What Research Says about Giving Habits Amid Coronavirus and Beyond
UChicago professor John List shares social sector research insights at On Board.
- May 17, 2021
- Rustandy Center - Social Entrepreneurship
In 25 years studying US fundraising data, University of Chicago Professor John List noticed a gap between what the science says about giving habits and what the charitable market and nonprofits raising money are actually doing.
His research aims to close that gap, examining the underlying motivations of donors across income groups and sharing insights with nonprofits, most recently during the virtual On Board conference on nonprofit board service, hosted by Chicago Booth’s Rustandy Center for Social Sector Innovation.
Instead of looking at charitable donations the traditional way—by examining the S&P 500 and the stock market—List studies data from 1971 to today looking at total giving as a percent of US income.
Through that lens, total individual giving to charities remains stuck at 1.8 percent to 2.8 percent of income, said List, the Kenneth C. Griffin Distinguished Service Professor in Economics.
So how can nonprofits, board members, and others move away from fundraising that leans on gut instinct to developing strategies to increase donations amid (and long after) the COVID-19 pandemic?
“I have tried to provide scientific insights and tried to explore where we can use science to enhance both what we know and also to enhance the [donation] bucket,” List said.
Below are four insights from List’s fundraising research.
Bigger Gift Matches Aren’t That Effective
Fundraisers laud a good “match rate,” List said, a gift that “challenges” or encourages someone to give because a donor has committed to matching every dollar raised. But do matches actually raise more money? And does a 2:1 match (or more) work even better?
Not really, according to . In a field experiment with a US nonprofit that works on social and political issues, List sent a letter to 50,000 households promoting one of four match rates: 1:1 match, 2:1, 3:1, and no match for the control group.
Nineteen percent more money was raised per letter in the matching treatments, but the effect mirrored the overall response rate, which was 22 percent higher in letters with matches.
So while matches did a great job of bringing in new donors, List said, no additional dollars were raised from the higher match rates. To avoid wasting matching gifts, he encouraged nonprofits to keep giving matches to 1:1.
“One Size Fits All” Leaves Money on the Table
Research shows people are motivated to give for different reasons, List said, so “one size fits all” fundraising campaigns likely miss out on donations.
Most donors, especially small donors, are not buying a private good when they give. “They’re giving because they enjoy the warm, fuzzy feeling of helping or doing a good thing,” List said. Data shows that women especially give more due to altruism or social pressure, while men are more price sensitive.
And while men give more than women on average—$2,127.09 annually compared to $857.50—List said women out-give men across every level of income (except the ultra-wealthy) when you look at percent of total income.
Non-price incentives—like offering a mug or tote bag—can have strong effects on giving, he said. And the success of long-run campaigns can depend on these incentives to attract first-time donors.
List encouraged fundraisers to lean on insights from behavioral economics and machine learning and algorithms to do a better job targeting your average or “modal” donor.
How COVID-19 Changed Giving Habits
The coronavirus has affected giving habits, with fewer people giving to charity since the start of the pandemic, according to recent data List has collected across the United States and Germany.
List also noted that the people who are giving are donating less, and they’re also giving to different types of causes than in the past—notably, charities tackling health-related issues.
People are also much more price sensitive than before the pandemic, something he’s seen while studying the same group before and after the pandemic started.
To help the philanthropic sector, List said the Biden Administration could provide more tax incentives for donors and continue allowing an existing non-cash giving loophole that lets high-capacity donors to directly donate an asset (like Google stocks) to a charity to avoid paying the LT capital gains tax.
Ultra-wealthy Donors are Similar to Small Donors, With Exceptions
In a recent analysis of 2012 IRS data, researchers see how much power ultra-wealthy donors wield. Looking at donations across adjusted gross income, List said the top 1 percent of people gave close to 40 percent of total US donations, and the top .01 percent accounted for 14 percent of total gifts.
List encouraged fundraisers to try some of the same tactics on high-capacity donors, since their giving habits aren’t that dissimilar from the modal donor, citing data from a field experiment he did in partnership with Chicago Booth before the pandemic.
Researchers sent solicitation letters with varying messages and prices to nearly 6,000 high-capacity Booth donors, each with a median annual giving of at least $25,000. Donors responded to letters emphasizing program quality, but different match rates had little impact and too many contacts could have a negative effect (aka donor fatigue).
Unlike small donors, who act quickly, high-capacity donors had a longer lag time between the ask and a donation. By leveraging List’s insights, researchers helped generate over $30 million in incremental donations for the university.
“Everything I’ve talked about today should be applied to make our donor pipeline bigger,” he said. “Thanks for trying to change the world for the better.”
