The corners hypothesis holds that intermediate exchange rate regimes are vanishing, or should be. Surprisingly for a new conventional wisdom, this hypothesis so far lacks analytic foundations. In part, the generalization is overdone. We nevertheless offer one possible theoretical rationale, a contribution to the list of arguments against intermediate regimes: they lack verifiability, needed for credibility. Central banks announce intermediate targets such as exchange rates, so that the public can judge from observed data whether they are following the policy announced. Our general point is that simple regimes are more verifiable by market participants than complicated ones. Of the various intermediate regimes (managed float, peg with escape clause, etc.), we focus on basket pegs, with bands. Statistically, it takes a surprisingly long span of data to distinguish such a regime from a floating exchange rate. We apply the econometrics, first, to the example of Chile and, second, by performing Monte Carlo simulations. The amount of data required to verify the declared regime may exceed the length of time during which the regime is maintained. The amount of information necessary increases with the complexity of the regime, including the width of the band and the number of currencies in the basket.