Circulation or Category? A Null Result on the Boundary of the Lender of Last Resort

What sets the boundary of the safety net? In a crisis a central bank rescues some private near-monies and lets others fail, and a tempting answer to which is that rescue tracks an entity's position in the payment, settlement and collateral plumbing, its "circulation-centrality", rather than its size or its legal category. This paper builds that hypothesis into a falsifiable test and reports that it fails. On a panel of thirty rescue decisions from 1970 to 2023, a blind-coded index of circulation-centrality at first appears to separate the rescued from the abandoned. The appearance does not survive scrutiny. Blind coding removes a thirteen-point hindsight inflation in the author's own scoring; dropping two index components that quietly restate the rescue trigger removes the construct's circular content; and correcting two contested keystone cases (Lehman Brothers, whom the Federal Reserve could have saved, and Washington Mutual, whose depositors were in fact protected) removes the rest. Conditioning on legal-political capacity rather than deleting the inconvenient cases, the circulation signal falls to a partial correlation of 0.13, and under an independent coder's blind reconstruction of the covariates it does not appear at all. The boundary of last-resort lending is governed by size and political-legal capacity, the familiar too-big-to-fail account, not by a distinct circulation construct. The paper offers the test as a reusable template and the false positive as a cautionary anatomy.