same-sex couple in the nation can obtain a marriage that their home state is constitutionally required to recognize, simply by traveling to a state that recognizes same-sex marriage. The practical effect of Obergefell is to create a national federal constitutional right to same-sex marriage, so long as the actual license is issued by another state's clerk.
Despite the court's initial claim that its ruling was limited, the court recognized later in a footnote that its ruling is much broader.
As a final note, although the question of whether Ohio's refusal to grant same‑sex
marriages also violates Ohio same‑sex couples' right to due process and equal protection is not before the Court in this case, the logical conclusion to be drawn from the evidence, arguments, and law presented here is that Ohio's violation of the constitutional rights of its gay citizens extends beyond the bounds of this lawsuit.
Id. at *21 n.22. Given this statement, the court's initial claim that its ruling was limited borders upon disingenuous. Further recognizing that the actual effect of its decision is very broad, the court spent most of its opinion providing a legal justification for its holding that states cannot take away the right to remain married. The court's focus upon the right to be remarried is also somewhat misleading, for the court placed utterly no jurisdictional limitations on the right to get married. Since a couple can functionally be married anywhere, the right to remain married is for all practical purposes no different from the right to get married. If Ohio must recognize a Maryland same-sex marriage, even where the parties plainly traveled to Maryland only for the purpose of getting married, then functionally all Ohio residents have a right to obtain a same-sex marriage that Ohio must recognize.
Whether this holding survives appellate review remains to be seen. If five members of the Supreme Court had been inclined to create a federal constitutional right to same-sex marriage in 2012, Windsor would have been a very different opinion. If Obergefell were to be appealed to the 2012 Supreme Court, it would likely be reversed.
But of course, any appeal in Obergefell is likely to end up in the Supreme Court in 2014 or 2015. Even if there is no turnover on the Court before that time, developments after 2012 are likely to influence the decision.
In particular, it is very possible that by 2014 or 2015, the more liberal Justices on the Supreme Court—Justices Breyer, Sotomayor, Kagan, and Ginsburg—will be ready to hold that there is a fundamental right to same-sex marriage. It seems unlikely that the most conservative judges—Justices Roberts, Alito, Thomas, and Scalia—would reach this result.
The question therefore turns upon Justice Kennedy. The actual basis for Justice Kennedy's Windsor opinion is that DOMA section 3 was enacted out of antigay prejudice. As previous posts on this blog have noted, that reasoning is potentially quite powerful, because essentially all
restrictions upon same-sex marriage were motivated to some material extent by antigay prejudice. Justice Kennedy's opinion can easily be used to strike down any and all provisions limiting recognition of same-sex marriages.
But Justice Kennedy's opinion also included a long discussion of federalism, stressing that marriage has traditionally been a subject for state law. The key issue is whether this discussion was intended to materially limit the much broader reasoning in the remainder of the provision,
or whether it was intended to distract more conservative readers from the very broad reach of the remainder of the opinion. The author suspects that in 2012, Justice Kennedy was not willing to concede that the entire discussion of federalism was irrelevant. But the law on same-sex marriage is evolving, and restrictions upon same-sex marriage are weakening all over the country. By 2014 or 2015, Justice Kennedy might see that discussion of federalism is considerably less important.
The author would nevertheless have preferred that Obergefell make a straightforward argument for a federal right to same-sex marriage—essentially to extend to same-sex marriages the rationale of Loving v. Virginia, 388 U.S. 1 (1967). The actual Obergefell opinion stresses its limited nature, functionally creates a national right to same-sex marriage, and then essentially admits in a footnote that the decision is actually not limited. One hopes that if the court's reasoning is reviewed on appeal, the appellate courts will recognize the opinion for what it is, and not for what it claims to be.
The other big news of the week was Kitchen v. Herbert, No. 2:13-CV-217, 2013 WL 697874 (D. Utah Dec. 20, 2013), which held that Utah was required to allow same-sex marriage. Kitchen is another in a growing line of cases giving very little weight to the discussion of federalism in Windsor, and holding that states are not free to choose either to recognize or not recognize same-sex marriage.
Interestingly, Kitchen was skeptical of the argument that Utah's restrictions upon same-sex marriage had been motivated by antigay prejudice. The court recognized that some degree of prejudice was present, as it almost always is when these sorts of provisions are enacted. But the court continued:
But there are also reasons why Amendment 3 may be distinguishable from the laws the
Supreme Court has previously held to be discriminations of an unusual character. Most notably, the Court has not articulated to what extent such a discrimination must be motivated by a "bare . . . desire to harm a politically unpopular group." U.S. Dep't of Agric. v. Moreno, 413 U.S. 528, 534, 93 S.Ct. 2821, 37 L.Ed.2d 782 (1973). . . .
