One Acre Fund

One Acre Fund: The (Green) Revolution will be Randomized

One Acre Fund is that rarest of gems: a development initiative that truly gets it all the way through. The program is built on a commitment to understand the specifics of what it is that’s holding down the incomes of some of the world’s poorest people.

One Acre Fund is now involved in the Mother of All Randomized Controlled Trials

For very-small-scale farmers in Sub-Saharan Africa, the answer isn’t really mysterious: people farm using crummy seed, no fertilizer and outdated techniques, and so yields suck.

How you go about addressing this makes all the difference. And One Acre Fund got the memo: when something genuinely helps people overcome poverty, they’re glad to shell out the money to pay for it.

This social enterprise model, where the project sets out to cover its own costs through payments made by participants, creates a subtle but profound shift in the way power flows through the organization.

It’s easy to by cynical, but One Acre Fund’s promise that its recipients are “the boss” isn’t just verbiage. The people who pay the bills are always the boss. Organizations evolve to serve the interests of the people who pay the bills.

This isn’t some fresh new insight. Everybody already knows that in traditional donor-funded interventions, feedback loops connecting recipients to donors are broken, ensuring all the power and most of the attention flow back to the donor.

It just takes some creativity to stand that equation on its head, getting beyond the verbiage of project ownership and aligning the ambition with the cold financial facts on the ground.

One Acre Fund is now involved in the Mother of All Randomized Controlled Trials, a massive J-PAL thing involving 232 farmer groups. So we’ll know to what extent it actually helps in practice. But, in principle, it’s hard for me to imagine that they’re too far off on the wrong track.

For much more, on One Acre Fund, check out Roger Thurow’s important book, The Last Hunger Season.

3 thoughts on “One Acre Fund: The (Green) Revolution will be Randomized”

  1. Hi Francisco, I am enjoying your posts and campaign! If you are looking for more examples like One Acre Fund, I would definitely check out Carbon Roots International (carbonrootsinternational.org). They have a very similar approach for a product called green charcoal. I have been working with them for the past year and just wrote a post on how the developed their marketing strategy by leveraging the knowledge of the “real experts.” http://www.theyoufinder.com/blog/empathy-in-development-the-real-experts

    Look forward to reading more!
    Katalina

  2. I remain skeptical of RCTs, borrowed from the stodgy pharmaceutical trial, where they became notorious for producing results rife with funder bias (and indeed, such experiments are deliriously expensive: how many organizations would be willing to run one, undergoing the years-long headache of ensuring that the results are reported in such a way to enable peer-reviewed and prime publication, unless they had a pretty good idea that the process of testing their hypothesis would not make them look like nincompoops?).

    At least in the pharmaceutical industry, the selection of a control group is fairly easy: there are people with established, agreed-upon maladies who are either administered the treatment or not administered the treatment. But how do you select the B group in a social sciences test? Given how fast urbanization is happening in much of Africa, maybe you can test against households given an opportunity to have a member work in the capital city; or, even bolder, perhaps you could test against a group of households where one member is given a work visa for a country in the European Union (http://www.theatlantic.com/international/archive/2014/02/why-brain-drain-can-actually-benefit-african-countries/283750/)?

    Those would be some wide-open tests, but the results are harder to predict than “do people with financial capital do better economically than people without financial capital?” Would the One Acre Fund submit a funding request for such a test? (The answer probably depends on what their surveys say is the economic benefit to a household of being given loans, versus what the literature says is the benefit of migration. If I were in their shoes, I would not submit a funding request for an experiment whose results might require me to fire some of the finest agricultural minds in development today, and replace them with a bunch of out-of-work travel agents, booking bus and plane tickets).

    (I don’t want to seem to pick on the One Acre Fund, which enjoys a sterling reputation. Indeed, I believe it is not alone in being an organization “glad to shell out the money to pay for” interventions that seem to “genuinely help[] people overcome poverty.” Indeed, I don’t know that you could find an interlocutor prepared to argue that there exist a large number of NGOs plagued by an unwillingness to spend as much money as they can raise on interventions that they are convinced will work).

    You may be right that development agencies and NGOs are a path-dependent bunch. If so, it may be worth being as skeptical of RCTs as you are of the last several generations’ magic bullets. RCTs are new and shiny right now, but the End to Poverty was new and shiny once, too. So was structural adjustment, and participatory development, and collectivization, and industrial infrastructure development. The fact that the new thing isn’t as rusty as the old things doesn’t count for much.

    That said, the RCT is a unique way of understanding impact that everybody in the development industry should be glad is now available. But an A/B test is not a silver bullet, and it won’t enable us fully “to understand the specifics of what it is that’s holding down the incomes of some of the world’s poorest people.” It will merely tell us that intervention A is specifically more likely than intervention B to improve incomes, saying nothing of interventions C, D, E, F, G, H, I, J, K. . . .

    . . . unless you run this “Mother of All Randomized Control Trials” 63 times over, switching out variables like an international development version of NCAA March Madness. The mind reels at the expense, but, with luck, afterward you could be justified in making the hubristic claim that rigorous scientific testing has revealed to the One Acre Fund “the specifics” that lay behind today’s poverty (at least in western Kenya).

    1. Waitaminute, I just assumed it was J-PAL picking up the tab for the RCT, right? I mean, now that you bring it up, I realize I don’t know that for a fact, but it seemed obvious – you can’t have OAF pay for its own RCT without fatally compromising it, can you? Now you’re making me doubt.

      I think your second-to-last paragraph is weak, though. A lot of RCTs these days – especially bigger ones like this one, compare A, not just to be but also to C, D, E, F, G, H, I, J, K.

      But, seriously, J-PAL pays for the RCT, doesn’t it?!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s