Does Dylan Matthews Read the Reports to Which He Refers?

There is an occasional conversation that happens in Leftmost Blogistan, and revolves around a simple question: Should we tax religious organizations?

The current iteration appears to have started with Matthew Yglesias over at Slate, who actually has a not-so-terrible point:

Meanwhile, nobody thinks churches and other religious institutions should silence themselves on the important issues of the day. On the contrary, discussing moral action is at the heart of many religious enterprises. And much moral action plays itself out in the arena of politics. So trying to say that churches should get subsidy when they don’t endorse candidates is de facto a kind of subsidy to religious doctrines whose views happen to lack strong partisan implications. So if your faith says “abortion should be illegal and spending on the poor should be increased and it’s too bad neither candidate supports that” you’re golden, but if your faith says “abortion should be legal and spending on the poor should be increased so good for Barack Obama” suddenly you’re in trouble. That’s perverse. Just make everyone pay taxes.

It’s important that he writes “make everyone pay taxes” at the end. After all, the same logic applies to all non-profits with an interest in public policy: presumably, non-profits interested in climate change will be better served by any given Democrat than any given Republican, non-profits seeking to end abortion will be better served by any given Republican than any given Democrat and so on. Indeed, it is telling that Yglesias’s description of Bible studies in the previous paragraph is, “(what are Bible study groups but the original book clubs?) that doesn’t need an implicit subsidy,” given that ‘literary purposes’ are among the exempt purposes explicitly laid out in the in section 501(c)(3) of the Internal Revenue Code.

So let me begin with this statement: I think that there might be something to a drastic revision of the tax code that would eliminate both the charitable deduction and non-profit status. This has a lot to do with the fact that 501(c)(3) status and the charitable deduction are poorly understood piecemeal affairs that I suspect served more to meet the cultural expectations of the governments of the time than any particular policy agenda we have today. Why not simply let organizations serving purposes that are “charitable, religious, educational, scientific, literary, testing for public safety, fostering national or international amateur sports competition, and preventing cruelty to children or animals” operate like any corporation? And why not simply let charitable personal giving – which, of course, is giving that is done without expectation of personal return - actually be charitable? There are historical answers to these questions that I need not approach here. Needless to say, we are permitted to rethink these questions and should probably do so in the context of rethinking the tax code as a whole.

Unfortunately, Yglesias’s article was picked up by Dylan Matthews at Wonkblog and this in turn was picked up by Steve Benen at The Maddow Blog. Both linked Yglesias’s article to a ‘study’ by a faculty member and two students at the University of Tampa – Ryan T. Cragun, Stephanie Yeager and Desmond Vega – purporting to show that Americans indirectly subsidize religion to the tune of $71 billion. Matthews adds charitable deduction to this amount and finds that Americans “give religion more than $82.5 billion each year.”

It should surprise no one to discover that there is at least one glaring problem with Cragun et al‘s article: Research Report: How Secular Humanists (and Everyone Else) Subsidize Religion in the United States. On a skim through the article, I was struck by this paragraph:

In order to calculate the government subsidy resulting from tax-exempt donations, we assumed that religions would be taxed at the maximum federal corporate tax rate, given their revenue (for example, the United Methodist Church parent organization falls into the highest corporate tax bracket, as would most local congregations). Using this assumption, the subsidy to religions in the form of lost corporate tax revenue to the federal government is about $35.3 billion annually. We estimated that states subsidize religions to the tune of about $6.18 billion per year as well by not requiring religious institutions to pay income taxes. Given the literally thousands of different local corporate tax rates, we did not calculate the subsidy to religions from local governments, but it would likely add hundreds of millions of dollars more in subsidies. [Footnotes omitted]

Slightly earlier, Cragun et al use a number from the 2010 Giving USA study to provide an overall amount of money donated to “religions”: $100.95 billion. They then apply the federal tax rate – citing this Wikipedia article – to this amount to  determine that the federal government misses out on $35.3 billion each year (100.95 x .35 = 35.3325).1 The problem is that corporate taxes don’t work this way. That is: it isn’t simply a matter of taking gross revenue and multiplying it by the marginal tax rate.

Rather, complex rules govern business taxes. For example, businesses are permitted to deduct a wide array of expenses including wages and salaries, repairs and maintenance, rents, certain other taxes and so on. While it is not possible – and I am not qualified  - to go into a detailed explanation of the corporate tax codes here, it is certainly the case that no religious organization (and no business) would simply pay the a single rate on all of the revenue it receives; hence the ability for some large corporations to pay an effective tax rate of zero. Instead, churches and other religious organizations would pay a taxes on a portion of the revenue they receive and that portion would likely be a minuscule part of the $100.95 billion that religious organizations received in contributions in 2009.2

While this is obviously not a detailed account the article cited by Matthews, it should at least call into question the quality of the amounts provided by that article. It would be intriguing to see the results of an actual study of religious congregations and potential taxes, perhaps by convincing a diverse group of congregations to calculate their hypothetical business taxes for one year. I, for one, would be very interested to see the results. Regardless of that, however, Matthews would do well to review the articles he refers to more carefully.


  1. This is actually a fun metric to apply to other areas of the non-profit sector. Using the same Giving USA 2010 study, we can determine that the federal government misses out on: $14 billion from educational institutions; $11 billion from grant-making foundations; $9.5 billion from human services organizations; $8 billion from public-society benefit organizations; $2.5 billion from health organizations; $4 billion from arts, culture and humanities organizations; $3 billion from international affairs organizations; $2 billion from environmental and animal welfare organizations; $1 billion from foundation grants to individuals; and $10 billion from ‘unallocated giving’. That’s a total of about $65.5 billion! Imagine what could be done with that kind of money or, of course, the more than $100 billion that would be gained simply by ending non-profit status altogether. 

  2. Just for fun: a GAO Report found that large profitable corporations paid an average federal tax rate of 12.6%. That would lower the so-called ‘subsidy’ to $12.7 billion or less than 40% of Cragun et al‘s original amount. Alternatively, that would be about $38 billion from the entire non-profit sector. Of course, being non-profits, many organizations including churches would likely pay far less in taxes. 

Leave a Reply