While the Plaintiffs argue that many Utah citizens voted for Amendment 3 out of a dislike of gay and lesbian individuals, the court finds that it is impossible to determine what was in the mind of each individual voter. Some citizens may have voted for Amendment 3 purely out of a belief that the amendment would protect the benefits of opposite-sex marriage. Of course, good intentions do not save a law if the law bears no rational connection to its stated legitimate interests, but this analysis is the test the court applies when it follows the Supreme Court's rational basis jurisprudence. It is unclear how a mix of animus and good intentions affects the determination of whether a law imposes a discrimination of such unusual character that it requires the court to give it careful consideration.
Id. at *23 (emphasis added).
This passage focuses directly upon the primary problem with the antigay prejudice theory. It is effectively certain that some voters and some legislators voted to restrict same-sex marriage out of antigay prejudice. But the mere fact that a law was supported by some small number of bigots does not make the law unconstitutional. As the court notes, a key question in applying the antigay prejudice theory would seem to be how many of a provision's supporters were motivated by bias.
In the specific case of measures restricting same-sex marriage, it is very possible that many persons supporting these measures had no prejudice against gay and lesbian persons generally, but simply supported the traditional rule that only opposite-sex couples can marry. Acceptance of this traditional belief is not equivalent to antigay bias. Windsor essentially recognized this by stressing that the validity of marriages is an issue of state law. If all opposition to same-sex marriage results from antigay bias, then the issue is not one for state law, and the Windsor majority opinion makes no sense. Courts should not invalidate restrictions upon same-sex marriage for antigay bias unless there is strong evidence that the restrictions
were primarily motivated by antigay prejudice. All opposition to same-sex marriage does not necessary arise from antigay prejudice.
The core reasoning of Kitchen is a straightforward argument that restrictions upon same-sex marriage violate the fundamental constitutional right to marry. The argument was expressly based upon Loving, and the court expressly noted that the State's argument was similar to the argument made against interracial marriage in Loving. Whether the appellate courts agree with the analysis remains to be seen. Kitchen is nevertheless a much more straightforward analysis than is Obergefell, and the author finds Kitchen much more persuasive.
Brett Turner, Senior Attorney, National Legal Research Group
The most recent American state to recognize same-sex marriage by statute is Illinois. A bill recognizing same-sex marriage was signed by the governor on November 20, 2013, but the bill does not take effect until June 1, 2014. S.B. 10, 2013 Ill. Legis. Serv. Pub. Act 98-597 (West) (codified primarily at 750 ILCS §§ 5/201, 5/212, 80/10).
Vernita Gray and Patricia Ewert have a long-standing same-sex relationship. In 2011, when Illinois first recognized same-sex civil unions, they immediately joined in such a union. When it became apparent that Illinois would soon allow same-sex marriages, they planned to enter into such a marriage.
But a cruel twist of fate intervened. Gray developed cancer, and her cancer advanced rapidly, leaving her with only weeks to live. It was gravely questionable whether Gray would live until June 1, 2014, the effective date of the new Illinois legislation. When the Cook County Clerk refused to issue an immediate marriage license, Gray and Ewert sued the Clerk in federal court, arguing that the Clerk's refusal to grant the marriage license violated due process and equal protection. To prevent irreparable harm, they sought a temporary restraining order ("TRO") directing the Clerk to issue them a marriage license.
The federal court granted the TRO. Gray v. Orr, No. 13 C 8449, 2013 WL 6355918 (N.D. Ill. Dec. 5, 2013). Failure to grant the TRO would cause irreparable harm, because Gray was
unlikely to live until June 1, 2014. If the plaintiffs were unmarried when Gray died, they would be denied important legal benefits such as "the right to take leave under the Family and Medical Leave Act, 29 U.S.C. § 2614(c)(1); the right to file a joint income tax return; spousal tax benefits such as exemption from certain estate tax obligations; and eligibility for Ewert for social security benefits as a surviving spouse." Id. at *4. The plaintiffs therefore had standing to seek the TRO.