Interested in attending a future On Board session? Learn more and register for the 5/21 event.
So, hello, everyone who is with us today. Welcome to the third day in this year's virtual On Board conference on nonprofit board service, which as I think everyone knows is hosted by our wonderful Rustandy Center for Social Sector Innovation at the University of Chicago Booth School of Business.
For those of you who may not know me, I'm Starr Marcello, the deputy dean for MBA programs at Chicago Booth. In my role as deputy dean, I have the great pleasure of working closely with Chicago Booth's research and learning centers, and that of course includes for Rustandy Center. And connect students with the events, programs, courses, experiential hands-on learning opportunities to apply the things that they are learning while they're sitting in the classroom, to tackle the things that they encounter in the real world. So now in its eighth year, which is just extraordinary 'cause I remember when this conference was getting started, On Board conference takes a page from this approach and by leaning into the belief that nonprofits and their boards of directors have the potential to drive real, social and environmental change.
So I'm excited for you to see the sessions today. Today's On Board sessions are gonna tap into this change making potential by honing in on a vital nonprofit topic, one that I think we all have different experiences with and different knowledge regarding just fundraising and philanthropy. My hope is that today's sessions will equip you with tangible resources to evaluate which new strategies and fundraising that might have emerged over the past year and could be specific to COVID in the pandemic, which ones are those strategies and which ones will persist through this year, and when we're on the other side of this experience we've all been through?
So first we are gonna kick things off with our keynote featuring UChicago professor John List, who will discuss his research on the motivations behind why people give. After that, we'll transition into a breakout session called fundraising for the future innovations that will last led by Margie DeVine. This session is gonna provide a roadmap for how you can start to plan out your fundraising strategies for this year and beyond this year.
If you are on this call and you did not register for the breakout session but are now considering joining, you can do so by going to the all sessions page on the Cvent Attendee Hub, and then selecting Add Session. So we are more than delighted to have you join that session.
So I'm gonna transition now and introduce our keynote speaker. John List is the Kenneth C. Griffin distinguished service professor in economics at the University of Chicago. John has revolutionized modern economics by studying people in the real world through field experiments. His research and publications include over 200 peer-reviewed journal articles, he's also written a number of books including an international best-seller called "The Why Axis: Hidden Motives and the Undiscovered Economics of Everyday Life." He is quite extraordinary, you are in for a real treat. It's really an honor to have him join us today. And I encourage you to read his full profile on the speaker's page, which you may have already done. It's on the Cvent Attendee Hub. You can also of course, google him and see all the wonderful things that he has done.
Before I turn it over to Professor List, I want to just tell those of you on the call today that we are also excited to hear from you. And you may have questions throughout the session, if you do have a question, there will be a chance to take yourself off of mute and ask your question. You can also choose to submit a question through the Q&A chat function within Zoom. You should be able to see a chat button at the bottom of your Zoom screen, and we'll try to keep an eye on that as well. I don't know how many questions we'll be able to get to. I hope you ask lots of good questions, professor List will try to get to as many of your questions as he can.
In case we run short on time, at the very end, I just wanna offer a couple of reminders ahead of our next session. The next session is gonna run at 12:30 to 1:30. After the keynote, after John talks, we'll take a quick five minute break. You can join the breakout session by navigating to My Schedule on the… The My Schedule page on Cvent and by joining Join Session, selecting Join Session.
We do also want your feedback that helps us plan for next year's conference. We'd greatly appreciate it, if you would take a brief survey by going to the My Event tab and clicking on Take Survey. Okay, that's all the housekeeping I think I am gonna provide. I want us to get to the main event. So with that, it's my pleasure to hand things over to Professor John List.
Starr, thank you so much for that overly kind introduction. And thanks for having me today, and Prentiss, and Caroline thank you so much for your professionalism through this in the onboarding. I really appreciate it.
So I have a slide deck that I would like to lead our discussion. Now, as Starr mentioned, as we go along here, what I'm hoping for is that we have a discussion. We have close to an hour to talk and my outline is given as follows. I'll start by talking about the charitable market and give some facts. Then I'll do something called opening my laptop, where I'll talk a little bit about some older results and then I'll talk about some newer results, and during the discussion hopefully, we can talk about some COVID results, what we've been finding so far.
The uniqueness of the talk will be that I will pivot from what I call the modal donor to the modal dollar. And what I mean by that is I'll have some data on the ultra wealthy and their giving patterns, which is highly unusual in the scientific area at least just because it's very difficult to generate data on these types of givers. And then I'll conclude with some thoughts at the end.
