Spreading the Wealth: The Preference to Distribute Donations

Manik, Tushar, Protoy, and Simanto have all been waiting about a year for a sponsor. Their pictures sit side by side on the homepage of Compassion International. For each of these 8- or 9-year-old boys, sponsorship means access to school, medical care, and better nutrition. Knowing their situations, you wish you could sponsor each of them, but your resources only go so far. In instances like this, would you prefer to donate as much as possible to a single child or donate a little to each of the children, trying to reach all four?

Most online giving platforms, from Kiva to Kickstarter to Compassion International, feature multiple profiles of people or groups requesting aid. Each requester’s profile may appear similarly needy and deserving, depicting compelling histories and reasons for their donation request.

As donors, we therefore face a trade-off in trying to help effectively. We can try to maximize “depth” by donating all of our money to a single requester (concentrating our donation). Or we can maximize “breadth” by spreading money across multiple requesters (distributing our donation).

As donors, we face a trade-off in trying to help effectively. We can try to maximize “depth,” or we can try to maximize “breadth.”

There are research-based reasons to think that either option could be appealing. We know, for example, that seeing a single victim tends to evoke more compassion and sympathy than seeing many victims. And people often define impact in terms of changing a single life. So maybe you’d be tempted to just pick one requester for your donation.

But donating to just one requester has a downside that makes this strategy particularly unappealing: it feels unfair to the other requesters. Our research, recently published in the Journal of Personality and Social Psychology, shows that the preference to be fair—even when nobody is watching—leads people to choose to distribute their money across multiple requesters, instead of concentrating it to a single requester. In other words, they prefer breadth over depth in their donations.

Demonstrating this, in one study we gave 294 people the option to donate up to $1 (out of their bonus payment for completing a survey) to groups of real requesters that we randomly selected from FundRazr.com. Participants kept any money they did not donate. Of those who donated (54.8 percent), 62.7 percent chose to distribute their donations across all of the possible requesters, so that each requester received something.

Why did they prefer to distribute their money? After making their decision, participants rated how fair, impactful, and efficient it would be to distribute their donation. Perceived fairness was the biggest predictor of their actual decision to distribute or not, suggesting it is an especially important psychological motive in the decision to distribute.

But what if one requester seems obviously needier than the other requesters? We tested this in another experiment by picking four real requesters from Kiva.org and manipulating the amount of money they had already earned toward the $600 they each requested. In one condition, all of the requesters had already earned the same amount ($100). In a second condition, three of the requesters had earned $400, but one had only earned $100, making the low earner seem needier than the rest. We then randomly assigned over 200 participants to one condition or the other, and asked them how they would distribute money among the requesters.

In the first condition, where all requesters earned the same amount, 86 percent of participants said they would completely distribute their donation. In the second condition, participants tended to give more to the low earner; here, only 58 percent of participants said they would completely distribute their donation.

After making donation decisions, participants rated how fair, impactful, and efficient different donation strategies would have been. They also rated how much requesters would appreciate different amounts of money and how badly they would feel about donating more or less to different requesters. Only one of these ratings mediated the decision to distribute one’s money more in the equal-earner versus low-earner condition: how fair it seemed. This suggests that fairness concerns are again driving the decision to distribute donations.

If people feel compelled to donate to each individual requester who they view, then having to make decisions about how to help multiple requesters might ultimately lead them to end up donating more.

This experiment further tests the impact of two different forms of fairness on donation decisions: fairness in the process of donating or fairness in the outcome of the donations. If donors are concerned with the process more than the outcome, they will tend to divide their money equally to each requester even if one requester ends up with more money. If they are more concerned with the outcome than the process, they will divide their money unequally to ensure the final amounts are equivalent. The data suggest that fairness in the process is more valued: 18 percent donated using equal distributions, but only 4 percent donated to create equal outcomes.  

In total, we conduced six experiments with 2,380 participants making real donation decisions and measured their strategies for donating to multiple requesters. In all experiments, we typically provided potential donors with a small bonus that they could choose to donate, and they viewed real requesters (from funding platforms like Kiva.org, GoFundMe, FundRazr.com, and Compassion.com) to whom they actually donated. Of those who donated, 78 percent chose to distribute their money to more than one requester. Of these, 77 percent donated to every single possible requester (“complete distribution”) and 51 percent chose to equally distribute to each requester (“equal distribution”).

So why does it matter that people prefer to distribute more than concentrate their donations?

For one, because the act of distributing money can increase the total donation amount. We reasoned that if people feel compelled to donate to each individual requester who they view, then having to make decisions about how to help multiple requesters might ultimately lead them to end up donating more. In this way, donation decisions that are “unpacked” into constituent requesters may lead to more donations than those that are “packed” (so that multiple requests are viewed in a single unit) or those that involve viewing just a single requester.

In an experiment testing this, 411 online participants completed a two-minute survey. In the survey, they viewed four needy animals (profiles from GoFundMe) requesting the same amount for various health problems. We gave participants a $0.50 bonus they could choose to spend on the animals or keep. Next, they either viewed each animal profile separately, deciding how much to donate, or viewed them as a single unit. Some of them also saw only a single animal, whereas others saw all four.

The different ways of viewing the animals did not influence the likelihood of donating: in every condition, between 62 and 71 percent of individuals chose to donate. But it did influence how much was donated among those who chose to donate. For those who viewed the animals separately, seeing more animals led to more donations (Ms = $0.42 vs. $0.33). For those who viewed the animals as a single unit, the number of animals in the group made no difference in the total donations (Ms = $0.35 vs. $0.35).

In other words, seeing more animals never reduced donations, and it increased donations when each animal was considered discretely.

One lesson that we think can be learned from these data is that fairness is a powerful psychological motive that can be leveraged in donation decisions.

Moreover, we found the same pattern of donation data in a follow-up experiment in which participants considered donating to needy children from GoFundMe. Participants had the same strong preference for distribution, at a rate of 98 percent, when they donated tokens to jars featuring children in need of medical care. Participants didn’t know that they were participating in an “experiment” or being watched by research assistants, which suggests that the preference for distribution persists even in private prosocial decision-making.

So what does this mean for charities looking to increase their donation totals?

One lesson that we think can be learned from these data is that fairness is a powerful psychological motive that can be leveraged in donation decisions. For example, charities might try designing a donation page with a small number of requesters who seem very similar in neediness or deservingness, as it may make it harder for donors to justify leaving some requesters unhelped. Similarly, charities could even try to amplify fairness concerns by making comparisons between requesters: “You donated $10 to [Requester 1]. Do you want to also donate $10 to [Requester 2]?”

By better understanding how donors make allocation decisions, it may be possible for organizations and individuals alike to more effectively direct donation strategies.