The question then became whether the plaintiffs were likely to succeed on the merits. The Supreme Court held in United States v. Windsor, 133 S. Ct. 2675 (2013), that a restriction on recognition of same-sex marriage violates due process and equal protection if it is motivated by antigay prejudice. "An examination of the Illinois Marriage Act's history and the 1996 amendment declaring same-sex marriages to be against Illinois public policy reveals a similar animus towards same-sex couples." Gray, 2013 WL 6355918, at *4. Like Obergefell v. Kasich, No. 1:13-CV-501, 2013 WL 3814262 (S.D. Ohio July 22, 2013), discussed here, Gray suggests that all state law provisions barring same-sex marriage might be invalid—a holding that as a practical matter would force recognition of same-sex marriages nationwide. Whether this is a correct interpretation of Windsor remains to be seen.
But Windsor further held that restrictions upon same-sex marriage violate due process and equal protection if they lack a rational basis. While Windsor suggested at other points that the
recognition of same-sex marriage was a matter for state law, Illinois has already decided to recognize same-sex marriage. The court found no rational basis for the delayed effective date:
There is no legislative history that the parties have pointed to, or that the Court could find, that provides either a legitimate governmental justification or a rational basis for the General Assembly's decision to delay the effective date of Senate Bill 10. Nowhere is there any mention or suggestion that a delay in the effective date is necessary to, for example, change forms in county clerks' offices to allow for an orderly transition to the new law. Indeed, the only reason the parties have cited for the delay is the functioning of the state's logistical process of passing a law.
Ultimately, the General Assembly's recent enactment of the new law permitting same-sex marriages and the attendant policy goal of that new law undermines any reason for applying the justification underlying the current law to these plaintiffs in these compelling circumstances. In any event, at least at this stage, Plaintiffs have demonstrated the requisite "some likelihood of success" on the merits of their as-applied equal protection claim, a conclusion that supports granting preliminary relief.
Gray, 2013 WL 6355918, at *5. The court also noted that the Illinois Attorney General had refused to defend the Clerk's refusal to issue the certificate, which was another indication that there was no rational policy basis for the Clerk's position. The court therefore granted the TRO requested by the plaintiffs, allowing them to marry before Gray dies.
The rational basis discussion is the strongest part of the Gray opinion. Windsor is very unclear on whether state restrictions (as opposed to federal restrictions) on same-sex marriage are unconstitutional. The broad holding that legislation is unconstitutional if based upon antigay
prejudice could be applied to state law, but Windsor also held at considerable length that states are free to recognize or not recognize same-sex marriage. Bluntly put, antigay prejudice is probably a material part of the publicly stated rationale for any provision disallowing same-sex marriage; this issue tends to provoke strongly worded public statements from social conservatives who vigorously oppose same-sex marriage. If these statements make state restrictions upon same-sex marriage unconstitutional, then states will functionally not be free to refuse to recognize same-sex marriage.
But other portions of the Windsor opinion stress that the validity of same-sex marriage is an issue for state law. "[T]he Federal Government, through our history, has deferred to state-law policy decisions with respect to domestic relations." Windsor, 133 S. Ct. at 2691. "[T]he federal courts, as a general rule, do not adjudicate issues of marital status even when there might otherwise be a basis for federal jurisdiction." Id. If marital status is a question exclusively for
state law, states must be free to choose to either recognize or not recognize same-sex marriage. Otherwise, the validity of same-sex marriage would depend upon federal law, not upon state law.
In other words, many proponents of same-sex marriage believed that the Court should resolve Windsor by expanding the principles of Loving v. Virginia, 388 U.S. 1 (1967), and recognizing a fundamental federal constitutional right to same-sex marriage. Windsor did not do that, and instead held that the validity of same-sex marriage depends upon state law. Necessarily, that holding implies that state law is free to either recognize or not recognize same-sex marriage. If that is not true, and the recognition is required by federal law, then the validity of same-sex marriages will be controlled by federal law, not state law.
That, in a nutshell, is the fundamental internal conflict within the Windsor opinion. The opinion states that marital status is an issue exclusively for state law. When the holding is logically extended, however, federal law will functionally require all states to recognize same-sex marriage, because essentially all provisions of state law refusing to recognize same-sex marriage were accompanied by antigay statements. Eventually, the Court will have to either give up the concept that state law is controlling and hold that federal law requires recognition of same-sex marriages, or hold that state law restrictions upon recognition of same-sex 0arriage are constitutional despite the presence of some level of antigay prejudice among at least some supporters. There are no provisions of state law preventing recognition of same-sex marriage that were enacted without the making of public statements suggesting antigay prejudice by someone—often social conservatives with strong feelings on the issue.
