Sorry, I Still Think MR Is Wrong About USAID

Scott's data-heavy rebuttal to Tyler Cowen (Marginal Revolution) on USAID/foreign-aid waste. Core reporting: the oft-cited ~30% 'NICRA' overhead is a negotiated legal-accounting fiction, not real overhead; Catholic Relief Services' true administrative overhead is 6.3% and JHPIEGO's 3.9% - comparable to Cowen's own 2-5% Emergent Ventures regrant overhead, which only looks lean because George Mason covers Mercatus's bills. Publicly corrects his own prior error (imagined 40% double-overhead; actually ~7.2% worst case), adding it to his Mistakes page. Broader point: the 'USAID is 90% grift' narrative collapses against where the money actually goes (food, local salaries, health, Ukraine), and the abrupt 2025 termination - with no transition to local charities - will likely kill millions. Lands a 'missing mood' critique (Caplan) and rejects the criticism-as-violence move, with venom (the Hell aside, the standing/credentials jabs).

Quality 73 · Strong. Strong band (upper). Rigorous, genuinely informative on how aid overhead actually works, and notable for the honest public self-correction; few would do this homework. Held in Strong rather than Excellent because it is a reply-in-an-ongoing-debate, heavily inside-baseball, topical to the 2025 USAID news, and not self-contained. A=73.

Claude’s paradigm shift 35 · Slight. Slight. A factual rebuttal in a live debate; the contribution is correction and reporting, not new ideas ('missing mood' is Caplan's). B=35.

Real-world impact 2 · Minor. A rigorous, genuinely informative rebuttal on how foreign-aid overhead actually works (the NICRA accounting fiction vs ~6% real overhead), notable for an honest public self-correction. Real topical relevance to the USAID debate, but an inside-baseball reply-in-an-ongoing-debate — low RWI.