But as Starr mentioned, along the way if you have any questions, I'll be glad to stop and take those questions. I'm sort of in the dark here. I don't see very many people, but you know, send it in the chat or just unmute yourself. I have no problem being interrupted. This is what this should be is a discussion, okay?
So when an economist talks about the charitable market, a very simplistic way to think about it is to say there are three major players. And those three major players include the government, which of course, I have at the top here, and the government of course sets rules about charitable giving tax breaks, for example. Sets rules on marginal tax rates, which affect individuals and they also affect charities 501c3s. The government also transfers through block grants and other ways, actual resources to charities, and that's a rope that we have studied. We've studied this relationship between governments and people as well. Where I'm gonna focus on today is a relationship between donors and charities. And I'll tend to focus on what works for charitable organizations to generate higher levels of participation from their partners and their friends.
Now, where I wanna start is just to run this time series from 1971 until today. So the dark line gives you the S&P 500 and it's normalized, so 1970 equals 1. And the red line is total giving from individuals. And again, I'm normalizing this, so it starts in 1971 at a level of one. What this shows you is by and large an aggregate, a good predictor of the total dollars going to charitable organizations in any one year, is the S&P 500, which is a metric of the economy, okay?
Now people tend to look at those figures and if you read things like philanthropy magazines and what experts say, they look at that figure and they say the big problem here is trying to figure out where to put the apple basket, because a tree keeps growing exponentially. But it's just trying to figure out where we should be asking and where we should be putting our resources when we invest in trying to induce people to give to our cause.
Now a very different way of looking at the data would be to say over that exact same time period, using the exact same data, what is total giving from individuals as a percent of income? Here's what that looks like. Now, when you see it (laughs) you might say, what just happened to my vibrant economy? This was a vibrant sector in one slide but it wasn't such a vibrant sector in another slide. Now I've written a paper where I go back to the 1920s and I have given to religious causes and other causes all the way back to the 1920s. And it follows this exact same tube-like relationship. In fact, religious giving ends up being like 1 to 1.3% over time, and the other giving of course follows this as well. So now you should ask yourself a totally different question. It's why are we stuck in that band of roughly 1.8 to 2.2% of income is given to charitable causes?
Now, I like to think of this as a Mark Twain problem. I think in areas where we don't have a lot of science, such as I work in early childhood as well, I help firms as well and I can use this quote for any number of sectors, that really all you need in life is ignorance and confidence and then success is sure. And that's exactly what we have in the area of fundraising. I've worked in this industry for about 25 years, and I'll talk a little bit about my story about how I got involved. Now, I blame both the practitioners and the scientists like me. The practitioners are to blame. When I first got into this at 1997, I talked to a lot of experts and by and large decision-making was made on gut instinct. It's how my boss used to do it, so that's how I'm gonna do it. It's how I feel we should be doing it, so I'm not gonna rock the boat. Even if people back then wanted to use science, here's where I blame us. It would be very difficult to use science because we have not produced the scientific insights that are necessary.
If you think about this, like the COVID problem, that was easy. We had in place ways to test the vaccinations, we trusted those tests, and then everyone now puts this foreign substance in their arm based on the scientific evidence that people have presented. We really we've… Ever since 1963, the FDA has adopted that approach to science. We haven't done that in the world of fundraising or even more broadly in the world of social sciences.
Again, I can say the same thing about early childhood. So my research, I have tried to provide scientific insights and try to explore where we can use science to enhance both what we know and also enhance the bucket, so we can get out of that 1.8 to 2.2% of income that's given every year to charitable causes. So let's use a simple example. We all know about match rates.
So early on in the late 90s when I started running my own fundraising (indistinct) at the University of Central Florida, I was given $5,000, and I was told, we want you to leverage that $5,000 to raise more money for the Center for Environmental Policy Analysis at the University of Central Florida. So you can think of various ways to use upfront money. You can announce it as seed money which we tried, you could announce it as matches. You could use it as… To purchase gifts to give to donors.
Now in the world of matches, of course, the experts have very strong opinions on that. And this comes from what people told me back then was the Bible. It was Kent Dove's handbook on charitable fundraising. And here's what Kent too is very good, and Kent's doing the best he can back here. Here's what he tells us "Never underestimate the power of a challenge gift in that obviously a one-to-one match is better than a one to two challenge, but a two to one is even better than both of these." What he's basically saying is what is the law of demand? We've all taken principles of economics and we know when the price goes up, we buy less of the good. When the price goes down, we buy more of the good. So if, when we give money to charitable causes, if we're buying Snickers bars and t-shirts, and Uber and Lyft rides, this is great intuition. This is the intuition from a standard Econ 101 model. So when I entered this area I said, called Kent and said, where the data that show that this is true?
