Billions of dollars are being donated without strong evidence about which ways of giving are effective
Philanthropists are flying blind because little is known about how to donate money well. Facebook co-founder Mark Zuckerberg’s US$100-million gift to schools in Newark, New Jersey, reportedly achieved nothing. Some grants to academic scientists create so much administration that researchers are better off without them. And some funders’ decisions seem to be no better than if awardees were chosen at random, with the funded work achieving no more than the rejected.
The recipients of funds are increasingly scrutinized, but the effectiveness of donors is not. Funders are rarely punished for under-performing and usually don’t even know when they are: if the work that they fund helps one child but could have helped ten, that ‘opportunity cost’ is felt by the would-be beneficiaries, not by the funder. The same is probably true of agencies that fund research.
I founded an organization that promotes charitable giving based on sound evidence. I am acutely aware of how scant the evidence is about which ways of giving work best. The solution lies in more research on what makes for effective philanthropy. A ‘science of philanthropy’ could enable more to be achieved with the tens of billions given each year by foundations and other donors and funders.
Only a handful of studies have been done on donor effectiveness. The Center for Effective Philanthropy in Cambridge, Massachusetts, found that the time spent on proposals for, and the management of, ten grants of $10,000 takes nearly six times as long as the time spent on one grant of $100,000. The London-based consultancy nfpSynergy found that UK charities value £2 (US$2.6) of unconditional funds as much as £3 of conditional funds, suggesting that attaching strings to donations reduces their value. And the Shell Foundation found that three times as many of its grants succeeded when the charity was heavily involved in creating and managing the work than when it had funded work based on a proposal from a non-profit.
Establishing the effectiveness of a donor is not straightforward. After all, donors have diverse goals, from funding basic research to testing interventions, providing services or promoting social policies. Nonetheless, answering three questions can provide useful insights for any donor. First, how many grants achieve their goals? (I call this the donor’s hit rate). Second, what proportion of funds are devoted to activities such as preparing proposals or reports for the donor? Third, how satisfied are the recipients with the donor’s process? Logging the goal of every grant and tracking whether these goals were met would be a big step forward.
Several fundamental questions about effective giving have yet to be studied. An obvious one is the role of grant size. Intuitively, larger grants should enable more impact and be proportionally less expensive to manage. But my organization’s analysis of ten years of grants by the ADM Capital Foundation in Hong Kong (published this month) found that grant size didn’t seem to affect success. Similarly, a study of the impact of arthritis research found that large grants were no more consequential than small ones, possibly because smaller grants were awarded for different types of work. Another key issue is whether a broad or narrow scope makes funders more effective. The dominant theory in business is that specialization boosts success; nobody knows whether (or when) that is true in philanthropy.
Other unanswered questions concern the appropriate duration of grants, whether funders do better operating alone or in partnership with other funders, how involved donors should be in the work that they support and how donors should find recipients. Is it better to open applications to everyone, or to approach prospective grantees?
How to select recipients also needs study. Almost all funders make their decisions subjectively, either by soliciting the opinions of experts about a proposal or by interviewing applicants. Research on everything from picking stocks to student admissions shows that humans show weaknesses and biases in allocating scarce resources. The role of biases in awarding philanthropic funds has not been examined. One funder of academic research found that shortlisting applicants on the basis of objective criteria was a better predictor of success (measured by scientific publications) than interviews were. Such findings are intriguing, but still too indiscriminate to yield broad implications.
When medicine became a science, health and longevity increased. Similarly, a science of philanthropy could reveal principles about which ways of giving are most successful. To move in this direction, every funder should gather data about its performance on the three metrics I have outlined, and share these data with researchers. Analyses should be done by researchers, not by the funders or the recipients. The analyses could be retrospective, for example, by assessing how performance and recipient satisfaction have varied with grant duration, or with how recipients were selected. Or it could be prospective: for instance, a funder could deliberately make some grants large and others small, and invite researchers to investigate how grant size affects hit rate and the cost of managing funds.
Such studies will, of course, require resources — from research councils or philanthropic funders. Although that might initially reduce the resources for the work being funded, it stands to improve the effectiveness of that work overall. More evidence about how to fund well could also increase the amount that donors are willing to give.