“Designing Donor Registries: Behavioral Drivers of Enrollment and Giving” is the title of the latest scientific article published in the Journal of Behavioral and Experimental Economics. Its authors are Matej Lorko (NHF EU in Bratislava), Maroš Servátka (MQBS Sydney and NHF EU in Bratislava), and Robert Slonim (University of Technology Sydney). The authors examine how nonprofit organizations should approach potential donors and present several surprising findings. And since the students in the experiment were quite generous, its impact can now be seen as far away as Nairobi, Kenya. We spoke with Matej Lorko about the experiment.
Could you introduce the topic of your research?
Most research on charitable giving focuses on situations where someone asks you for money. But there is often a prior step. First, a nonprofit reaches out with a smaller request: fill out a survey, sign up for a newsletter, leave your contact details, follow us on social media, and so on. This is how you get recorded as a potential donor. The actual donation request comes only in the second step.
We asked whether the form of this first step matters—that is, how the donor is registered. It turns out that it does matter, and more than we expected. The way you invite someone onto your donor list influences not only whether they join, but also how much money they later donate. We tested four mechanisms widely discussed in behavioral economics: aversion to overhead costs, the status quo effect, reciprocity, and so-called moral consistency.
Could you explain these mechanisms?
Imagine I run a nonprofit and want you to join my donor list. Maintaining such a list—and running the nonprofit itself—costs money. These are called overhead costs. Should I tell you about them or not? When registering you as a donor, I also have several options. For example, I can sign you up automatically and give you the option to opt out, or I can wait for you to opt in yourself. As a thank-you for joining, I could give you a small gift. Finally—and this is the least intuitive option—I could ask you to contribute to the nonprofit’s overhead costs before even registering you as a donor.
In our online experiment, more than 400 students went through these situations. Each received real money, which they could either keep or donate to a good cause across four phases spread over several months. This was not hypothetical—donations were real and went to specific projects.
What are your main findings?
There are two key findings. First: when you openly tell donors that running a nonprofit costs money, they give less. Students who were informed about (relatively modest) overhead costs donated significantly less than those who were not told. People simply don’t like hearing that part of their money “goes to administration,” even if that part is minimal. The takeaway for nonprofits: it’s better not to mention overhead costs during the initial contact with donors.
The second finding may be surprising. When we offered students a small gift for signing up to the donor list, more of them joined. However, reciprocity did not work—these students did not donate more than those without a gift, and once we subtracted the cost of the gifts, the nonprofit was worse off. By contrast, when we asked students to pay a small €2 contribution toward overhead costs upfront, we expected it to discourage them. It didn’t. Just as many signed up—and those who paid went on to donate more and more consistently. It seems that once people invest their own money, they start to see themselves as donors—people who support the cause—and stick to that role. This is what we call moral consistency. In our experiment, this approach generated the highest net returns for the nonprofit among all tested methods.
Was there any mechanism that didn’t work at all?
Yes, and it’s instructive. The popular “nudge” of using a default option—automatically enrolling someone with the option to opt out—had no effect compared to letting them opt in voluntarily. In areas like organ donation or pension savings, default options are often very powerful. But in the context of financial donations, the difference between opt-out and opt-in completely disappeared—we found no differences in either enrollment rates or donation amounts.
This is an important finding: no nudge is universal. What works in one domain may be completely ineffective in another. That’s why these mechanisms need to be tested rather than assumed.
You mentioned that the donations were real. Where did they actually go?
We selected the Slovak Integra Foundation, which implements development and humanitarian projects in sub-Saharan Africa, the Middle East, and Ukraine. Participants contributed to three specific goals: planting macadamia trees on small family farms in Ethiopia, humanitarian aid in rural Ukraine near the Russian border, and completing a children’s home in Nairobi, Kenya.
The children’s home has a particularly personal meaning for me. It’s called Praise Gate, and for years, abandoned children from the streets of Nairobi lived there in a makeshift structure made of corrugated metal with a dirt floor. Today, they live in a new building—with running water, flush toilets, and a functional kitchen. So our study is not just a table of numbers in an academic journal; part of the money that “flowed through” it, together with contributions from many other donors, helped build a home for these children. For me, it’s an unusually tangible outcome—normally, research leaves us with a dataset, a published article, or a few citations, not a roof over children’s heads.
By coincidence, I was able to visit them. A year after the experiment ended and the donations were sent, an association of experimental economics researchers organized a conference in Nairobi. So I didn’t hesitate—I got on a plane and visited Praise Gate during the conference. It was an extremely powerful experience. The children live in very modest conditions, but they help each other in incredible ways. All the older ones attend school, some even university. And they still play and laugh.
By the way, did the data reflect anything happening in the world at the time?
In general, we found that donations did not decline over time—students who wanted to donate did so consistently across all phases, even when we asked for contributions to one project (Praise Gate) twice. The best predictor of a donation was the amount the donor gave previously. So giving is largely about identity or habit.
There was one exception. In the phase that took place in June 2023, students donated significantly more to humanitarian aid for Ukraine. This aligns with Google Trends data—the topic of the war was particularly prominent in Slovakia at that time. It’s a small reminder that donors are not detached from reality; they respond to what they perceive as urgent.
What can nonprofits take away from all this?
Three simple lessons. First, during the initial contact with potential donors, it’s better not to mention overhead costs—they reduce willingness to give. Second, don’t expect miracles from small gifts for potential donors. You may get more people into your donor database, but it doesn’t seem to be profitable. Third, don’t be afraid to ask potential donors for a small commitment right away—a reasonable, modest request for a small contribution can turn people into more regular and persistent donors.