And I talked to several practitioners and they all knew it was true, but none of them could give me data that showed that it was true. So now that's easy for a person like me who loves to run field experiments. So one of the early field experiments was a partnership with the ACLU, and we simply put roughly 50,000 households in one of four groups. So the ACLU sends letters every month. They just gave us access to one of those months to try out our experiment. That's typically how our partnerships work.
So in this particular experiment, if you were in group one, you received a letter that was identical to groups, two, three, and four, except in this letter, we said, we have some money upfront, every dollar you give, we can match it dollar for dollar. So if you give &100, the ACLU will receive $200. That was group one. Groups two and three are identical, except we're just changing the match rate. In group two, if you gave $100, the ACLU would receive 300, 'cause it's a two to one match. Group three, they'd received 400 because it's a three to one match. Group four are households that receive the same letter but with no mention of upfront money or a match. Okay, so simple test. Let's see what happens.
So, the first result from the simple test is if I just compare group four, which was the control group, and I group all of the other match groups together. So I'm just looking at if you had a match versus if you didn't have a match, what we find is you raise about 20% more money in the matching treatments. That's great. Check. Now, you might ask where does that effect come from? In these data the effect comes purely on what economists call the extents of margin.
There are total new set of people who decide to give in that the response rate was about 22% higher. Oftentimes, we talk about establishing a really strong base, a donor base. What we find in not only this study but in studies that replicate this work, is that matches do a really good job of bringing new people who would not give without a match into the organizational quote warm list, okay? So this result all was levered on the extensive margin or the response rate.
Now, you can say, well, John, what about the match rates? Three to one, two to one, one to one, are people buying Snickers bars? That's what you should ask. In fact, they're not. When you look at the three to one, two to one, and one to one, all you're doing is effectively wasting match money because just the fact of having a match, that's what worked, the actual match ratio did not push more donors or higher levels of gifts.
Okay, so now in this world where we're to the point where you say, well, how low can you go? What we found is you can go to about one to two in the typical drive even though there's heterogeneity across people in drives, but in general, I won't go lower than one to two, but I certainly would not go above one to one, because the match tends to work as a signal of quality. So when people see an anonymous donor or Bill, we've done some experiments with Bill Gates and Melinda Gates giving money that works even better than an anonymous donor if they have more insights into the charitable area. And I can send that paper, if any of you would like to see it. But just the fact of having a match is a really important signal of quality of the drive.
Now I'm not gonna go through 20 years of research but let me back up and say, well, first of all, one thing we've learned is that most donors, especially small donors, they're not buying a private good. Many of them are driven by warm-glow. They're giving because they enjoy the warm fuzzy feeling of helping or doing a good thing, okay? They tend not to be driven by prices. The smaller donors. Now, what we find is that non-price incentives can have very strong effects on givers, especially cold-lists givers. So what I mean by that is what we find is that if you have a small donor gift, and you look at the effect on cold-list and the effect on warm-list people, the small donor gift can make your cold-list look like and act like it's a warm-list. So that's the general effect that you find and we all know that warm-list people tend to be more committed to the cause, and if we compare warm and cold, we get a bunch of zeros on cold. But what we find is that a small gift does a really good job on cold-list people, it does okay old warm-list people, but not as well as on cold-list people.
Third, it's… This is a warning that when you do these types of interventions, that it's one thing to measure what I would call a short-run substitution effects. So, you know what works in the here and now over the next six months. But it's really important for the long-run success of your organization, how you initially bring people into your cause. In particular, the best type of incentive that we have found is signals of quality of your organization or the drive. If you get people in with, for example, matches that are signaling that some insider has information about the quality of your drive, that ends up leading to people being more likely to stay committed to your cause.
Number four, and where I think we can make a big dent is we should be using machine learning and artificial intelligence algorithms to do a better job on targeting. For example, in most of our datasets, we find that men are much more price sensitive than women and that women are giving more out of altruism and social pressure compared to men. So that's just one cut of heterogeneity but using machine learning, say random forest and some causal type of approach that is now becoming standard, and it's exactly what we're doing at Lyft and what organizations are doing. We should be doing the same thing because you could leverage a lot of insights based simply on heterogeneity of people or targeting.
Now fifth, behavioral economics. Now the Booth folks will like this. Dick Thaler, one of my colleagues and a member of Booth, one of the pioneers in the area of behavioral economics, pardon me, will like this point. Is that behavioral economics can really go a long way to help you raise money and I have several examples of that. I have one at the end of the slide deck that I typically go into right now during this talk. I saved it for the end because I wanna bring out some high-net worth giving type of evidence, but if we wanna back to this example in the end, it's an example that I have with Smile Train. And it's a partnership that we worked on that leverages a behavioral economics insight to really help them raise a lot more money in some of their charitable drives. Okay, so we can come back to the behavioral economics should you have interest?
Okay, so everything that I have just told you about is really what I would call the modal giver. These are givers who will give up to 500 or $1000. Those are the givers who tend to be in the experiments that people run. Now you can say, well, is that the modal dollar? Okay, so here is where I'm going to give you work. We're digging into the IRS data and I think I've discovered some really interesting facts, and I wanna give you some of those facts, and then I'll talk about a field experiment that I've done with Booth recently. So this is hot off the presses. This is just taking one year, 2012 tax year. We've looked, we have data now from 2007 to 2018 but this will kind of give you a sense. So in this figure, I have proportion of nationally declared charitable giving on the y-axis and on the x-axis, this will be in a lot of the coming figures. This is a national adjusted gross income distribution.
Okay, so what this tells you is each of these dots gives you the fraction of overall charitable giving that comes from that percentile of adjusted gross income. So you're going, going, going, going, going, wow, look at this. The top 1% of people in AGI, not the top 1% of givers but the top 1%. In the AGI distribution top 1% of income earners are given close to 40% of the gifts. Okay, now what I'm gonna do for you, is I'm gonna dig into that particular dot right here, okay? So now what I have here on the x-axis is the 99th to the 99.9th percentile people. As you going, boom, boom, boom, boom boom, oh, wow. So now the 99.9th percentile people are giving 25% of the overall charitable giving in the IRS data. Now I'm gonna dig into that dot there. Boom, boom, boom, boom, boom, wow, the top 1000 people. So to be in the 99.999th percentile, you needed to earn about $32 million. In adjusted gross income in 2012, it represents about the top 1056 people, so roughly what you can take from this is the top 1000 people are giving 14% of the charitable gifts that individuals give in 2012.
And this isn't just cherry picking, this happened from 2007 to 2018. Okay, it's what's called a power law, where you start breaking these distributions down and you look at city size, you look at growth rates. A lot of things follow power laws and charitable giving is no different. What's interesting about charitable giving, you can do these exact same cuts on income and to explore income inequality. What's interesting about charitable giving, which is different than income inequality, is there's a lot of transition between people across these different dots with income. There's very little transition through these dots between these dots in charitable giving.
The charitable giving distribution is much stickier. And it kinda makes sense because charitable giving is based on both income and wealth. And even though your income might go down, your wealth tends not to change a lot over time in relation to other people, okay? So that's sort of an interesting first fact.
Now what's also interesting is how the rich give, and this is gonna have very important implications for the tax law changes that people are talking about. So let's have an example. I've recently sold Google stock to give to my alma mater, the University of Wyoming. And I followed number one here. I had Google stock, I sold it, I then paid long-term capital gains and I donated the rest. Call that the dumb job list approach to giving. Because number two, what I could have done is I could have said, "Hey University of Wyoming, you just take the Google stock and let's talk about what the numbers look like." They would have been better off and I would have been better off. Okay, so strategy two means that the full asset value can be deducted from income. Now it introduces a really interesting tax incentive for asset holders that clearly is highly regressive. And I'll show you in tax data why that's the case.
Now, again, my strategy, I'm a meager fellow. I had $11,000, huge capital gains of $10,000. So what I do is I sell it, I pay the capital gains tax and I can give 9,000, and now my tax refund is 3,150. That's a simple math from John List dumb strategy one. Okay now, if I could do it all over again, here's what I would have done. I would have just taken Google, and I would have said, look, I have $11,000, I'm gonna donate that. Now I'm gonna have my refund of 3,850 from that gift. University of Wyoming gets more, I get more in the end and well, the federal government is out, but University of Wyoming and dumb John are better off.
Okay, let's now dig into the data and see if that kind of giving is evident using the exact same figures. Okay, so what I have here again is this is the share of the total in the y-axis, and I get the AGI distribution here. So this is the non-cash share of giving and this is a cash share. From this it looks fun, looks kind of normal. I'm gonna now dig into this 1%. Here's what that looks like. Again, this is non-cash share. Now we're up to about 10 or 20% in cash shares, 50 to 70, I'm gonna dig into here now. Whoa, now we get to 20 to 30, 40 to 50 here, let me see if I can dig one more time. And now I'm getting that the ultra-wealthy are using this stage, let's say tack number two, 40 to 50% of the time for their dollars. That's a very different type of giving than what we observed, of course, across the income distribution. That's how it becomes regressive.
