Deborah Small on Charitable Giving

Is giving to a charitable cause essentially equivalent to any other economic decision made by a human being, bounded by the same rational and irrational inputs as any other expenditure? Based on research by psychologist Deborah Small and others working in the area of philanthropy and altruism, the answer is a resounding no.

In this Social Science Bites podcast, Small, the Adrian C. Israel Professor of Marketing at Yale University, details some of the thought processes and outcomes that research provides about charitable giving. For example, she tells interviewer David Edmonds, that putting a face to the need – such as a specific hungry child or struggling parent – tends to be more successful at producing giving than does a statistic revealing that tens of thousands of children or mothers are similarly suffering.

This “identifiable victim effect,” as the phenomenon is dubbed, means that benefits of charity may be inequitably distributed and thus do less to provide succor than intended. “[T]he kind of paradox here,” Small explains, “is that we end up in many cases concentrating resources on one person or on certain causes that happened to be well represented by a single identifiable victim, when we could ultimately do a lot more good, or save a lot more lives, help a lot more people, if – psychologically — we were more motivated to care for ‘statistical’ victims.”

That particular effect is one of several Small discusses in the conversation. Another is the “drop in the bucket effect,” in which the magnitude of a problem makes individuals throw up their arms and not contribute rather than do even a small part toward remedying it.

Another phenomenon is the “braggarts dilemma,” in which giving is perceived as a good thing, but the person who notes their giving is seen as less admirable than the person whose gift is made without fanfare. And yet, the fact that someone goes public about their good deed can influence others to join in.

“[O]ne of the big lessons in marketing,” Small details, “is that word of mouth is really powerful. So, it’s much more effective if I tell you about a product that I really like than if the company tells you about the product, right? You trust me; I’m like you. And that’s a very effective form of persuasion, and it works for charities, too.”

Small joined the Yale School of Management in 2022, moving from the University of Pennsylvania’s Wharton School, where she had been the Laura and John J. Pomerantz Professor of Marketing since 2015. In 2018, she was a fellow of the American Psychological Society and a Marketing Science Institute Scholar.

To download an MP3 of this podcast, right-click HERE and save. The transcript of the conversation appears below.

For a complete listing of past Social Science Bites podcasts, click HERE. You can follow Bites on Twitter @socialscibites and David Edmonds @DavidEdmonds100.

TRANSCRIPT

David Edmonds: Profit-focused companies and charities have entirely different objectives – but certain things in common. Companies want to persuade customers to purchase their products. Charities want to persuade potential donors to give them money for the good causes they fund. Deborah Small is a professor of marketing at the Yale School of Management and has studied what motivates people to be altruistic. Deborah Small, welcome to Social Science Bites.

Deborah Small: Thanks so much for having me.

Edmonds: We’re talking today about charitable giving. Why do people give to charity? What works in persuading them to open their wallets? Now, I don’t normally ask interviewees why they’ve ended up in a particular discipline. But this is such a specific area, so I’m interested in what first attracted you to philanthropy as a research topic.

Small: A little bit of happenstance, I didn’t intend to study philanthropy, I was getting my PhD in judgment and decision-making. And so I was interested in biases in the way we make all sorts of judgments. And I was interested in public policy. And the first area of bias that kind of drew me in was bias in the value of human life. For a long time, I didn’t even think of myself as someone who studied philanthropy, and then that kind of brand was thrust upon me, I really thought of myself as studying questions around distortions and altruism and self-interest and how that affects our decision making. But, of course, the real-world analog of that is philanthropy. And it’s very practically important and useful for taking research into practice for teaching and basically being in the real world.

Edmonds: Let’s talk a bit about some of the research you’ve conducted, starting with what’s called the ‘identifiable victim effect.’ Before you tell us the findings of this research, you’d better define what the identifiable victim effect is.

Small: That term came from Thomas Schelling, who wrote about this, I think, very relatable tendency, we tend to focus our attention, our sympathy, and behavioral responses, oftentimes giving money when we learn about a specific identified victim. And yet, there’s evidently thousands, millions, of needy people around the world, who remain hidden to us, Schelling refers to them as statistical victims or statistical lives, that don’t move us in the same way.

