Philanthropy Daily recently reached out to Benjamin Priday, a doctoral candidate in economics at Texas A&M. Priday has been involved in some interesting work on charitable giving and economics, including recent work exploring how charitable giving varies (or doesn’t) by income. A summary of those findings can be found at The Conversation: “Rich folks aren’t that stingy after all.”
This interview was conducted by phone and was lightly edited for length and clarity.
There aren’t very many people out there studying the economics of charitable giving. How did you first become aware of this area of study and interested in the topic as a whole?
My initial interest was kind of an accident: the result of a professor bringing me on board to review some of his own research, which turned out to pique my interest and turn me on to the field. But over time, I've developed my own appreciation for why the study of charitable giving is really crucial in our conversations about the role of society, government, spending and social service provision, and many other core public economic questions.
I know you've published some work on the effects on charitable giving of the Tax Cuts and Jobs Act (TCJA) of 2017. When that was being debated in Congress, I remember hearing cataclysmic predictions about how much charitable giving would drop. What were your findings about whether that tax law change had a big effect on charitable giving overall?
In our study, we approached the question of the effect of TCJA in two parts. First, we made an estimation of the tax price “elasticity” of giving— which, broadly speaking, is just a measure of how responsive people are to the tax incentives for charitable giving. So, we were asking, “Do people respond to these incentives, and how do they respond?” After that, we actually applied our estimates to the TCJA's provisions and got some estimates.
Admittedly, many researchers were doing this. But our contribution was that we were able look at people who didn't itemize and look at their giving. We found that, due to TCJA, there was going to be a decrease in the number of people itemizing. Quite a dramatic decrease—and that this would be true across the income distribution, but especially among people making over $100,000 a year. The cost for them to make a donation was going to go up, because they wouldn’t be able to itemize those contributions for deductions anymore. And so, presumably, these people were going to donate less as a response to the change.
But the TCJA also dropped marginal tax rates down, so people had an effectively higher take home income. So, we knew people were going to donate out of that additional income and wanted to account for both of the above factors, and their countervailing effects.
When we combined both of these factors to get an estimate, we found about a $9.8 billion decrease in charitable giving for the year, which is about 3.5% of what total individual giving is already.
You also ran a lab experiment in which you were trying to better understand people's propensity to give to organizations versus individuals in the context of disaster relief. Am I getting that one right?
Yes. We ran a lab-type experiment with adults from areas that had experienced a natural disaster, like a hurricane, in recent memory, so that our sample would be better equipped to relate to a hypothetical disaster relief situation.
We allowed these people to make a $60 allocation between themselves and a charity and to do so four distinct times, with each decision treated as its own independent exchange. In each decision, we gave people a different level of charitable giving to assess. One level of giving we offered in the experiment was to a real individual person impacted by a disaster. Then we compared that level of individual giving to organizational giving on the local, statewide, and national levels.
The really clear thing that came out of the experiment was that people decidedly like giving to individuals above organizations. That's not really a novel finding by any means, but it confirms a trend in research. Being able to identify an individual you are giving to increased giving tremendously among our sample. People would give about 50% of the $60 to an identifiable person— significantly more than what they chose to give to different charities.
Giving levels between local, statewide, and national organizations were more comparable. But we also found that people who reported a high level of trust in institutions were more likely to donate money to the national level organization. And people who were disposed to trust individuals over institutions donated more to individuals and local charities than the nationally-based organizations.
So, one of our important findings from this was that people's degree of trust in institutions and individuals mediates their giving decisions.
Now I’d like to talk with you about your research that's just been released. What are the common narratives about rates of giving from the wealthy and the less wealthy that you see out there that this work was meant to address?
There are quite a few narratives out there, and a lot of them are quite frustrating because they're based on research that I find quite suspect.
You can find articles with suspect science by searching Google for “stingy rich people” or “poor people more generous”. In hit after hit the takeaway message would be that "Yes, rich people are super stingy, and poor people are really generous." But a lot of those findings were based on means that were just not really corrected for anything, any sort of mediating circumstances in the data.
For example, in some of these other analyses statistical outliers or people who weren’t really low income were often being counted as low income. Other times there will be references to studies where researchers are drawing these really broad conclusions from watching a handful of people drive luxury cars at an intersection. Then from watching the behavior of these drivers they’ll make broad conclusions about how rude and stingy rich people are.
That's just not a very scientifically rigorous way to approach this question. I would not base any sort of broad understanding of the generosity of people with different income levels off of something like that. We wanted to document this question more rigorously than has been done to date.
