When Religious Tax Accommodations Are Inconsistent

On Wednesday, October 24, the Seventh Circuit is going to hear arguments in the appeal of Gaylor v. Mnuchin. I’ve written about this parsonage allowance case a number of times in the past (see here and here for examples), but as a quick summary: section 107(2) of the Code says that “ministers of the gospel” don’t have to include rental allowances in gross income. Several years ago, the Freedom From Religion Foundation challenged this parsonage allowance on the grounds that it violated the Establishment Clause of the Constitution. They won in the district court, but the Seventh Circuit found that the plaintiffs didn’t have standing to challenge the provision.

The Seventh Circuit also suggested, in a footnote, that if they claimed a parsonage allowance and the IRS rejected their claim, they might have standing. So they did, the IRS did, and the district court again found the provision unconstitutional. And now the Seventh Circuit will weigh in (again).

As a side note, this provision (as well as a bunch of others) made their way into God and the IRS, the book I wrote that was recently published about tax accommodations of religious individuals. The fundamental purpose of the book was to illustrate the ad hoc nature of religious accommodations in the tax law, and develop a framework that could provide some consistency as Congress and the IRS consider providing these accommodations.

In anticipation of the Seventh Circuit hearing (and, perhaps, of a presentation I’m going to give at the law school), I was reading through the briefs, and something leapt out at me that vividly illustrates the danger of the current ad hoc version of accommodation. In describing why the parsonage allowance is not unconstitutional, the government makes a couple fundamental arguments. One is that the accommodation avoids excessive entanglement between the government and religion. There’s a general exemption for some employer-provided housing, as long as the housing is provided for the convenience of the employer. By exempting employer-provided housing (and, apparently, housing allowances), the government doesn’t have to investigate whether the provision of housing to ministers is for the convenience of the church.

The second is an argument that the tax-free provision of housing allowances eliminates discrimination among various religious denominations:

Congress enacted § 107(2) to ensure that all ministers who were similarly situated were treated equally by the Government, tax-wise. Id. And, in doing so, Congress simply codified the results of litigation brought by ministers seeking equal tax treatment of parsonages and parsonage allowances. See, above, pp. 7-8. Because § 107(2) has the permissible, secular purpose of avoiding governmental discrimination among religions, it furthers one of the core purposes of the Establishment Clause. See Larson v. Valente, 456 U.S. 228, 246 (1982) (determining that law that applied to some, but not all, religions violated the Establishment Clause by running afoul of the “principle of denominational neutrality”).

Brief for the Federal Appellants, BL-16, pp. 38-39 (emphasis added).

Leaving aside whether this is a compelling argument for the constitutionality of section 107(2),[fn1] the Larson citation suggests that another tax accommodation is insupportable. See, section 1402(g) provides an exception from SECA taxes. In order to be exempt, though, self-employed individuals must (a) be a member of a recognized religion, the established tenets of which don’t allow them to accept social security (or other public or private insurance similar to social security), and (b) they may be required to provide evidence to the Secretary of their membership in the religion and of its teachings.

Essentially, as I discuss in my book, this was an exemption created for the Amish. But when Congress created the accommodation, it wasn’t thinking broadly about how that accommodation would fit with other accommodations, and so it doesn’t fit. If the government is right that the touchstones of permissible accommodation are denominational neutrality and avoiding excessive entanglement, well, the SECA exemption does neither. Congress didn’t want social security to be an optional program, so it deliberately created a very narrow exemption, one that, as far as I know, applies only to the Amish.[fn2]

And the exemption demands entanglement: it requires individuals who want to take advantage of it to prove that they belong to a religion that forbids the use of this kind of insurance and that they comply with the religious prohibition. So the government has to evaluate what the tenets of the religion are, as well as an individual taxpayer’s compliance with those tenets. I can’t imagine a much more intrusive religious inquiry.

That’s not, of course, to say that if the court upholds section 107(2) that the Amish accommodation will go away. It would be extremely difficult, I suspect, to have standing to challenge it. Still, because Congress (and, for that matter, the DOJ) has an atomized view of religious tax accommodation, the policies underlying the various accommodations are unrelated and, at least in some circumstances, inconsistent with each other.

Does this have anything to do with Mormonism? I mean, it’s a question of tax accommodation of religious individuals, which definitely has some connection to Mormonism (in fact, two chapters of my book deal extensively—albeit not exclusively—with litigation that dealt specifically with Mormons). Notably, at least some General Authorities have a portion of their paychecks designated as a housing allowance, meaning they receive that portion of their salary tax-free. So this month’s litigation could eventually impact the after-tax pay of church leaders.

