Taxing Churches: A Response

Oh no—somebody on the Internet is wrong while I’m on vacation! But duty calls.

Recently, Ryan Cragun, a sociology professor, along with students Stephanie Yeager and Desmond Vega, argued that the government subsidizes religion by about $71 billion a year. He thinks this is wrong, and that religions should pay their fair share.

I have no problem with his making this argument—tax exemption costs the government significant revenue (though his $71 billion is based on really, really poor assumptions—more on that later), and should be examined carefully and critically.

But Prof. Cragun’s analysis is not the careful and critical examination that the tax treatment of churches deserves. His piece has a number of significant problems. I’m not going to address all of the problems, including the fact that he appears unaware that there is an extensive academic literature that explores the place of a tax exemption for churches,[fn1] but I am going to address a handful of his assertions. In the end, though, what bothers me most about Cragun’s piece is that he’s taken an important topic and made it into a polemic. Those who agree with him now have “facts” to bandy about, while those opposed have a specious argument they can treat as an easily-dismissed straw man, and can ignore engaging in a valuable tax policy discussion.

Before I get into my specific criticisms, though, I want to make a couple points upfront. First, although I find lots of legal and methodological problems with Cragun’s analysis, I wouldn’t consider it a salvo in some sort of war on religion. Arguing that churches should not be exempt from taxation is not an inherently anti-religious question. Especially in a world of fiscal shortfalls and increased need, it’s important to critically evaluate expenditures, including tax expenditures.

My impression is that Cragun saw what looked like an interesting topic, one related to a subject he was interested in, and ran with it. It turns out, though, that there’s a steep learning curve associated with his chosen topic. Frankly, I hope that, if I were to explore the sociology of nontheists, that I could come as close to asking the right questions as he comes.

Second, I’m not attempting to make an affirmative case for tax exemption generally, or for exempting religious organizations in particular. My sole project here is to react to some of Cragun’s assertions. With that said, on to specifics.

1. Tax-Exempt Organizations. Cragun begins by arguing against churches as “charities.” If they’re not charitable organizations, he says, then taxing them is fair game. He proposes that, in order to be charitable, an organization should be required to spend at least 50% of its revenue on “charitable work addressing physical needs.”[fn2]

He doesn’t provide any support—other than a reference to the Red Cross—for that definition, but I’m not actually interested in arguing about what constitutes a charity. Because tax-exempt and charitable are not coterminous.

Did you know that the NCAA is tax-exempt? The NFL? Labor unions? Some fraternal organizations? Universities, including the one at which I teach and the one Cragun teaches at? Some of them probably shouldn’t be exempt from taxes (I’m looking at you, NFL and NCAA!), but I’m almost certain that none would qualify as a charity under Cragun’s definition. Which makes singling out religious tax-exempt organizations disingenuous at best.

2. Donations. Per Giving USA, religions received nearly $101 billion in donations in 2009. Cragun estimates the churches’ exemption from tax cost the federal government $35.3 billion in revenue on this income, and cost state governments an additional $6.18 billion. (That $41.5 billion represents more than half his estimate of the tax cost of churches’ tax exemption.)

Except that it doesn’t cost the government $41.5 billion. Under the Internal Revenue Code,[fn3] gifts do not constitute income to the recipient.[fn4] And, according to the IRS and DC Circuit dicta, donations to churches count as gifts.[fn5]

This lost revenue, then, doesn’t qualify as lost revenue. Under generally applicable tax law, churches would not pay taxes on donations[fn6] even if they weren’t exempt from tax, and even if they weren’t churches.[fn7]

3. The Parsonage Allowance. If you’re lucky, maybe your employer provides you with housing at a below-market price. If you do get employer-provided housing at a below-market price, though, the difference between what you pay and what you would pay without the employer subsidy is considered compensation to you, and you owe taxes on that difference.

Unless you’re a “minister of the gospel.”[fn8] If you are, you can exclude the value of employer-provided housing, whether it is provided I the form of actual housing or in cash as a housing allowance.[fn9]

Cragun takes issue with the parsonage allowance. As, frankly, do I. It’s purely a political animal; not only is there no tax policy reason for it, but it flies in the face of important tax policy considerations, including the fundamental equity consideration of horizontal equity, which holds that similarly-situated taxpayers should pay similar amounts of tax.

