Wharton’s Katherine Milkman and Judd Kessler discuss their new research on charitable giving.
Charitable organizations and nonprofit institutions that rely on donations to fund their activities have to strike a delicate balance between aggressively soliciting money and not turning off donors. New research from Wharton shows that timing is the key to maximizing donations, particularly from people with an existing connection to the organization.
Katherine Milkman, professor of operations, information and decisions, and Judd Kessler, professor of business economics and public policy, joined the Knowledge@Wharton show, which airs on SiriusXM channel 111, to discuss their study, which focused on donations to a hospital system. While the findings have practical implications for nonprofits, the study also offers some interesting insights into reciprocity and human nature. Their paper, “A Field Study of Charitable Giving Reveals that Reciprocity Decays over Time,” was recently published in the Proceedings of the National Academy of Sciences (PNAS). The paper was co-authored with Wharton Ph.D. student Amanda Chuan.
Listen to the full podcast using the player at the top of the page. An edited transcript of the conversation follows.
Knowledge@Wharton: Let’s dig into the idea behind the research. Charitable giving is such an important component to many organizations, but they have to get the timing right. That’s the key theme of this paper, correct?
Judd Kessler: Yes, that’s exactly right. A lot of charities will send solicitations when they think people will be generous or in the mood to give. They might send around Christmas or leading up to Giving Tuesday. But we’re looking at a nonprofit that solicits funds but also provides a service to the people they solicit from. It gets a little bit more interesting because if I’ve done something for you, then ask you to make a donation, I want to think not just about when might you be in a generous mood, but also when might you be reciprocal towards me given the fact that I’ve done something that’s helped you in the recent past.
Knowledge@Wharton: Your research was with a hospital system. Is that different compared with charitable organizations such as the American Red Cross or the American Heart Association?
Katherine Milkman: We believe our findings are likely quite generalizable, not just to other hospital systems but to all organizations that are providing services and asking for donations in return. You could think about a church or religious organization that’s asking for people to give. You could think of a school that asks alumni for donations after providing a service in the form of education. You could think of the humane society.
There’s a very large range of organizations that provide services and then seek donations from the people they’ve provided those services to. The fundamental question we’re asking should really apply to all of them, which is: Can you ask too early? If you ask too quickly, will it perhaps offend people that you are reaching out and asking, “Hey, we just did a nice thing for you. Can you make a donation?” If you do that really fast, maybe people will be put off. Maybe you should wait. On the other hand, you could imagine that if you wait too long and don’t strike while the iron is hot, you miss your opportunity. This paper addresses that question as well as telling us something more fundamental about reciprocity.
Kessler: You might also think about how this applies in interpersonal interactions. When somebody asks you for a favor, you do it partially out of the goodness of your heart — maybe also because you think when you need a favor from them, they’ll be ready to reciprocate. Maybe you might be worried about asking too soon, but also potentially about waiting too long. That warm feeling of you having helped somebody out — maybe they don’t feel that as much anymore.
Knowledge@Wharton: Were you able to gauge the success rate of donations based on the timing of the solicitations?
Milkman: That’s exactly what this paper looks at. It uses a quirk of the institutional setting we were studying, which is that mailings asking former patients to make a donation to the hospital system go out in batches that are scheduled well in advance. If you come into the hospital during a certain period of time, your mailing is going to be queued up shortly thereafter. If you miss a deadline — say, you come in on the last day of the month — you get into the next batch that’s going out the next day. If you come the first day of a new month, you’re going to get an ask many months later. There’s this quirk, which is that the mailings all get queued up and sent in one big batch, and that gives us an opportunity to see what happens if you get an immediate mailing and what happens if you have a really long delay.
“Life is all about give and take, and this emphasizes how critical it is to recognize that there is an expiration date on reciprocal behaviors.”–Katherine Milkman
What we see is a very clear, steep decline in generosity towards the hospital the longer you wait. Specifically, donation rates drop about 30% every extra 30 days the hospital system waits to contact a former patient.
Kessler: It takes a little while to figure out exactly who visited and who you’re going to mail to. Then there are quirks where they try to time the solicitations nearer to the holidays, so that can create even longer delays from the time of your visit to when you get the solicitation. They’re doing OK in terms of being within a month or two of your visit. But if they could figure out how to get everybody who is receiving their mail late and after a longer delay, sort of pushing that up a month, then they’d increase the rates of donation to the hospital system quite dramatically.