Addendum 10/1: Just after the post went live, I was reading some of the legislative history of section 107(2). The argument that this provision is intended to eliminate discrimination between clergy of different religious goes back to the testimony of Rep. Peter F. Mack Jr., who introduced the legislation. He testified that the provision occurred to him when a Baptist official called it to his attention. “Upon looking into the matter, I realized that the present tax laws are discriminatory among our clergy.” That statement certainly seems to support the idea that section 107(2) was not intended to advance religion, but rather to create denominational neutrality.

But there’s a problem with that argument: at the very end of his testimony, Rep. Mack says, “Certainly in these times, when we are being threatened by a godless and antireligious world movement we should correct this discrimination against certain ministers who are carrying on such a courageous fight against this foe. Certainly this is not too much to do for these people who are caring for our spiritual welfare.”[fn3] And that sounds an awful lot like the provision was intended specifically to favor religion. So oops, maybe?

Cross-posted at Surly Subgroup.

[fn1] Though frankly, I don’t find either argument super-compelling.

[fn2] There’s another exemption in 1402(e) that applies to clergy and Christian Science practitioners generally; that accommodation does seem to have denominational neutrality, but the Amish exemption applies to all Amish members, not just clergy.

[fn3] Both of these quotations come from Statement of Hon. Peter F. Mack, Jr., on H. R. 4275, Concerning the Taxability of a Cash Allowance Paid to Clergymen in Lieu of Furnishing Them a Dwelling, Bernard D. Jr.; William S. Hein Co. Reams, Inc., Internal Revenue Acts of the United States: The Revenue Act of 1954 with Legislative Histories and Congressional Documents at 1574-78.


  1. Last Lemming says:

    I don’t remember if I have mentioned this before, but my Jewish boss would argue that simply by employing the phrase “ministers of the gospel,” section 107(2) fails the denominational neutrality test regardless of how expansively the phrase has been interpreted by the courts. She won’t let me use the term even in the precise context that the IRC uses it.

  2. Yeah, every time I present about this, I have to air-quote “ministers of the gospel,” and often I go on to explain how the IRS has broadened it (though, again, that broadening requires entanglement between government and religion. Like, rabbis and imams almost certainly count, but the IRS has found that “ministers of music” and “ministers of education” don’t. And it’s gone both ways on Jewish cantors. So even the argument that this prevents entanglement is disingenuous at best.

  3. J. Stapley says:

    Thanks Sam! Really interesting. I’ll have a review this excellent little volume up this month for sure!

  4. Aussie Mormon says:

    I’m only about a quarter of the way through your book after getting it when it first came out as I only get to read a bit every so often, (and funnily enough, am currently at a section talking about Amish&Social Security), but so far it’s been an interesting read. Seeing what different people have tried over the years, and in some cases received what seems like harsh treatment, is a fascinating look into the other side of the big pond.

  5. I can’t remember if I’ve said this before (Gaylor v Mnuchin has come up), but at least today I’m thinking about entanglement. It seems to me that if one were to start from general principles (Section 119 and convenience of the employer) without the specific parsonage allowance, and examine Mormon practice, we would discuss mission presidents and temple presidents who have assigned locations, and would not think general authorities as a class (with some special assignment exceptions) would be entitled to consideration. But that would require a fairly detailed examination of a particular church’s practices and requirements.

    I think that’s an argument that there is something to the concern about entanglement. On the other hand, it suggests that the parsonage allowance as it exists permits something extraordinary.

  6. Thanks, J. and Aussie Mormon; I look forward to hearing what you think of the book!

    Chris, I’m not sure I’m convinced that the question convenience of the employer would be that entangling for purposes of section 119. I mean, you’re absolutely right that it raises some questions (for example, it probably wouldn’t apply to mission presidents, as long as the church argues that they’re not employees), but the questions feel less like questions of religious determinations and more like questions of fact. Honestly, though, even if those questions do require some entanglement, I think it pales in comparison with the entanglement required by section 107, which requires the IRS to determine who qualifies as a “minister of the gospel,” which is a determination that requires it to look into the beliefs and practices of the underlying religion. So even if 119 would require entanglement, it’s not like we’re looking at entanglement vs. no entanglement; we’re looking at degree of entanglement.

%d bloggers like this: