Three Cheers for Failure

“Fail often, fail fast, fail cheap.” This well-known principle of innovators in Silicon Valley is still anathema to much of the philanthropy business, where failure is a dirty word. We argued in “Philanthrocapitalism” that the failure of foundations to embrace failure is a huge brake on their effectiveness, so we were excited to read the new book by Tim Harford, the Financial Times Undercover Economist’, called “Adapt: Why Success Always Starts with Failure“.

“Adapt” ranges across a wide range of topics from regime change in Iraq, through the global financial crisis, to the efforts to eradicate malaria. Across all of these problems Harford demonstrates that success requires the cultivation of lots of different potential solutions, many of which will fail, rather than single top-down strategies, and rigorous screening of the evidence to find out which of these solutions work.

One of our main arguments for philanthrocapitalism is that, as private donors can do things that government cannot, it can be the supporter of the risky (some might say crazy) solutions that would not get the backing of public funders. This is exactly what Harford finds, including some great examples that we had not heard of. Take, for example, the British Spitfire fighter plane that is widely credited with winning the Battle of Britain in the Second World War, which struggled to get adequate government funding throughout the 1920s and 1930s. “Adapt” tells the story of how a philanthropist called Dame Fanny Houston, the widow of a shipping millionaire, bailed out the project in 1931 with a donation of £100,000.

Highlighting individual successes, of course, is subject to the potential bias of cherry-picking and Harford could equally have written a book about crazy philanthropic failures. Yet another story that was new to us was a comparison made by a team of economists between grants to scientific research from the National Institutes of Health (NIH) and the philanthropic Howard Hughes Medical Institute (HHMI). The publicly-funded NIH, he describes, funds projects with a high chance to succeed in achieving concrete measures of success. The privately funded MMHI, on the other hand, funds for longer and on looser terms, preferring to back individuals rather than specific projects. The MMHI, it turns out, funds more duds – research that makes little contribution to scientific knowledge (measured by citations in scholarly journals). MMHI also funds more stars and, on average, outperforms NIH.

One lesson from this example is the benefits from trying risky strategies. Equally important is the need for measurement. Our critics love to rail against measurement, saying that the work of philanthropy cannot be captured in numbers. That may be true, sometimes. But the rejectors of measurement are playing a dangerous game. In an excellent presentation at the recent Global Philanthropy Forum, the economist Howard White, who is leading the 3ie evaluation initiative started by the Hewlett Foundation, used the example of a World Bank nutrition project in Bangladesh to show that it was only through a randomised control trial that a project that had widely been seen as a success was, in fact, found to have been a waste of money. Philanthropists, just as much as organisations such as the World Bank, need to invest in hard-nosed research to see if what they are doing is working. If they don’t, they may be wasting money which could otherwise be saving lives.

The importance of randomised trials is one of the big messages of “Adapt”, where Harford praises the work of economist Esther Duflo and others to bring this approach to finding out what works in aid. (We will comment in more detail in a future post on her remarkable, must-read new book with Abhijit Banerjee, “Poor Economics: A Radical Rethinking of the Way to Fight Global Poverty”.) He also sees the potential of incentive prizes as a way to encourage innovation yet only pay for success, citing examples such as the X-Prize and the Gates Foundation’s work on Advance Market Commitments that we discuss at length in “Philanthrocapitalism”.

“Adapt” is both entertaining and inspiring in its call for new thinking in how we tackle social and environmental problems. Indeed, it is a must-read for anyone who wants to make the world a better place. We hope that it will both inspire more foundations to learn to love failing and give the critics of measurement cause to reconsider.

Our own modest proposal in this campaign for failure in philanthropy is that we should be doing more to celebrate when bold projects don’t work out. To that end, we would love to see a donor step up to fund a ‘Heroic Failure’ prize. That’s right, a prize for the biggest, boldest failure in philanthropy to reward those who have failed by giving them a chance to try again, to
make failure something that philanthropists talk about, and (perhaps most importantly) so that we can all learn from what doesn’t work and thus stop repeating the same mistakes.