01| When a disaster hits, the question has an answer. Did the food arrive? Were the tents up before the rain came? Did people live through the week? It is a brutal question, but it closes. Somebody can stand up at the end and say yes or no. 02| Poverty work outside a crisis has no such moment. An organisation can work the same district for twenty years, raise money every year, publish a report every year, and never arrive at a point where anyone says plainly whether it worked

01| In April 2017, at the World Bank's Spring Meetings in Washington, the institution's president described how his own organisation rewards its staff. "Inside the World Bank now, if you do so well that it begins to look commercially viable, the incentives are aligned so that you quickly get it through the board and get your credit for having got that loan out the door."— Jim Yong Kim, World Bank Spring Meetings, April 2017 02| He went on to describe what he wanted instead —

01| The Center for Global Development's review of how evidence gets used in aid produced one finding that deserves to be better known than it is: only about one in five impact evaluations considers what a programme achieved per dollar spent. 02| Read that carefully. It is not a statement about how much aid goes unevaluated. It is about the evaluations that do happen — the serious ones, with budgets, designs and published results. Four out of five establish whether something w

01| A dollar committed by a donor passes through several organisations before anything reaches a household. A bilateral agency funds a multilateral or a contractor, which funds an international NGO, which funds a national organisation, which runs the work. Each layer costs something. The obvious question is how much of the original dollar survives. 02| Figures circulate. Ten percent reaches the poor. Only a third arrives. Somewhere between a third and two-thirds, depending on

01| Everything wrong with poverty work at a distance — no test, no verification, nobody checking — should disappear when the work happens close to home. A neighbourhood is small. You can walk it in twenty minutes. The people affected live at addresses you can visit. The statistics already exist: school results, crime figures, employment rates, health records, collected by somebody anyway. If any kind of poverty work can be tested, this is the kind. 02| That reasoning is why e

01| If nobody can prove a neighbourhood programme worked, something else decides whether it continues. Usually two things do. The organisation asks the people who pay for it, and the organisation asks the community. 02| Both sentences hide a problem, and the problems are opposites. The first is the one everybody complains about. The second is the one almost nobody wants to look at. Fifty-seven years of saying the same thing 03| Almost every organisation working in a poor neig

01| Stand on a paved footpath in an upgraded section of Mathare, in Nairobi, and ask a set of ordinary questions. Who built this? What did it cost? Who decided it should be here rather than two streets over? Is there a plan for the parts of the settlement still under mud? How many people does the organisation that paid for it say it helped, and how would anyone know? 02| Every one of those questions has an answer somewhere. What is interesting is how far you have to travel to

01| A common charge against development work is that it has no coordination architecture. Humanitarian response, the argument goes, has clusters, country teams, response plans and pooled funds — a system that assigns responsibility and produces a single plan. Development work in the same city has a dozen agencies operating in parallel with no shared plan, no shared data and no shared accountability. 02| The observation about the dozen agencies is accurate. The explanation is

01| The claim is easy to state and awkward to prove. A minister has discretion over a pot of money for small social projects. The projects are approved, announced and delivered. Their effects fade within a few years because nobody follows up. But the minister's network of obligated people grows, and that network has value long after the projects stop working. Nothing illegal occurs at any point. 02| This is not a suspicion in search of evidence. It is one of the better-studie

01| The idea is elegant enough that it keeps being reinvented. Stop paying organisations for activities and start paying them for results. Define the outcome, agree a price, verify independently, pay on delivery. Everything wrong with input-based funding — the disbursement pressure, the unfalsifiable reporting, the absence of any test — dissolves, because nobody gets paid unless something measurably happened. 02| This has not been a thought experiment for some time. It has be