Still, notwithstanding my fundamental agreement in this point, I’m deeply skeptical of his quantification of the cost. With 600,000 ministers, Cragun assumes an average marginal tax rate of 25% and an $8,000 annual allowance. He takes his average salary from, but doesn’t explain how he arrives at his annual allowance, or why he assumes that each of the 600,000 ministers receives a parsonage allowance. I don’t have any idea what the true number is, but I suspect that $1.2 billion is not a good estimate.[fn10]

4. Property Tax. I’m a federal income tax guy, not a state and local guy, so I don’t have much to say about this part of Cragun’s analysis, except that his numbers are unconvincing. Though my stats knowledge is rusty, I’m pretty sure that a non-random selection of 47 church buildings (for six Protestanty religions) in the Tampa area is not a statistically valid sample for estimating an average value for all religious real property nationwide.

Moreover, Cragun doesn’t explain how he determined the property tax rate he used, but determining an appropriate rate is tough. It turns out that propert tax is not a state-level, but a local-level tax, meaning there are probably hundreds, if not thousands, of property tax jurisdictions in the US. You can estimate some sort of weighted average rate to use, of course, but it’s complicated, and it would be nice if Cragun had told us how he arrived at whatever rate he used in his calculation.

5. Deductions. Cragun asserts that the deductibility of donations to churches increases donations. That may be, but it’s not an uncontroversial assertion. It depends on whether donors are determining the amount of their donations pre-tax or not. That is, if I want to donate $10 to a church, and am in the 35% tax bracket (and I itemize—more on that in a minute), the church gets $10, but that $10 only costs me $6.50. In that case, the entire subsidy went to me. On the other hand, if I want my donation to cost me $10 after taxes, I can donate $15.38, and the church gets the benefit of the deduction. It’s not clear to me, in the real world, that most people think in after-tax dollars.[fn11]

Moreover, only about a third of taxpayers (give or take) itemize. Taxpayers who don’t itemize don’t get any tax benefit from their charitable donations (or deductions for mortgage interest, or other itemized deductions). And donors to churches are more likely than other donors to be non-itemizers.[fn12]

Conclusion. Cragun and his coauthors ask good questions about the tax status of churches. Sadly, the resulting article is marred by mistakes of fact and law and lack of familiarity with the work that already exists about this topic. Whether and how religion (and, more broadly, tax-exempt organizations) should be subsidized by the government is an important topic, and one that deserves airing beyond the rarified world of tax policy people. Still, as I hope I’ve demonstrated, there is a certain level of homework that needs to be done before stepping into that discussion.

[fn1] Why not? Basically because I’m on vacation, and I’m writing this on my phone. And it’s kind of a pain to do Lexis research on a phone.

[fn2] For a lot of reasons, I assume he means 50% of expenditures, rather than revenue, maybe with some minimum distribution requirement, but maybe he doesn’t.

[fn3] Not, as Cragun writes, the “Internal Revenue Service tax code.” I apologize for the pickiness of this footnote, but that’s a mistake my first tax professor mentioned being annoyed by, and I share his annoyance. The tax law is written by Congress; the IRS administers it.

[fn4] Section 102.

[fn5] Branch Ministries v. Rossotti.

[fn6] I’ll qualify this statement a little: some donations probably wouldn’t qualify as gifts, and would be taxable to a non-exempt church, but the vast majority would qualify as gifts.

[fn7] There’s an odd discursion here about the Mafia being able to use churches to launder money. Frankly, if any Mafioso is doing that, he or she really needs to hire better tax counsel. As I understand it, the purpose of money-laundering is to take dirty money, invest it in something, and get it back as clean money. Based on the constraints imposed by the tax law, a tax-exempt organization such as a church will lose its exemption if it operates for the private benefit of anybody particular. Treas. Reg. § 1.501(c)(3)-1(d)(1)(ii). So laundering money through a tax-exempt seems like a really bad idea, at least if (a) you want your money back and (b) you want to maintain the exemption.

[fn8] Note that, in the wonderful world of tax law, a minister of the gospel can and does include non-Christian clergy.

[fn9] Section 107.