And the kind of paradox here is that we end up in many cases concentrating resources on one person or on to certain causes that happened to be well represented by a single identifiable victim, when we could ultimately do a lot more good, or save a lot more lives, help a lot more people, if – psychologically — we were more motivated to care for statistical victims.

Edmonds: So that’s the definition of it.

Small: Yes.

Edmonds: What did you do to test it?

Smalls: We and other researchers have tested it in a variety of different ways. And in hindsight, now, sometimes I think this gets a bit confusing, because sometimes it’s operationalized by contrasting helping towards one victim versus helping towards some larger set of victims. And that’s certainly part of it.

Other times, it’s tested, including some my own work, by showing a picture of a particular victim in one case, versus statistics. In another case — perhaps the cleanest lab-based experimental test of this, which is also the most experimentally cleanest, which means it’s the most unnatural and most different from the real world way, is in a lab-based experiment — we have participants playing a version of the Dictator Game, where some participants are randomly assigned to get some sum of money, and then they’re matched with another anonymous person, they don’t meet this person, are never going to meet this person, they’re matched with someone else. And they’re given an option to give away any of the money for which they’ve just been endowed to this other anonymous person. And the way that we sought out to kind of cleanly test for identifiability was we enabled in one condition, so for some randomly determined set of participants, they drew the number of the person for whom they were matched before they made their decision. And then the other condition, the others flip a coin, those who flip tails instead of heads, they don’t learn the number of the person they’re matched with until after they’ve already made their decision of how much to give.

Now, in both of these cases, the number is meaningless because they know that that number can be any of the other people, they’re never going to find out who it is or anything like that. Nonetheless, when say imagine you draw number six, right before you made your decision, that person has now been determined. It is a specific certain person. It’s an anonymous person, but it’s a certain determined person.

And just that subtle manipulation affects how much people give in this Dictator Game. We used to call this determinateness. We think it’s kind of the most stripped-down part of the identifiable victim effect. You know, in the real world, being identifiable often means you learn information about a person and you see their picture, and maybe they’re attractive, maybe they’re unattractive. Maybe you relate to some of the content in the appeal, or you don’t. It’s very difficult to tease out whether any effect is driven by one of those characteristics or by identifiability itself. And so this was an aim to subtract out all that other information. And it seems to matter.

Edmonds: And what’s amazing about that is how little it takes to trigger this effect, you just have a number which you know, is linked to a determinate individual, you don’t know a name, you haven’t seen a face, you don’t know where they live. And the number alone is enough to increase the amount of donations that person receives.

Small: Exactly.

Edmonds: And in the real world, we have much more than that. So, in the real world, it means that when we see an image of somebody on a leaflet, we’re much more likely to support that charity than if we’re presented with bold statistics, about thousands of people are affected by this particular disease, or illness, or whatever it happens to be.

Small: Exactly. In fact, we’re recording in mid-October, Israel was recently attacked by Hamas, and I just a few minutes ago, walked by a number of signs, with images of children who have been kidnapped, and then hostages of Hamas. So, leveraging this exact psychology to try to raise support for their side.

Edmonds: This seems to be a finding in a way that academics have uncovered but the charities have always understood implicitly. And that’s why they do exactly that they put a picture of a single child on their leaflets.

Smalls: Totally, yeah. They knew it before us, and not just them. I think journalists, maybe you can relate to this. journalists understand this, too. So oftentimes, even in a pretty dense, heavy news article about some misfortune, that does eventually get into lots of statistics and details. They don’t start with that, they don’t lead with that. They lead with a story of a particular person.

Edmonds: We’ve heard about a replication crisis in psychology over the past decade or so, often, these studies are based on small samples. Are you confident of the result?

Smalls: I’m confident of that key effect. We haven’t even gotten into all of the many, many, many publications about this now. And many of them show specific moderator effects you get it if it’s one versus many, in-group, out-group, all these kinds of things. I think I’m less confident about all of those. But you mentioned that this is based on lab data. There is pretty good field-based evidence field experiments, and also just observational field data.

Again, as I mentioned before, there’s a lot of different features of the identifiable that can affect but certainly it’s consistent with a preference to help smaller numbers, like one or two as opposed to larger numbers.

Edmonds: And do you think that part of the reason that most of us give away so little to good causes you would talked about the numbers is that the problem just seems too vast in so many cases that there were just too many millions of people who need help, and it’s overwhelming for us.

