In a 5-4 opinion written by Justice Kennedy, the United States Supreme today held in United States v. Windsor that the provisions contained in the Defense of Marriage Act (“DOMA”) that exclude same-sex relationships from the definition of marriage and spouse for federal law purposes is unconstitutional as a deprivation of the liberty of persons that is protected by the Fifth Amendment of the Constitution of the United States. In doing so, Justice Kennedy has highlighted once again his role as a critical swing vote on the Court. He also has rendered a decision that seems likely to have far reaching implications for the design and administration of employee benefit plans in this country.

In Windsor, Edith Windsor, the plaintiff, and Thea Spyer, her partner, were residents of New York who got married in Canada back in 2007. Ms. Spyer died in 2009, and left her entire estate to the plaintiff. The plaintiff thereafter sought to claim the federal estate tax exemption for surviving spouses. Relying on the definition of spouse found in DOMA, the IRS denied the availability of the exemption. After paying the resulting estate tax, the plaintiff sued for a refund and in doing so challenged the constitutionality of DOMA. The plaintiff’s motion for summary judgment was granted by the District Court and the Second Circuit Court of Appeals affirmed that decision. The Supreme Court agreed to hear the case.

At issue in Windsor is Section 3 of DOMA, which states that the word “marriage” means only a legal union between one man and one woman as husband and wife. Section 3 goes on to provide that the word “spouse” refers only to a person of an opposite sex who is a husband or a wife. DOMA, which was signed into law by President Bill Clinton in 1996, also empowered states to refuse to recognize same-sex marriages recognized in any other states (note that this latter provision was not considered in the Windsor case).

Now that Section 3 of DOMA has been struck down, there will be numerous changes ahead under federal laws, including those laws that regulate employee benefit plans (e.g., ERISA and the Internal Revenue Code (the “Code”)). It is far too early at this point to identify all of the changes that will be required as we await reaction (hopefully in the form of helpful interpretive guidance) from the federal government and from the states.

In general, since DOMA has limited the required definition of a “spouse” under ERISA and the Code to an opposite-sex spouse, there has been no requirement to extend protections for spouses under qualified retirement plans to same-sex spouses (although plans generally were free to do that voluntarily). Health care benefits extended by plan sponsors to same-sex spouses resulted in imputed income to the covered employee unless that spouse properly could be treated as a tax dependent under applicable rules. Moreover, several other welfare benefits requirements otherwise required under federal laws did not have to be provided to same-sex spouses. This all has changed.

Some provisions in the Code and ERISA prohibit same-sex spouses from receiving certain benefits otherwise available to spouses while some permit same-sex spouses to receive such benefits but do not require it. For retirement plans, some examples include survivor benefits in the form of qualified joint and survivor annuities and other forms of death benefits, designation of plan beneficiaries, timing rules for the payment of death benefits and access to qualified domestic relations orders. With respect to health care plans, some examples of areas directly affected by the Windsor decision include the ability to extend health care coverage without the incurrence of imputed income regardless of dependent status, required extension of COBRA continuation coverage benefits, coverage under the FMLA, special enrollment rights under HIPAA and the ability to make tax-exempt reimbursement of expenses incurred by same-sex spouses under flexible spending arrangements, health reimbursement arrangements and health savings accounts.

With today’s rejection of Section 3 of DOMA by the Court, the definition of marriage for federal law seems to once again depend on state law. Thus, it appear plan sponsors are consigned to dealing with the current patchwork of state laws (that either permit or prohibit same-sex marriage and/or civil unions). In this case, finality does not necessarily bring along with it ease of design and administration. Plan sponsors are going to have to make some decisions, and some of those may be difficult and at least at first may have to be made in a regulatory vacuum.

Plan sponsors also are well advised to consider the effects of the Windsor decision on how a plan should define a child. The extension of employer-provided health care coverage to the child of a same-sex spouse also could result in imputed income to the employee. Now that Section 3 of DOMA has been rejected, it seems that this result may change so that imputed income no longer results. The favorable tax treatment extended to step-children of employees in a heterosexual marriage seemingly now may be extended to step-children of same-sex spouses – at least in states where same sex marriages are permitted. This may suggest the need for plan design changes.

It seems that plan sponsors also may want to revisit the propriety of domestic partner coverage. To the extent that the rationale for offering such coverage is based on the notion that an employee’s same-sex domestic partner otherwise could not be treated as a spouse, then perhaps this category of coverage no longer is needed (at least in those states that permit same-sex marriage). While eliminating this coverage certainly could have negative human resources implications, it will be interesting to see how plan sponsors respond.

Plan sponsors should review their employee benefit plans in light of the Windsor to begin to identify the impacts the elimination of Section 3 of DOMA will have. Some of this will necessarily depend on how the applicable government entities interpret the Windsor decision (for example, it will be necessary to determine if the case should be applied prospectively only or whether it may have retroactive applications, which could raise multiple difficulties) and on how plan sponsors wish to contend with varying state laws on same-sex marriage that affect their plans. Plan sponsors operating in multiple states obviously need to consider a variety of additional issues (including the adoption of a uniform approach regardless of state law differences). While it certainly is not too early to begin this analysis, it unfortunately may take some time to identify and answer all of the questions.