So, your paper finds that people with higher incomes are giving a similar percentage of their income as people with lower incomes. That finding is probably surprising to hear for many because, as you mentioned, there's a narrative out there about the wealthy giving less—one based in bad numbers, because previous studies have only considered people who itemize and not people who take the standard deduction. Not factoring in those who take the standard deduction skews the data to make it look like people with low incomes are much more generous than the wealthy. But you’ve found that that's not the case, at least where your metric is a percentage of income that's being given away. Is that correct?
Exactly. We were looking at the percentage of family income, which is the whole income that a household gets. We importantly made a distinction between wealth—the value of everything you own—and income, which is just an annual snapshot of what you brought in.
A lot of people, if you read the econ literature, say things like "Wealth matters, but we don't have good data on wealth so we can’t include it." Well, we did have data on wealth, so we included it, and we show that wealth really matters. Even separately from income, wealth matters in the way people give.
As a percentage of family income, giving is relatively flat across the income distribution, but it's actually increased with higher rates of wealth. People with more wealth—assets and savings and things like that—give a higher percentage of their income than people with lower wealth on average.
There’s a question I’ve wondered about that ties together those two different studies you've been involved with. In your field work experiment, you found that people have a higher propensity to give to an individual than to an organization. And then in this study, you found that the wealthy are, depending on how you measure it, equally or slightly more likely to give.
It's interesting to put those two observations together, because I’ve long considered that for low income people, much of their charitable activities may be things that would never show up as charitable giving because they're not going through a 501(c)(3). If I give $20 to my neighbor who can't pay his utility bills, or I help my neighbor change his tire to avoid having to pay a mechanic, these things won’t show up anywhere in statistics. Those are charitable acts, broadly speaking, but none will be recorded as charitable giving because, they’re not going to a 501(c)(3) organization and showing up in their revenue stream.
But if I’m a wealthy person, I may not know very many people who need direct charity because my social circle of people contains fewer people in need. So, even if I have the same charitable intent with my money, it seems I'm much more likely to direct my dollars to a charitable organization that is in touch with those who need help.
That seems to establish a bias towards the type of charity that I would think a wealthier person is more likely to pursue, doesn’t it?
That's a good point. In economic terms, we will often talk about the substitutability of donations for volunteering: you might imagine that as people's income goes up, they have less time, but they have more money, so they donate money in lieu of volunteering. But, broadly speaking, you're exactly right. There are multiple possible expressions of generosity. You can give time, you can give through bequests, you can give through foundations: there's a whole array of ways to be generous. And the data that picks up monetary donations is quite possibly going to pick up moreso on the donations from rich people, because they’re expressed through things like 501(c)(3) donations.
So, what you’re suggesting is possible, I think. There is a lot of anecdotal evidence that poor people are more likely to give time or money to others in a more directly interpersonal, off-the-record way. I don't know of any systematic studies that would demonstrate that idea, though. I accept anecdotal evidence, but I’m hesitant to make broad generalizations about behavior based on anecdotal evidence.
I’d also note that if we’re talking about volunteering and other non-monetary expressions of generosity, on the wealthier side you have things like board involvement— such as running for school board for your kids’ school district because you care about providing your business expertise for the school board. Those kinds of non-monetary contributions, I think, also don't get picked up on the wealthier people's side. And so, while I'm not disagreeing with your observations, it's not exactly clear to me how everything would shake out if we broadened the definition of generosity.
Last question here for you: What's a question about charitable giving or a study you'd really like to tackle next—or maybe something you'd like to see someone else work on?
Something I'm working on right now as part of my dissertation is a hypothetical policy analysis of what would happen if churches lost their tax exempt status: what would happen to donations, what would happen to government revenues, tax expenditures, and what's going to happen to the churches on their side.
What are the implications of churches having to pay property taxes and income taxes and donations going down? It's a huge puzzle: a big mess that nobody has touched, because it seems so politically infeasible.
I want to do a first pass on the question—one that amounts to something like a policy brief, but the idea of this needs to be investigated more rigorously. I think that this is something that will be a policy discussion in the not-too-distant future. There are many people already raising the question of whether churches should really be subsidized by the government through the tax code. So having analysis on the question—both my work but also, hopefully, more rigorous analyses to come—that could point to the implications of such a change of policy feels really important to me.
That's really interesting. I think it was Beto O’Rourke, on the Democratic debate stage, who said something about how churches with certain beliefs should have their nonprofit status revoked. I think I agree with you that this may well be an important policy consideration sometime in the future.
I think the issue will likely come to a head in the form of a civil rights discussion, exactly because of the legalization of same-sex marriage and what that means: cases where religious freedom butts up against civil rights law will catalyze the conversation.
And then, other people who are tired of taxes subsidizing religious groups when they're not religious will likely join the conversation, saying, "Yeah, regardless of the potential civil rights violations of churches, we don't think their tax exempt status should be a thing on principle." So I think the question of it will come up in the future, and that it’s important to start exploring the implications of churches’ tax exempt status today.