[fn10] Though I can’t answer the question of how much the parsonage allowance costs the government, I can at least provide Cragun and his coauthors some enlightenment: they started down this trail when they discovered they couldn’t find how much Rev. White paid in taxes. That’s clearly true: the Code prohibits the disclosure of tax return information by any government official. Section 6103. Unauthorized disclosure is a felony. Section 7213. But wait, you may be asking, how do we know how little GE paid in US taxes then? We don’t, actually. Journalists have tried to reverse-engineer the amount based on SEC filings.

[fn11] Moreover, in a number of religions—mine included—believers are expected to give a set amount. The bulk of my contributions to the Mormon Church consist of tithing, a donation of 10% of my income. Though there is some wiggle room about the income base on which I pay, the 10% number isn’t tax-dependant.

[fn12] You can look at this post of mine, and especially fn4, to run that down.

27 comments for “Taxing Churches: A Response

  1. Excellent analysis, Sam. Minor quibble – of course, determining that donations to churches are otherwise non-taxable as gifts is not the end of the analysis, as the current gift tax regime imposes its own tax and compliance costs.

  2. DCL, absolutely true. I decided to ignore gift tax consequences because I’m assuming most donors’ donations would come in under the annual exclusion in any event, and because the donor, not the church, would bear the tax. But you’re right—there would be tax consequences if churches lost their exemption. It would just be nowhere near the scale that Cragun estimates.

  3. I don’t follow the “give $10/costs $6.50” (or in my case, $8.50, I guess) line. Is this thinking that proceeds from other-than-cash-accounting standards?

    In my view, the $10 cost me $10. If another mostly independent entity wants to charge me less than my 15% because I’m a good chap, I don’t think of that as them having taken the money first and then rebating it back to me, thus reducing the cost of my *donation*. Rather, I think of that as a reduction in the cost of my *social contribution* and don’t try to conflate the two. After all, for wage earners like me, at years’ end all the money is basically gone anyway.

  4. Rob, if you itemize, your charitable contributions reduce your tax bill by your marginal tax rate. If, then, you make a $10 donation and pay taxes at a 15% rate, your donation reduced your tax bill by $1.50. Thus, your $10 donation cost you $8.50 after taxes.

  5. Excellent post. Thanks.

    When my students ask me what I did with my summer vacation, I’m going to tell them I read blog posts about tax law.

  6. You wrote this on your *phone*??!!??

    What kind of phone do you have? I can’t imagine doing anything this long and complicated (footnotes! bold headings!) on a phone. You must have a lot of patience.

    P.S. – very interesting post

  7. Thanks, Sam. It’s so nice to have a coherent analysis. I especially like that you call him on throwing around numbers without telling how he arrived at those amounts.

    Are PACs tax-exempt? Are contributions to them tax-deductible and/or considered gifts?

  8. not sure how any of the details matter one way or the other. US tax policy is entirely political and the churches are a powerful lobby with broad public support. no surprise then that they get generous tax treatment.

  9. Interesting and insightful post! I am curious about one aspect of tax exempt churches that I don’t see addressed – maybe Crager didn’t bring it up? That is, the for-profit arms of churches wind up being effectively tax exempt entities as well as the arms we think of as churches. As I understand it, a for-profit entity (say Bonneville International or City Creek Mall) can take its profits and “donate” them to the tax-exempt parent organization. The for-profit entity can always receive investments from the parent corporation as needed – but those are used as business expenses and also tax exempt. Are there any estimates for how much this type of tax sheltering costs in tax revenue?

  10. Cragun asserts that the deductibility of donations to churches increases donations. That may be, but it’s not an uncontroversial assertion.

    Among economists, its pretty uncontroversial. A 1994 paper in the Public Finance Review estimates the price elasticity on contributions to churches to be -1.0. (For other charitable organizations, it was -2.) To put that into meaningful terms, such an elasticity implies that the change in tax price associated with increasing the top marginal rate from 35% to 39.6% (scheduled to occur in January) would increase contributions to churches by roughly 7 percent. One of the authors sits next door to me, and although he won’t vouch for that number’s validity in 2012, he says that the price elasticity is definitely not zero, and he knows of no published article asserting that it is.