There may, of course, be room for looking at the role of the persons making the statements. Perhaps state law restrictions on same-sex marriage are unconstitutional when major legislators make prejudicial statements, but not unconstitutional when such statements are made by outside interest groups. But cases like Gray seem to be suggesting that state law restrictions preventing recognition of same-sex marriages are unconstitutional if anyone, anywhere, made statements prejudicial to gay people at the time the restriction was enacted. This rule will functionally create a federal right to same-sex marriage, because the enactment of provisions restricting recognition of same-sex marriage are almost always accompanied by statements by someone showing a high level of prejudice against gay people.
The internal conflict within Windsor was not relevant to the facts of Gray, however. Illinois has already crossed the Rubicon on same-sex marriage, by enacting a bill to allow it. The narrow question was whether there existed a sufficient rational basis to justify the delayed effective date, in a situation where the delay would probably prevent the plaintiffs from ever being able to marry. Given the lack of any substantial policy reasons for the delay, the court's holding that no rational basis existed is not surprising.
A similar result was reached in Lee v. Orr, No. 13-CV-8719, 2013 WL 6490577 (N.D. Ill. Dec. 10, 2013). Lee involved two same-sex couples, each of which included one partner in critical medical condition who might not survive until June 1, 2014. The plaintiffs sought a TRO
not only for themselves, but also for a putative class of "medically critical plaintiffs." The class had not yet been formally certified, but the court found that injunctive relief could still be granted:
The court may conditionally certify the class or otherwise order a broad preliminary injunction, without a formal class ruling, under its general equity powers. The lack of formal class certification does not create an obstacle to classwide preliminary injunctive relief when activities of the defendant are directed generally against a class of persons.
Id. at *2 (quoting Ill. League of Advocates for Dev. Disabled v. Ill. Dep't of Human Servs., No. 13 C 1300, 2013 WL 3287145, at *3 (N.D. Ill. June 28, 2013) (quoting 3 William Rubenstein et al., Newberg on Class Actions § 9:45 (4th ed. 2002))).
On the merits, Lee agreed that enforcement of Illinois's prior law against recognition of same-sex marriage would be unconstitutional:
The Illinois law, particularly the 1996 amendment to the Marriage Act declaring that same-sex marriages are against public policy, is clearly intended to single out gays and lesbians for disparate treatment based solely on their sexual orientation. Moreover, the new amendment contained in Senate Bill 10 legalizing same-sex marriage is an express repudiation of the state's earlier position and undermines the traditionally invoked justifications for the prohibition on same-sex marriage. As set forth in Windsor, denying same-sex couples equal status under the law serves no constitutionally legitimate purpose.
Id. Like the reasoning in Gray, the literal reasoning in Lee suggests that Illinois's refusal to recognize same-sex marriage was unconstitutional even before it was repealed. If prior law was
"intended to single out gays and lesbians for disparate treatment based solely on their sexual orientation," id., it was certainly so intended from the moment it was enacted. But the Windsor opinion anticipated at various points that states would be free to recognize or not recognize same-sex marriages. The validity of state law restrictions upon same-sex marriages, essentially all of which were accompanied by statements by conservative social groups suggesting antigay prejudice, is one of the crucial uncertain points in the Windsor opinion.
In the specific context of Gray and Lee, however, where the crucial decision to recognize same-sex marriage has been made, and the question is only one of effective date, the courts seem to be holding that a delayed effective date is unconstitutional to the extent that it prevents marriages of same-sex couples when one member of the couple may not live until the effective date arrives.
In United States v. Windsor, 133 S. Ct. 2675 (2013), the Supreme Court held that federal law must defer to state law on the validity of same-sex marriages. After Windsor, when a valid marriage exists for purposes of state law, it exists for purposes of federal law.
Is federal law likewise required to recognize a civil union as a valid marriage? A split in authority seems to be developing. One view of the issue is taken by the leading IRS Revenue Ruling on same-sex marriage:
For Federal tax purposes, the term "marriage" does not include registered domestic partnerships, civil unions, or other similar formal relationships recognized under state law that are not denominated as a marriage under that state's law, and the terms "spouse," "husband and wife," "husband," and "wife" do not include individuals who have entered into such a formal relationship. This conclusion applies regardless of whether individuals who have entered into such relationships are of the opposite sex or the same sex.
Rev. Rul. 2013-17, § 4, 2013-38 I.R.B. 201. Thus, for tax purposes, the IRS will not recognize any state civil union as a valid marriage.
