NYU Tax Policy Colloquium, week 7: Richard Reinhold's Does the Parsonage Exemption Violate the Establishment Clause?

Yesterday at the Tax Policy Colloquium, Richard Reinhold presented Does the Parsonage Exemption in Internal Revenue Code Section 107 Violate the Establishment Clause of the 1st Amendment?

Here is a bit of background. Section 107 of the Internal Revenue Code provides: “In a case of a minister of the gospel, gross income does not include (1) the rental value of a home furnished to him as part of his compensation, or (2) the rental allowance paid to him as part of his compensation, to the extent used by him to rent a home or provide a home and to the extent such allowance does not exceed the fair rental value of the home, including furnishings and appurtenances such as a garage, plus the cost of utilities.”

This is potentially suspect under the Establishment Clause of the First Amendment, which provides that “Congress shall make no law respecting an establishment of religion.” Indeed, it would unambiguously violate the Establishment Clause, as that clause has come to be understood, if the rental exclusion were limited to Christian “ministers of the gospel,” as distinct from, say, rabbis and imams. But it has been interpreted as applying to the equivalent of “ministers of the gospel” for religions generally – a concept that may involve considerable ambiguity! (E.g., does “religion” have to involve belief in a God or gods, should the IRS be on the lookout suspected sham religions, how broad is “minister” and how does it track across religious affiliations.)

In Gaylor v. Mnuchin, 278 F. Supp.3^rd 1081 (D. Wi. 2017), the section 107(2) exclusion for cash rental allowances was held to violate the Establishment Clause. The case is on appeal to the Seventh Circuit, oral argument took place last October, and a decision is expected imminently. The main legal issue in dispute, other than section 107(2)’s constitutionality, is whether the plaintiff has standing. In an earlier case, the same judge had found standing but the Seventh Circuit had reversed. Here, the plaintiffs sought to create standing (without which, there is no justiciable case for the court to hear) by filing for a tax refund under section 107(2) despite being employed by the Freedom From Religion Foundation, Inc., rather than by a religious institution in the conventional sense. Thus, the options available to the Seventh Circuit include reversing and dismissing the case on standing grounds. If that happens and is upheld more generally (e.g., by the Supreme Court upon appeal), then section 107(2) (in common, it appears, with section 107(1)) might prove, as a practical matter, to be wholly immune to constitutional challenge through litigation even if it does in fact violate the Establishment Clause.

Reinhold’s article argues that section 107 IS constitutional and ought to be upheld by the  courts even if there is standing (as to which he is skeptical).  The paper raises 3 topics that I’ll (in varying detail) address here: constitutionality, standing, and how we should think about the role of the courts and of constitutional litigation.

1. Is section 107 constitutional? – Suppose Gaylor v. Mnuchin ends up in the Supreme Court, and is definitively addressed on the merits (i.e., the court finds requisite standing). Then we might end up with a decisive precedent, to the effect that section 107 (or at least section 107(2), the cash allowances aspect) either is or isn’t constitutional.

If one were handicapping the outcome of this still-hypothetical case, it would of course be of enormous predictive moment that the Scalia seat ended up going to Neal Gorsuch, rather than to Merrick Garland. I’d personally place a large bet (if this were the type of thing I did) on the Court’s 5 conservatives finding section 107 constitutional, if they reached the issue. But it also wouldn’t surprise me if, upon reading such an opinion, I personally found it unpersuasive and ascribed it to political and policy preferences (as distinct from the legal grounds offered for the Court’s conclusions) that differ from mine.

If that happened, I might find myself concluding that the Supreme Court had (in my view) gotten it “wrong,” and that section 107 “is” unconstitutional. But what does that mean exactly, once we’ve untethered the notion of constitutionality from blind adherence to precedents? The more general problem I am adverting to here is the difficulty of setting forth definite criteria for determining constitutionality. The inquiry is not well-specified, nor can a dispute between A and B on any such question be resolved, other than if they sufficiently accept common premises and one of them has made errors in reasoning from those premises.

Con law, in my view, tends to be a lot murkier and less well-specified in this sense than, say, disputes over tax policy between people who accept generally welfarist frameworks and are open to empirical evidence that those frameworks make normatively relevant.

But with all that said, I am strongly inclined to view section 107 as unconstitutional under the Establishment Clause. I view it as favoring religion, and funding it at the expense of the non-religious, without sufficient justification. (For much more detail than I will offer here, see this article by Adam Chodorow )

One can think of the Establishment Clause as limiting permissible subsidies to religious institutions and actors, and the Free Exercise Clause as limiting the permissible imposition of special burdens on them. This suggests looking for neutrality in some sense, notwithstanding baseline issues in defining it.

Consider sections 170 and 501(c)(3) of the Internal Revenue Code, which provide tax benefits to charities, including religious organizations. I would question their constitutionality, on Establishment Clause grounds, if they were ONLY for religious organizations. I would also question their constitutionality, on Free Exercise grounds, if they offered tax benefits to all charities EXCEPT religious organizations. I don’t think we need to ask ourselves any sort of neutrality question (requiring that one specify the baseline) with respect to having versus not having special treatment for charities in general.