Knowledge@Wharton: What about the perception of the potential donor? Someone who receives a letter after 30 days may feel like they’re being pressured into giving.
Milkman: We thought maybe you would see that kind of backlash, but we really don’t find any evidence whatsoever in our data. We look at some different sub-populations to see if there are different patterns, and we never see backlash. We do see some groups that respond even more dramatically to time delays, meaning the decay is faster in their willingness to reciprocate. It goes down faster over time.
For instance, we see the steepest decay among patients who visited some of the departments that deal with the most severe cases, like surgery and oncology. We think that’s interesting because one concern about our findings is maybe you’re just forgetting about the interaction you had with the hospital. But it’s not about reciprocity decaying particularly, which we think is really fundamental about human nature. Maybe they forgot that they even had this interaction. But there are two things that counter that. One, the mailing literally reminds you that you were a patient of this hospital. And two, you’re definitely not going to forget when you went and had surgery. There’s just no realistic way you’re forgetting, and that’s actually where we’re seeing the strongest effects. So, that makes us pretty convinced that this is more about gratitude and less about forgetting.
Knowledge@Wharton: How much variation is there from department to department within the hospital system?
Kessler: Giving rates do start out higher at the departments that treat the most severe illnesses. If you go in for something minor, giving rates are quite low. But the fact that giving is happening in response to visits to these severe cases, that leads the hospital system and organizations like that to maybe want to think, “Who are the people that might feel most generous to us? Let’s target them.”
We were particularly worried about this question you raised earlier of what if it’s too soon? So, we talked about the three-week delay between the mailings. A nice feature of the hospital system is that they classify you into these buckets based on the first visit. But many people visit multiple times before the solicitation. That last visit could come days before you get the solicitation. We do some empirical work to recognize that people who visit multiple times are going to be different from people who visit once. Even when it’s a couple of days before your [last] visit took place, you’re still giving at much higher rates, and we still see the sharp decline.
“What we see is a very clear, steep decline in generosity towards the hospital the longer you wait.”–Katherine Milkman
Knowledge@Wharton: Should hospitals consider setting up their solicitation system to be on the last visit, timing out 30 days or 60 days from that point? It seems like some hospitals do it from the front end.
Kessler: Hospital systems have many other ways that they fund themselves. But for other organizations, solicitation of donations from people that they’re serving might be the main source of revenue. Religious organizations, for example, might want to be particularly interested in whether their systems could be improved to better timed solicitations, relative to when people come in.
We do a little calculation in the paper where we say, “There are other strategies you might want to use to increase donation rates.” We benchmark our effect to others that have been tried in the literature, and we find getting solicitations a month sooner is about equivalent to offering a one-to-one match in a solicitation mailing, which has also been shown to increase giving rates.
Knowledge@Wharton: Are hospitals and charitable organizations starting to figure out the quantitative loss because of timing?
Milkman: Yes, and that’s exactly the exercise we go through in this paper, in addition to doing a bunch of analyses to try to isolate the fact that this seems like a gratitude effect rather than something else. We can point out that this is really a large reduction — about 30% — in the likelihood of a donation for every 30 days they wait between the time of your visit and the time of their solicitation.
Knowledge@Wharton: Does that delay also affect the amount that somebody will donate?
Kessler: It doesn’t affect the amount once a gift has been made. But given the fact that you’re missing out on all these donations, it does hit the bottom line pretty hard. When you wait, it’s fewer donations — which translates pretty directly into fewer donations.
Knowledge@Wharton: Was there a difference between male and female donors?
Kessler: We didn’t find any differences based on gender or other demographics. In the statistical approach we take, we control for those things. Men and woman may donate at different rates at baseline, but the only patterns that we saw are the ones that Katy already mentioned, that the more severe departments were the ones that showed the sharpest rate in decline.
Consistent with that, we also saw that if you visit the hospital system more times, then you demonstrate a directionally larger decay rate. We think that that has to do with having stronger feelings of reciprocity towards the hospital system. I’ve had a good enough experience that I’ve bothered to come back, and that means I might be more inclined to be generous back to the hospital system when they ask or funds.
Knowledge@Wharton: Beyond the numbers, what’s the other big takeaway here?