Now, if the new tax code comes into being the new marginal tax rate in the 43% on the long term capital gains and they don't close this loophole, that's the best chance for the new tax overhaul to lead to the charitable sector growing. Because this incentive ends up being very strong, and if you have a high marginal tax rate of say 39%, and you have the capital gains tax of 43% and you continue to allow the non-cash giving, those two together will lead to a lot more giving and aggregate according to our best models. So those laws alone will significantly help our sector as long as they don't close this loophole.
Let's look at men and women. Now the world is actually becoming a society of singles. What I've taken here is prosperity, a little prosperity figure and the darker shade is a greater percent of singles by state. So as you can see in the US, roughly 50% of taxpayers are singles, they're filing as singles. So now I'm gonna look at all of those data, and the first thing that jumps out. If I just look at annual averages of giving, men give about two and a half times more than the average woman. Average woman will give about 900, $850, average man about $2,100. But what's interesting is that fact is entirely misleading because it hides a super interesting fact. And the super interesting fact is in this figure, okay?
So I'm gonna take a minute to explain this. Again, this is the share within the income bin here, this is the AGI distribution. So the way I want you to think about this is, at the 50th percentile of AGI, that's people earning about $60,000, women represent about 50% of the people in that bin, okay? That's what the lighter dot means, but they give about 70% of the charitable dollars from that bin. So as you can see at every income level, except for the very end, women are more generous than men, which you wouldn't get that from the averages but if you look by percentile bin or how much people actually earn, women are much more generous than men.
Now, here's what happens in this fairy tale here. This point out here is women are just much less represented than men in the deep right tail. And then secondly, amongst the ultra-wealthy, men actually give more than women. So the big reason why the average looks the way it does, is because men are more represented in the deep right tail of the income distribution and in the deep right tail, those men tend to give more than those women. It's kind of an interesting fact that I didn't know until I started to dig into the distribution. Now, from there, you can see that high income earners are driving the market. That's also kind of unique in that they give non-cash gifts. Now, women at every level of income are more generous, but they don't look more generous in the average data just because of the deep right tail.
So to bring all of this together, you might say, well, John, you gave us some facts about the modal donors from experiments, you gave us some facts about ultra-wealthy donors give a lot of money, which you already knew, you just didn't know the exact facts like I told you, can you put those all together? Yes, we can.
So recently we've partnered with Booth to run the same kind of experiments we've run with "ordinary people" but instead run these experiments with high capacity donors. So we have 6,000 of these donors, a median annual giving capacity is about 25,000. Now, just to give you a sense of the sample, about close to 50% made a gift in the last two years prior to our experiment. In a typical year, our subjects aggregate donations to the school is about 20 million. Okay, so this is a sample that we're talking about.
Now Booth sends out an average of about four donation letters per year to individuals. They gave us access to one of those mailings and we will essentially follow the literature, and we're gonna vary those same things that the literature has varied to see if the ultra wealthy are behaving in the same way as the modal donors. So here's what we find. In many ways, high capacity donors are similar to model donors. That's sort of good news from a perspective of trying to translate the science that we've generated in the last 20 years to perhaps wealthier potential donors. How are they similar? Warm-list people are much more likely to give than cold-list people. That's sort of an obvious result.
Interestingly, signals of program quality using match dollars or donor gifts, those work. Those work to influence high-capacity donors. Now again, much like the modal donor, the price of giving doesn't matter that much in terms of match rates. We know the price of giving matters when it comes to tax rates. We do know that from the IRS data, but it doesn't matter very much if at all, in most cases, it doesn't matter when it comes to the actual match rate.
Now we also find, and this is kind of opening up a whole Pandora's box, is that you do see a lot of donor fatigue. So we did some experiments on number of contacts to high capacity donors, and that can be detrimental. And we sort of have an idea that that's true, we just don't know how far to push. As fundraisers we don't know how far to push that. Now, unlike the typical smaller donors, our givers are actually responding only on the intensive margin, and often with a longer time lag. So the time lag tends to take months and months after the initial drive, whereas a lot of small donors will act immediately.
So in the end, and I can send all of you the academic paper that we've written, if you look literally at the data, our intervention generated over $30 million in incremental donations to Booth, okay? And that's simply using the behavioral insights, and the behavioral motivations, and the science that we had learned about in the previous 20 years. Okay, so I'm gonna stop there and I've gone a little bit over my time, but I would love to take any questions that you have, and if you'd like to hear about Smile Train or patterns in the geography, or any questions like that, I have a few more slides, if you'd like to take a look at those. But I'll stop there and turn it over. I think Prentiss is in charge, but I'm not sure.