Now let’s look at section 107 from this standpoint. It is JUST for “ministers of the gospel,” as generalized to be nonsectarian within religions. This strikes me as exactly the sort of thing that the Establishment Clause is supposed to inhibit.

As Reinhold’s article notes, one counter-argument is that we actually have a broader category of exempt housing, akin to the broader section 501(c)(3) that I noted above with respect to exempting churches from federal income tax and allowing charitable contribution deductions to those who donate to them. But it’s really just dribs and drabs, apart from the big entry here, which is section 119, excluding the value of housing provided by an employer to an employee, if (a) it is provided for the convenience of the employer, (b) the employee is required to accept it as a condition of employment, and (c) it is on the employer’s business premises (although this can be satisfied where the employee’s off-site house is sufficiently used in the business, e.g., as a site for business meetings).

Although I question section 119 – it over-responds to valuation problems that arise when the home serves business as well as consumption purposes, hence I believe that a value greater than zero ought to be includable – I’m perfectly fine with “ministers of the gospel” (in the broader sense) seeking to qualify under it. Indeed, there would be Free Exercise problems with not allowing them to do so. But section 107 is far broader, especially (although not solely) given section 107(2)’s exclusion for cash allowances. The difference is great enough that section 107 has been scored (I believe, in official tax expenditure budgets) as costing nearly $1 billion a year.

Boris Bittker, a rightly towering figure but one who liked to argue absurd things sometimes, apparently claimed that section 107 should be viewed as merely an evidentiary rule establishing a presumption – albeit, an irrebuttable one even when demonstrably false – that one qualifies under section 119. Or perhaps the argument is better put as: section 107(1) might be reasonable as an evidentiary presumption re. section 119, then once we’ve gone that far why not vastly broaden it (on the claim that this increases neutral effects between religions) by adding section 107(2).

This line of argument strikes me as rather a triumph of bootstrapping. But the rationale for the initial evidentiary presumption is that testing “ministers of the gospel” for qualification under section 119, potentially subject to audit, would raise undue “entanglement” concerns under the Establishment Clause. But to my mind, these concerns are small not only absolutely, but also relative to the entanglement concerns raised by needing to define “ministers” who can claim the benefit of section 107.

So I regard the Establishment Clause case against section 107 as extremely powerful, although (as Reinhold’s article helps to show) not everyone agrees about this.

2. Standing – The standing issues in Gaylor are clearly significant, and I won’t address them here. But I agree that standing is not just some sort of arcane technical requirement (although it is that, too). Rather, it serves substantial purposes in limiting courts to resolving the sorts of disputes that they are best equipped to handle. It would presumably be a big mistake to offer “taxpayer standing” whenever a taxpayer happened to dislike a particular appropriation or tax preference, and thus wanted to raise all conceivable challenges.

Still, if section 107 is unconstitutional, yet no one has standing to challenge it (or at least section 107(1), even if the taxpayer in Gaylor succeeds in establishing standing to challenge section 107(2)), then we have a potential problem of under-enforced constitutional norms. This is especially troubling if one believes that a key reason for having constitutional limitations, such as those in the First Amendment, is to allow the courts to address violative behavior by the executive and legislative branches, in circumstances where our supposedly (whether or not actually) majoritarian politics may fall short of offering adequate protection.

3. Role of the courts and of constitutional litigation – While the paper mainly makes specific legal arguments, it’s also about what one might call sound social practice: Is constitutional litigation really a good way to handle the sort of issue that section 107 raises? Does it overly politicize the courts, and/or overly judicialize politics? Rather than providing a way to reach consensus in a diverse and pluralistic society, does it instead end up intensifying social discord?

These are serious questions, and hard to answer confidently. With respect to institutional choice, I am not confident that Congress and the executive branch can be trusted to adhere sufficiently to Establishment Clause norms, without judicial oversight. But then again, my confidence in the Supreme Court as a good faith arbiter has been higher at times in the past than it is now.

In terms of consensus versus discord, I admittedly identify with the plaintiffs in Gaylor. As one whose own descent and set of personal beliefs about what one might call cosmological questions places him outside of the U.S. majority, I would like to be able to invoke the protection of the courts when Establishment-type issues arise. I remember, as a child in public school, being offended by the public assemblies in which I was expected to recite the “under God” language from the Pledge of Allegiance. To this day, I am affronted when ignorant Supreme Court justices falsely (although, I presume, sincerely) assert that a crèche, or even crosses in a public graveyard, are not religious symbols and are not rightly perceived by minorities / outsiders as such.

But even granting that I (and “we”) are rightly affronted, do we help ourselves by bringing these cases?  Are we better off just swallowing the ill feeling than creating larger controversies that may inspire pushback? Maybe yes, maybe no.