  11. “He doesn’t provide any support…”

    “This lost revenue, then, doesn’t qualify as lost revenue.”

    “I’m deeply skeptical of his quantification of the cost.”

    “…his numbers are unconvincing… Moreover, Cragun doesn’t explain…”

    “…it’s not an uncontroversial assertion.”

    “Sadly, the resulting article is marred by mistakes of fact…”

    What mistakes? You haven’t told us what those mistakes are, only that you aren’t sure if they’re mistakes or not.

  12. Thanks, everyone, for your contributions!

    Edje, I’m always happy to give someone summer vacation reading.

    Charlene, I think PACs are tax-exempt, but I know you don’t get a deduction for contributions to PACs.

    NorthboundZax, corporations can only deduct up to 10% of their taxable income in charitable contributions. So even if Bonneville donated all of its income to the Church (which it may or may not be able to do—if the Church is a shareholder, Bonneville would be paying nondeductible dividends), it would owe taxes on 90% of its income.

    ed, it’s an iPhone. I started on it out of necessity, and eventually it became a matter of pride (though still necessity).

  13. LL, good point. The elasticity is not 0, but how elastic charitable giving is remains, to the best of my knowledge, unresolved. This reflects roughly my understanding of the state of the art:

    I’m interested in your neighbor’s study—from what I’ve read, charitable contributions are most elastic among the rich, who tend to donate more heavily to non-church charities.

  14. The elasticity is, indeed, unresolved (and most likely will remain so). But even the paper you link to points to an estimate of -0.4, and a nonzero elasticity makes it a legitimate avenue for Cragun to pursue. (I have no idea what elasticity he assumed.)

    And your statement that “It’s not clear to me, in the real world, that most people think in after-tax dollars.” misses the point. Rich people frequently do and they make a disproportionate share of the contributions–even to churches. It is true, however, that they donate more heavily to nonchurch charities–hence, my colleague’s estimate of a (huge) -2 elasticity for such contributions.

  15. Ah. Sorry LL, I misread your colleague’s general charitable elasticity as -.2, not -2.

  16. Sam, I understand the logic. But the logic doesn’t square with my picture of the world. It appears to say that if I itemize, then $1.50 of my donation came from the government. That leaves openings for critics of churches to make claims about the government subsidy of churches, which has its own toxic slippery slope in a post-prop-8 political climate.

    At any rate, I don’t donate to reduce my tax bill, which it seems to me is required motivation to undergird the thinking that an itemized tax return reduces the cost of donations in the first place.

    But that makes my original question more important to me: Am I missing this because I manage my household on a cash basis, rather than any other kind of accounting?

  17. Sam, how much of the concern over the parsonage allowance might be a reaction to mega-church ministers and televangelists living in palatial housing provided by the church they founded and being able to count those extravagant accommodations as a parsonage allowance – and how might the housing at that extreme affect a reasonable average?

    I know of a church near where we used to live where the founding minister and his wife (no children) live in a church-provided house of multiple thousand square feet that would appraise for multiple millions of dollars in that area. Are you saying they can write it off through the parsonage allowance? I can understand resentment and even anger at that situation.

  18. Armand Mauss just emailed me to let me know about your post, so I thought I’d stop by.

    First, thanks for reading my article and taking the time to critique it. It wasn’t perfect and no doubt could have benefited from having someone with better tax law chops on board.

    Second, if you don’t mind, I’d like to respond to some of your critiques.

    1) I’m aware of the fact that “charity” and “tax-exempt” are not coterminous. I don’t think I said they were. However, one of the more common arguments for religions being tax exempt is that they are charities. That’s why I addressed whether or not they are charities by illustrating that they do not donate a substantial amount of money (or goods or time) to addressing the physical needs of others (at most about 30% of their revenue goes toward that). I also addressed the other activities they engage in and argued that they do not merit tax exempt status based on those.

    2) On point 2, you failed to state one of my assumptions. I’m going to call it an unfair criticism given that you are working under a different assumption. I clearly stated in the article that I was treating religions as though they were for-profit corporations providing entertainment, like movie theaters, and defended that assumption because I had just argued that religions were not “charities”. I also argued that, if not charities, it could be claimed that their primary purpose was to entertain. Under THAT ASSUMPTION, the money they receive would not be donations but rather income (well, technically, revenue).