Yeah, I'm happy to help out. If you raise your hand or submitted a question in the chat, I can certainly monitor that.
I've got a question.
Is there any research on bequests as a portion of giving and what influences that? I've seen that to be a huge, huge driver at least in the sector that I look at?
Yeah, so we've looked a little bit at bequest. When you look at bequests, you're right. They tend to be a tiny fraction. Historically, they've tended to be a tiny fraction of overall individual giving, but when you look at the last 10 or 20 years, that is probably the most vibrant sector in terms of growth. Now, I've done a little bit of work in partnership with religious causes, and bequests, and trying to explore what are the reasons why people become re-engaged with religion and they bequest dollars to religious causes. I can give you those insights, but I haven't worked with any other organizations. In the science and economics, it's pretty thin there in terms of, do the same types of features work? I think warm glow will likely have a significant impact in terms of bequest giving, so I would say that drives that tend to tickle or induce feelings of warm-glow would be most likely in the area of bequest to work, but by and large, we don't have a ton of science around it.
And I think as a whole, that's an area where I myself would be super interested in partnering and working on, because like you, you can see our populations age distribution and you can see how people are making commitments too. They wanna make sure they have good end of life care. They wanna make sure that they live the last several years in comfort, but they also want their money to do good. And they want to make as deep of impact as possible. So that, you know, how to illustrate that to donors is a process and much like calling a warm-list or calling a high-capacity donor. But to answer your question for one last time, there's not a ton of science around that.
John, we have quite a few questions in the chat, I'll read out some.
One question from Diane Rodriguez Brand saying, do your modal giving studies apply to corporate foundation giving and the types of appeals like match giving?
Yeah, so that's a great question. So when we've been looking at both giving by corporations and giving by individuals within corporations, a lot of the insights that I just talked about in terms of the modal giver and how they translate to the modal dollar, they also translate to those parts as well. And by and large, what happens is, the magnitudes change a little bit each time but the general direction stays the same. So there are ways to induce more people to give and keep them committed to your cause using the insights that I was talking about today.
Right, and we also have some questions related further information regarding donor fatigue. And so do you have any data on, you know, when does donor fatigue tend to occur? How much over time?
How much is too much?
No, absolutely. So, yeah, we've done these types of experiments where we randomize people into receiving one letter a month versus two letters a month, versus one letter a quarter, versus one letter every six months, versus one letter every 12 months, and there is a lot of heterogeneity in that. So what I mean by that is this is where machine learning really helps because after you have a bunch of data and that type of experiment, you really learn the types of people based on both their previous giving patterns, you know, what the trajectory is, and also their demographics. Those combined are pretty predictive of a donor fatigue model.
And well, let me give you kind of some general thoughts. Never ever in any of our data would you tweak a person more than once a month. Most people in the data you would tweak once every other month. Some of the data, like as you move up the income distribution, we're talking one call every six months and of course, you guys all know this, no letters or a phantom phone calls from our phonathons. So those are sort to be general guidelines that we would learn from our data, and with the caveat that there's a ton of heterogeneity. In that if I just followed the general guidelines, so that's what I would tell you, never more than once every month.
Ron, it looks like you came off mute, do you have a question? No, okay. I will keep finding if there's other questions coming in. If you would like to come off mute to ask a question, feel free to raise your hand. Oh, we got a request to stop sharing the screen, John if we could for now.
While we ask you questions, let's do that.
Done with a thank you.
I've seen enough thank yous in a hundred different languages.
Okay, here's another question or clarification. John, did you say there is 30 million incremental fundraising beyond the 20 million expected for a total of 50 million? Is that correct?
Yeah, so it's 30 million above the control group. That's correct.
And in this, this is pre-COVID now. So this experiment would have been wrong around 2012 to 2015. Now in COVID what we found… So we have now data from the USA and Germany, and kind of three general trends that we've been finding, maybe four general trends. One, fewer people are giving. Two, the people who are giving are giving less. Three, they're giving two different types of causes. And what I mean by that is it tends to be more tilted than in the past toward health issues, and which kind of makes sense in the short-run. And four, people are much more price sensitive. That's the exact same people we have data on. In pre-COVID times and post-COVID times, that same person's more price sensitive now. So those are sort of four headlines that I can give you in pre versus post-COVID.
So, I do think, if President Biden is sympathetic to our sector and the correct, let's say that the correct measures are put in place, I think that's the best chance for the sector to stay vibrant and potentially a breakout of the 2.2% of income that we have found in the last several decades.
Great, there's another question that was asked to fit earlier on, it says how important is maintaining status to the UHNWI or to put it another way, can philanthropic giving be considered a luxury good?