But a recent Ninth Circuit decision takes a conflicting view. In In re Fonberg, No. 13-002, 2013 WL 6153265 (9th Cir. Nov. 25, 2013), a female law clerk in the District of Oregon filed a petition to review an employment dispute. The plaintiff was a party to a same-sex civil union that was valid under Oregon law. She had attempted to enroll herself and her partner in a family health insurance plan offered to married employees of federal courts. The Office of Personnel Management ("OPM") denied her claim, reasoning that an Oregon civil union is not a marriage. OPM relied on a formal opinion letter somewhat similar to Revenue Ruling 2013-17:
The Supreme Court's decision addressed the constitutionality of a statute that defined "marriage" and "spouse" for purposes of federal law to include only opposite‑sex couples. Therefore, same‑sex couples who are in a civil union or other forms of domestic partnership other than marriage will remain ineligible for most Federal benefits programs.
OPM Ben. Admin. Ltr. No. 13-203 (July 17, 2013).
After mediation failed, the plaintiff filed a formal employment complaint alleging discrimination. The district court held that the federal government was required to recognize the plaintiff's civil union as a marriage but held that there was no authority for granting reimbursement for insurance payments made before the date of the order.
On appeal, the Ninth Circuit agreed that the federal government was required to recognize
the Oregon civil union as a marriage. Oregon's civil union statute provides:
Any privilege, immunity, right or benefit granted by statute, administrative or court rule, policy, common law or any other law to an individual because the individual is or was married, or because the individual is or was an in-law in a specified way to another individual, is granted on equivalent terms, substantive and procedural, to an individual
because the individual is or was in a domestic partnership [civil union] or because the individual is or was, based on a domestic partnership, related in a specified way to another individual.
Or. Rev. Stat. § 106.340(1) (Westlaw current with 2013 Reg. & Sp. Sess. legis. eff. through 10/8/13) (emphasis added). Thus, Oregon state law expressly provides that partners to civil unions are entitled to all of the privileges and benefits of marriage. A civil union under Oregon law essentially is a marriage in all aspects except the name.
The Ninth Circuit held that the federal government is required to defer to Oregon law and
recognize an Oregon civil union as a marriage:
Fonberg and her partner are treated differently in two ways. First, they are treated differently from opposite-sex partners who are allowed to marry and thereby gain spousal benefits under federal law. This is plainly discrimination based on sexual orientation, which
the District of Oregon's EDR Plan prohibits. They are also treated unequally vis-à-vis same-sex couples in other states in the circuit, who may marry and thus gain benefits under Windsor. This violates the principle that federal employees must not be treated unequally in the entitlements and benefits of federal employment based on the vagaries of state law. Here, Oregon law suffers from precisely the same deficiency that the Supreme Court identified in Windsor with respect to the Defense of Marriage Act. Both these forms of discrimination are prohibited under the Oregon EDR Plan.
Fonberg, 2013 WL 6153265, at *2. The Ninth Circuit also held that the plaintiff was entitled to reimbursement, reversing the district court's decision on that issue.
Fonberg does not directly hold that the position taken by OPM violates the federal Constitution. The narrow holding of the case is that the position taken by OPM violated a specific plan for federal employees (the Oregon EDR plan) that prohibited discrimination based upon sexual orientation. But given the court's express statement that "Oregon law suffers from precisely the same deficiency that the Supreme Court identified in Windsor," and given that Windsor construed the federal Constitution, there is a high degree of likelihood that the court would reach the same result in a case that directly presented the constitutional issue.
At a minimum, Fonberg suggests that the IRS's position on civil unions is likely to be contested by taxpayers who are partners to civil unions. The IRS seems to be reasoning that when a state describes a relationship as a civil union, it is expressly refusing to describe the relationship as a marriage and that, therefore, the federal government is not required to recognize the relationship as a marriage.
But many civil union statutes, including the Oregon statute at issue in Fonberg, are intended to give the partners literally all of the benefits of matrimony except for the label "marriage." Fonberg holds that when federal law looks to state law on same-sex relationships, it must look to the substance of that law—the rights actually granted to partners in civil unions—and not to the mere label.
It is possible, of course, that other courts will reach a different result and uphold the IRS's position. It is also possible that even the Ninth Circuit might distinguish Fonberg and reach a different result under the law of a state that does not define civil unions so broadly and does not award to partners in civil unions literally all of the benefits of marriage. But Fonberg certainly suggests that in some situations, Windsor may require the federal government to treat a civil union as a marriage.