Small: I think that’s part of it. What you described is sometimes known as “the drop in the bucket effect” where we feel like we can’t really have that much impact when the problem is so vast. Paul Slovic has this nice quote, when talking about what we might do to deal with those vast large problems. He says, “We need to put tears on the statistics.” That could be by using an identifiable victim, as a poster child, it could be other ways of humanizing large numbers or making them more emotional in some way.

Edmonds: Right. Another point about the identifiable victim effect is that if we’re looking at the impact of poverty, and we hear about one person dying from malnutrition, and then there’s another report, which talks about two people dying of malnutrition, well, that seems obviously twice as bad to us. But if I hear about an earthquake on the other side of the world, I can’t really grasp the distinction between 17,100 people dying, or 18,300 people dying.

Small: Yes, we don’t perceive numbers linearly. So big numbers, there’s, you know, diminishing marginal utility. So first of all, there’s I think numeracy challenges with just even comprehending large numbers, but also the difference between how one and two feels. We perceive it as much larger than the difference between you know, 9,000,871 and 9,000,872. We barely perceive any difference between those two numbers.

Edmonds: The single identifiable victim effect — what are the lessons for charities in all this is it to concentrate on narrative basically.

Small: I think that’s the main takeaway. I think that’s one aspect of what they need to do, which is to make it more emotional, tug at our heartstrings, make it more personal for people. And so when you see a person, you can almost imagine having a relationship with them, right? I think relationships really matter for charitable giving to.

I think another related set of findings looks at how personal experiences affect charitable giving. For instance, if you had somebody that you loved and lost, like, say someone in your family who died of, say, lung cancer, you’re going to be much more likely to support lung cancer causes in the future than someone who hasn’t had that personal experience. And so, this is separate from the identifiable victim effect. But it’s something that people’s attention tends to narrow in on things that they can relate to things that tug at their heartstrings. It’s another approach.

Edmonds: Most people, at least in the developed world give some percentage of their money to charity. How do we perceive generosity? How do we perceive our own generosity? And how do we perceive the generosity of others?

Smalls: That’s a great question. I think that people tend to rely a lot on what I call motive inferences. We observe others behavior, and we learn about them through their behavior. So if I see you donate to charity, or help a homeless person, or help in some way, that action alone affects my impression of you. But it’s not the only thing that affects my impression of you. I tend to then question, ‘Well, what was David’s motives there? Did he actually care? Or did he only act that way, because he knew I was watching, or because he expected some reciprocity or some benefit from that action?’ And so people tend to rely heavily on motives in general and assessing our behavior and others’ behavior. But I think it’s particularly important aspect of human judgment, social judgment, when it comes to altruism, because there’s such a strong expectation that altruism needs to be pure, free of or devoid of any sort of self-interest. And so, people get pretty irritated, pretty cynical, pretty upset by what they perceive as insincere, inauthentic forms of altruism.

Edmonds: But it’s interesting, isn’t it? Because I don’t know whether giving to a very rich institution like Yale counts as altruism. But if you go to Yale or Princeton, and you hear about people who’ve donated. Well, the reason you hear about them is their names are plastered on the corridors, or the libraries, or the buildings in some way, that doesn’t look like a very pure form of giving.

Small: Totally Yes, I’m fascinated by this. I actually live across the street from a museum that built an addition about five years ago, and we waited for months for all this construction to end. And then the grand finale. Finally, we weren’t hearing the noise of construction workers outside anymore. And but instead, they slapped on the new edition the name of the donor. So now I see it on my window.

And this is ubiquitous as university campuses and hospitals. We call this the “Braggarts Dilemma.” The braggarts dilemma is the idea that it helps to get the word out about our generous actions, and helps our reputation. I can be the most generous person in the world. But if no one finds out about it, then I don’t get any credit at all. And so, I think that motivates people to want their names on buildings.

Also, people’s sometimes it’s known as virtue signaling, but posting online about actions they do that might be perceived as generous by others. So, there’s this motive to get the word out. But at the same time, when we get the word out, we risk appearing as insincere as braggarts. And so that’s the dilemma.

Edmonds: Right? So, the tension for a charity, like Oxfam, or Save the Children, is that they want you to boast about your giving, because that encourages others to give. But you yourself are reluctant to do that, because that reflects very badly on you.