So if you define luxury good the way that economists do, so how economists think about luxury goods, for example, is if your income goes up by 10%, does your amount of giving go up by more than 10%? So under that definition for the deep right tail people, giving is a luxury good, for the modal donor, giving is not a luxury good. Giving would be a normal good. What a normal good means, for example, if your income goes up 10%, you're giving might go up 2%. That's the modal donors that act, so to speak in the data. Now, I thought where you were going was along the lines of, how well does binning work?
So for example, gold givers give $500 or more, platinum give 250 to 250, silver give 100 to 250. We developed a model that we can actually take your distribution of givers that you have right now. Like, let's say you have a distribution of givers but you don't have these bins or these groups. We have a model whereby we can tell you how many bins you should have and where to place those bins. Because the idea is, you don't wanna place them too high because people will give up, say, I'm not gonna reach it. You don't wanna place it too low because then people say, well, a $25 gift is fine for me. What you wanna do is place it say at 100 to where people who would've given 60 to 95, move up to 100. And you know what, too many people who would have given 125 to move down to 100. So that all depends on your distribution of what people give without these bins. So we do have an academic paper on that should you want one, and I call it the optimal bin number in the optimal bin placement in your donor base. No good fill, so just reach out to me. I appreciate that. That you’re interested in it.
Great, we've got time for a few more questions I think. One came in that says, do you think women are more caring than men? So give out more but men only care about giving if it offsets tasks, which maybe that'd be a little hard to say but we'll throw it out there.
Yeah, hard to say but that's what our data say. Yeah, it's not nice to say it, I guess I can say it more easily because I'm a man. No, absolutely. Women are more altruistic, check. Women give more because of altruism. They do give more because of social pressure too. but there is a reason why that slide to me is so fascinating because in every income bin, so if you just set a man in a woman's incomes equal, women will out-give men on average every time until you get out to the 99.99th percentile. That's fact. Okay, now you say, well, why do they give in that bin? Our data suggests men are much more price sensitive than women, check. Our data suggests women are a lot more altruistic than men, check. At science.
Great, Thank you. I see another question that just came in saying--
Can I follow up on that Prentiss, 'cause I think in the end
Yeah, please go, yes just do.
You might say well, John that's a great fact. And now I can use it at the local cocktail party and people might be bemused by it, but how can I actually put it in action? Here's how we put in an action in Alaska. So Alaska has an annual giving fund where they ask each of their citizens to give back part of their oil royalty checks. And we tested the warm-glow model versus the altruism model.
So what do I mean by that? The altruism model is, give today so you help our citizens in Alaska. The warm-glow model is, give today because you're gonna feel better about yourself. Guess what works better? The second one. Guess what works better for women? The first one, because that's tweaking altruism. That's a simple nudge that we raised and I can send you the paper, it's coming out in nature. It'll be coming out next month, probably in nature. So that's a simple fact that you can leverage now with targets because you know the underlying motivation for why someone gives and the differences between men and women. Okay, I'm sorry Prentiss, go ahead. Because...
No, John, that's great.
We gets caught up in these facts, our list is just spinning vignettes that are great for the cocktail party, but I don't know how to take them from here and actually apply them. Everything that I've talked about today should be applied to make our donor pipe bigger. You might not readily see how to apply it. Ask me for the papers we say in the papers and we show how you expand and keep people committed, okay?
So I'm sorry, Prentiss, go ahead.
No, I think we're coming up on time here. And so, unfortunately we're not gonna be able to get to all the other questions. But John, thank you so much. I think that's a great takeaway and hopefully folks will, you know, use this in the real life situations.
The session recording will be available on the Cvent Attendee Hub, We will also be sure to add the slides and the articles that John has alluded to throughout the entire session. So thank you again, if you will be joining us for our session at 12:30, you have a few minutes before, but you can join that through the Cvent Attendee Hub. And if you have a chance, please fill out the survey. We really appreciate your feedback, and we hope you enjoyed today's session.
And again, big thank you to John, it was a pleasure hearing from you.
Thanks so much Prentiss and thanks for having me and I look forward to continuing the dialogue and the partnership.
Thanks everyone. And thanks for trying to change the world for the better. I appreciate that. I really do so. Thanks everyone, have a great rest of the day and have a super weekend. Bye-bye.
It uses speech-to-text machine learning algorithms to transcribe video footage evidence, speeding up pre-trial preparation and helping public defenders better defend clients.JusticeText Helps Public Defenders Better Serve Their Clients
Dean Madhav Rajan speaks with Walker about innovation in the philanthropic community, his journey into social justice leadership, and other topics.Distinguished Speaker Series with Darren Walker, President of the Ford Foundation