by Dan Mahoney
Abstract: In this article we consider an argument put forth by Selgin (1988) in support of the claim that there exists a mechanism for limiting coordinated expansions of fiduciary media under a system of fractional reserve free banking. Selgin argues that such banks hold risk-adjusted reserves against expected losses, and even if the expectation of reserve losses remains zero, the variance of such losses (adverse clearings) increases under an in-concert expansion (if such expansion is unwarranted by demand). It is this increased variability that is claimed to act as a brake on the expansion. We take issue with this argument on the basis of the fact that such a characterization of observed clearings would require that characteristics of the underlying data-generating process be obtainable from pathwise realizations of that process. In other words, there is an implicit assumption of stationarity (or more strongly, ergodicity) in Selgin’s argument, and this assumption is at odds with well-known empirical facts of non-stationarity associated with most economic time series. We also point out ways in which techniques of risk management commonly found in the modern financial industry are unlikely to be effective in addressing this problem.