    3) Throughout the article I note that I am estimating these values. I think $8,000 per year for housing is, well, remarkably cheap. I tried to be pretty conservative in my estimates. Is there a reason you think it should be lower? I know my mortgage payment is more than that, and I live in a home that is below the average value in my area.

    (Also, I do want to note here that you said, “I don’t have any idea what the true number is…” That’s precisely why I wrote this article. No one knows. If I’m wrong, I want someone to show me definitively where and how I’m wrong. And that means people need to turn their attention to this issue. Please, please, prove me wrong with solid numbers! That way, we’re both more informed.)

    4) Again, you’re kind of poking holes in my argument with a needle. Sure, I could have taken a bigger sample of churches. But property values in Tampa are about the national average (which is why I used Tampa). I didn’t try to cherry pick expensive churches; we found all the churches we could for the denominations listed in the article. I’m also surprised that you didn’t mention the value – $1.7 million. Does that seem unreasonable for a church – land and building?

    5) One of the other commenters already addressed this issue. There is decent evidence from economists suggesting that the ability to write off donations increases the odds of donating to churches. I’ll leave it at that.

    Finally, I’d like to see what you think the subsidy to religions in the US is – though, of course, that requires you to work from my assumption that religions are not charities and should be taxed like for-profit corporations like movie theaters. If you reject the assumption, that’s fine, but it is a key assumption of the article.

  19. Churches provide “entertainment.” That is, as you probably won’t hear from any Mormon pulpit this Sunday, one hell of an assumption.

    And it suggests precisely why churches should be tax exempt–because they represent another sovereignty, separate from the state, and we believe that they should be accorded the respect that a separate sovereignty deserves. And shouldn’t be subject to the state’s power to destroy (which Supreme Court decision said that?) if a majority of our citizens should decide that it’s simply another class of entertainment.

    That goes double for the entire “tax expenditure” and “subsidy” argument. They both appear to be built upon the premise that all money is owned by (or owed to) the state, and any choice not to take some of that money is either an expenditure or a subsidy. Some of us find those assumptions troubling, frankly.

  20. Dr. Cragun, thanks for stopping by and for your reactions to what I wrote. Your comment warrants a response, but I’m in the middle of family obligations for at least the next couple days. But I’ll get back to you as soon as possible.

  21. Okay, Dr. Cragun, a quick response while everybody’s still asleep.

    (1) Maybe in popular discourse, being a “charity” is a reason for churches to be exempt. But that’s fairly meaningless in the tax discussion—it’s brought up, of course, but nobody knows what “charity” means or why it is relevant. At most it matters because it underlay the original tax exemption. But churches are exempt because they’re listed in section 501(c)(3) of the Internal Revenue Code, along with a host of other groups. So creating a narrow definition of “charity” doesn’t mean anything for the tax exemption.

    Moreover, as I said, a host of other entities that wouldn’t meet your “charitable” definition are also tax-exempt (and I’ll get to those in (2)). If you want only charities that spend a high proportion of their revenue (again, I think you want expenditures, but I’m fine either way) on bettering people’s physical welfare, you’re not arguing against a church exemption, you’re arguing for a fundamental shift in tax exemptions generally. Which is fine, but then addressing solely churches is fairly irrelevant.

    (2) I didn’t mention your “entertainment” argument, sure, but let’s go with it. Assume that churches are in the business of providing entertainment, contra Mark B. So what? Other exempt entities under 501(c)(3) include the Lincoln Center, the Shedd Aquarium, Hubbard Street Dance, WBEZ, WNYC, WTTW (the main PBS station in Chicago), the New York City Ballet, the U.S. Olympic Committee, the NCAA, etc. In fact, you could conceivably have a tax-exempt movie theater. So saying “entertainment” doesn’t get you anywhere.

    And saying “for-profit” it pointless. If a church is for-profit (and note that, for purposes of 501(c)(3), that means that shareholders have a right to a share of its profits, not that it makes a profit—as long as it doesn’t distribute its profits to shareholders or other interested parties, a tax-exempt organization can make a profit—see, hospitals), it’s not exempt from tax, even under current law. And if it does distribute profits, it’s already not tax-exempt, even if it’s a church.