Small: Totally, yeah. So I teach marketing courses. And you know, one of the big lessons in marketing is that word of mouth is really powerful. So, it’s much more effective if I tell you about a product that I really like than if the company tells you about the product, right? You trust me; I’m like you. And that’s a very effective form of persuasion, and it works for charities, too.

But as you described, donors are often reluctant to talk about their charitable giving, because they want to seem to others and perhaps also want to feel themselves like they are being generous for purely altruistic reasons, rather than to get social credit.

We actually ran a field experiment with an organization that was struggling to get donors to share talk about their charity. So with this organization, when you donate online, there’s a pop-up window that appears thanking you for donating. And then it there’s some links to Twitter and other social media and email. And they say thank you for donating, please share about this cause online and very few people do. And so we ran a field experiment with them, where we change the wording of that message. So rather than just thanking them and asking them to share about it online, we pointed out to donors, that sharing is a way to do more good and magnify your impact. And it worked. I mean, it didn’t get everybody to share it, sharing rates are still pretty low, but it raised sharing rates by about a percent.

Edmonds: So, it raised a number of people who shared but I guess the crucial question is, did it increase the number of donations? Did it?

Smalls: Yeah. So, what was cool about this data, because this was all digital, was that when you share through this pop-up window, we can observe what people post on their own private social media. But there’s a unique traceable link that gets shared. And so then we could trace any subsequent donations back to the person that shared and we know what message they saw whether they saw the standard message, or the one encouraging them to share as a way to have more impact. And so, we ran our experiment just for a few weeks, but kind of a back-of-the-envelope calculation of the impact of this message was that it ought to raise about $170,000 more per year just by changing that word.

So, it’s not like millions of dollars or anything like that. Most people are still quite reluctant to share about charity online. Even after doing this research, I still cringe and cannot bring myself to post on social media about my donations. Even though I advocate for it, I think it’s a good thing. But maybe I’m not quite generous enough to sacrifice my reputation in that way.

Edmonds: Some of these phenomena, we’ve been talking about the identifiable victim effect, and so on, do you regard them as biases, irrational responses that just are difficult to overcome because they’re such a deep part of our psychological makeup?

Smalls: I’m so hesitant to use irrationality language. So, what we were just discussing, for instance, I think there is some good reason to care about motives, I don’t think it’s completely wrong to care about motives in assessing other people’s altruism, and perhaps even for ourselves, this we live in a social world, we have to interact with people repeatedly. And so we want to try to figure out who’s truly good. Motives are pretty good signal. Like I want to know not just whether you behave generously, whether you behave generously consistently, absent other incentives to do so. Right. And so, it’s a useful social process.

What fascinates me is when that comes into conflict, or when there’s a tension between that and doing the most good in the world. And so, if we take social utility maximization as a normative model — not everybody agrees with that, of course — but if we say that that’s the definition of have rationality than certainly the identifiable victim effect, but also our attention to motives in say, decisions of whether to share about charity seem to be a bias, right? They’re limiting the amount of good we can do in the world because of our psychology.

Edmonds: Right. So where does that leave you, as an academic who’s studied philanthropy, you say you’re reluctant to boast about your charitable giving. But are you able to put most of these other effects aside and think, you know what, I’m just going to donate to the charity that does the greatest good for the greatest number.

Smalls: I do do that. I tend to read the research provided by effective altruists and make my decisions of where to give based on those recommendations, which are, as you know, designed to maximize impact. So that part I think I’m semi-rational about.

I still sometimes give to other causes, too. But my way of rationalizing that is, in my mental budgeting, I treat that as personal consumption. So, if for instance, I know someone who’s passed away and their family asked for donations to their favorite charity, I don’t mentally encode that as my charitable giving, that’s more my personal consumption or being a loyal friend or something like that.

Edmonds: You’ve persuaded yourself that that’s a reasonable approach because after all, as well as your obligations to strangers at the far end of the world, you’ve also got obligations to your friend who’s got a special cause.

Small: Yeah, I think that’s a good way of saying it, it would seem less rational if I took from my charitable giving budget. So, if I gave less to effective causes, but if I can do both the opportunity cost of me giving to my friends, charity is less lattes or vacation or more work or something like that. I’m OK with that.

Edmonds: Deborah Small, thank you very much indeed.