“We find getting solicitations a month sooner is about equivalent to offering a one-to-one match in a solicitation mailing.”–Judd Kessler
Milkman: I think this is a really important finding for organizations that survive based on this kind of reciprocal donation. I actually think it’s an important finding for humans in general. One of the reasons Judd and I are so excited about this paper is that it’s not just about hospital system and reciprocity. What we think we’re able to isolate here is something much more fundamental. Life is all about give and take, and this emphasizes how critical it is to recognize that there is an expiration date on reciprocal behaviors, which is something that was just completely missing from both psychology and economics previously. It’s really important because in much of life we are doing favors for someone. You ask a friend to do something for you at work and hope that they’ll help you in return. If you don’t recognize that you need to ask for that return favor quickly, you may really miss out.
Knowledge@Wharton: Why is there a lack of understanding of this concept to begin with?
Kessler: I think people recognize that you might forget about an interaction. That’s a fair thing. I do a favor for you, and maybe you don’t remember when I turn around and ask you for one. But there are ways of getting around that. When you ask me for a favor via email and I do it, I can reply to that and say, “Hey, now I need your help.” People understood that.
What we’re finding is that, even though we think you’re likely to remember your visit, even though the solicitation request has in it the reference to the fact that you came back, we’re still seeing people are less inclined to reciprocate. We think that that means there’s something else going on, and it’s subtler. It’s probably that I feel a certain way when you do me a favor, where I feel indebted to you, I feel grateful to you, I feel these positive emotions. If you were to ask me for a favor while I’m feeling those very strongly, I would be happy to. I’d enjoy it, or maybe I’d feel really guilty if I didn’t. We can’t separate those apart. But I’m at the peak of my emotions regarding our interactions.
Even though you might continue to remember the event, if I wait too long to ask you for that return favor and those feelings have abated — feeling good about you but not as elated — then it’s going to be less likely that the reciprocation will take place.
Milkman: There’s a lot of research showing that we experience much of life in different visceral states. We can have hot states of anger or hunger, and they are passing. They’re fleeting. One explanation for our findings is maybe that gratitude is like that. You feel this intense wave of gratitude when someone does something important or special that’s meaningful to you. When I think about your feelings, you don’t recognize that that’s fleeting in the way that you experience it. So, there may be a disconnect between the fleetingness of these kinds of states of gratitude and our expectations.
Knowledge@Wharton: In this digital age, does the type of contact matter?
Kessler: In this paper, we’re looking specifically at mailed solicitations, but your intuition is exactly right. If you look across studies, an email is very easy to ignore, and it’s going to be ignored more often. If it’s in the spam folder, it’s definitely ignored by default. But a physical piece of mail maybe gets more attention. A phone call even more so, if you actually pick up. That is going to be underlying a lot of the way organizations use this kind of insight from our work. It’s not just about striking while the iron is hot, so to speak, but you have to get people’s attention for them to be inclined to engage reciprocally or not.
“It’s not just about striking while the iron is hot, but you have to get people’s attention for them to be inclined to engage reciprocally or not.”–Judd Kessler
Knowledge@Wharton: What’s next for this research?
Kessler: Any time you have a result that surprises you, you have a ton of follow-up questions. This is a situation where we’re seeing this decay. Your intuition might be that many charities, like the American Red Cross, assist somebody during really hard times. If I suffer a fire and need help getting ready to find a new place to live, that’s not going to be a setting where I’m going to be flush with cash to be able to make a donation. I’d be interested in exploring how our findings interact with situations where I’m down on my luck. Maybe that’s when I feel particular reciprocity towards an organization. If you help me when I’m struggling, I might really care about returning that favor. But again, I might not be able to at that moment in time. Unpacking that kind of question is something that seems exciting to me.
Milkman: I think really interesting questions for follow-up would be what other aspects of timing are relevant to decisions about reciprocity and giving? Not just the time delay after a service interaction, but maybe there are certain moments besides the holidays when we’re particularly motivated to give. I’ve done some other research on what I call the “fresh start effect moments,” when people are particularly motivated to achieve new goals in their lives or turn over a new leaf. There may be similar moments when we’re feeling like we want to give back to the world. If organizations could figure that out, it would have a big impact.
Kessler: Yes, and it might vary organization to organization. Religious holidays might be that point in time for religious organizations. There’s evidence from charities that deal with natural disasters that it doesn’t matter where the natural disaster is. You are more likely, for example, to donate blood locally when you hear about some tragedy in some other distant part of the world, even though the blood you donate is not going to the people who are suffering from that particular event. I think Katy is exactly right. These temporal spillovers or moments where you’re particularly keen to give back — identifying what those are could be particularly interesting.
Article by Knowledge@Wharton