KERALA RISING · FROM EXPORTING PEOPLE TO EXPORTING IDEAS framework is built to absorb that: each program publishes its independent assessment in full regardless of outcome, and a published unwind protocol applies if an assessment is negative. Being wrong in public, on the record, is part of the design, it is what makes the audit credible for the programs that follow, and it is the difference between an institution and a sales pitch. Why equity is the risk we take most seriously Of all the structural risks, capture by the alreadyadvantaged deserves the most attention, because the populations these programs most need to reach are the ones least able to participate on equal terms. The evidence is stark in the road-safety case: World Bank analysis finds that poor households bear a disproportionate share of the burden of road crashes, over 70 percent of crash victims in poor households lose income, against far lower rates among the better-off, and that the same households have the least access to social safety nets. A recognition program that inadvertently rewarded only the smartphone-owning, time-rich, already-safe would widen exactly the gap a public institution exists to close. [9] This is why equity is engineered into the mechanics rather than appended as a hope. The populations most in need of the loop are exactly the ones a careless design would leave out, so equity is treated 268