    What’s more, I referred to the Branch Ministries case. In that case, the DC Circuit (one of the most prestigious federal Courts of Appeal, FWIW) held that the church violated the campaigning prohibition. As result, it was no longer exempt from tax under 501(c)(3) which, as I understand it, is the assumption you want to go with. Nonetheless, notwithstanding the fact that it was not exempt from taxation under 501(c)(3), the court said that donations received by the church would not be taxable to it. Unless I’m missing something, that meets the criteria you’ve set up. Although it would pay taxes on its investment income and certain other income, if any, that it received, it would still not pay taxes on donations it received.

    And now my family’s starting to get up, so I’m going to have to leave it here for now. As you’ve seen, I overspoke on the deduction side, but I’ll reply to (3) and (4) later.

  22. Other charities that would be no longer tax exempt are all kinds of noteworthy environmental charities, such as the Nature Conservancy, the National Geographic, Southern Utah Wilderness Alliance, etc. Sure, the end result of these charities might be, in part, to better people’s physical welfare, but that’s not the immediate impact, and I doubt their tax status would survive such a strict definition of “charity.”

  23. Dr. Cragun, sorry it took me so long to get back to you. I’m just going to respond really quickly to the other two points, then I’ll make an alternative proposal that would be more palatable to me.

    On the $8,000/year, I agree that that is probably a very low number; I’m more interested in the assumption that all 600,000 ministers get that. As to the property tax issue, I assume that churches in Chicago and New York are generally worth far more. But I’ve just spent the last week in South Carolina, where there is a church on every corner, many of which, I suspect (based on real property values in S.C. and the look of many of the churches) aren’t near $1.7 million. And I suspect that there are more churches in S.C. than in Chicago or New York. I don’t know how Tampa would fit into that.

    But as to my proposal: your main concern is the cost to the government of the various special tax treatments of taxes. But we don’t know what they cost, principally because there are a couple significant differences between churches and other tax-exempt organizations. One of the big ones is that churches don’t have to file a Form 990 (unless they have UBTI). The IRS Statistics of Income department has some wonderful information about the revenue, expenditures, etc., of non-church tax-exempts, but doesn’t have data on churches.

    In order to determine the cost of the tax exemption, then, why not propose that churches be required to file a Form 990. I doubt that would raise constitutional issues (though, notwithstanding yesterday’s big decision, con law rarely interacts with my specialty, so I could be wrong). Then the IRS would have all of the relevant information and could let us know the cost.

    Why is that important? Because, even if there’s no constitutional issue with removing the tax exemption for churches, it would require significant political will and expenditures of political capital. If it’s only going to provide a few billion dollars in additional revenue, it may be that the reform effort would work better aimed toward another target (like the Realtors).

    That said, though, there really is no reason to keep the parsonage allowance (which, Ray, would mean that the value of employer-provided housing wasn’t included in the televangelist’s income).

  24. Considering taxation is theft I applaud any group/person that can escape its clutches.

  25. Dr. Cragun’s point 2): On point 2, you failed to state one of my assumptions. I’m going to call it an unfair criticism given that you are working under a different assumption. I clearly stated in the article that I was treating religions as though they were for-profit corporations providing entertainment, like movie theaters, and defended that assumption because I had just argued that religions were not “charities”. I also argued that, if not charities, it could be claimed that their primary purpose was to entertain. Under THAT ASSUMPTION, the money they receive would not be donations but rather income (well, technically, revenue).”

    Hmmm… On of the reasons that there are advocates for separation of church and state is BECAUSE American citizens can, and do, define religious organizations differently. As a cleric, I would define them as manifestations of faith that are much greater than charities and certainly provide deeper and broader service than merely providing entertainment. I think you have to belong to a religion before one can pass beyond the “religion as entertainment” label. And I suspect that those who are offended by such a label have good cause — although I also suspect that that those compelled to use it have little understanding as to why. Value conflicts like this one among divergent people are often deep and uncompromising — and create tensions that the founders of this country were attempting to relieve by means of the first Article of the Bill of Rights. Government policies designed to tax religious organizations should be considered with great care (and caution).

Comments are closed.