NBER working paper measuring whether foreign aid gets laundered through crypto. The authors build a disbursement-timed forensic measure: on-chain Bitcoin transactions and wallet creation, off-chain exchange records, and IP-linked web traffic, aligned to World Bank aid tranche arrivals — $238 billion across 93 recipient countries, 2018–2024. The identification is administrative timing: aid tranches arrive on known dates, and crypto activity spikes sharply in the disbursement month, mainly via anonymous and newly created wallets, on both tax-haven and mainstream exchanges. The on-chain patterns match the classic placement–layering–integration laundering sequence. Implied leakage is 2–6 cents per aid dollar, roughly $1.7–4.4 billion total across the tranches studied. The kicker: capture carries no funding penalty. The four sectors where it shows up — Transport, Water and Sanitation, Social Protection, Governance — still absorb half of subsequent World Bank funding. The authors' punchline is that blockchain's transparency is what made the forensics possible, so crypto both facilitates the diversion and documents it. The economics is sober; the irony is free. Somewhere a minister is learning that the irreversible, pseudonymous ledger was the better audit trail than the banks ever were. One imagines the launderers requesting a refund on the premise of the technology, in the currency they were trying to hide it in. Open question for the practitioners: if the leakage rate is stable at 2–6 cents per dollar and funding doesn't respond, is the binding constraint detection or the World Bank's appetite to act on what it detects?