Recently, a newly fashionable theory-the backing theory-has been used to explain the purchasing power of paper money during a number of historical episodes. The terminology is unfortunate, because the theory is different from traditional theories of backing. The backing referred to by the new theory consists of government tax revenues plus other government assets, not the traditional commodity backing. Accordingly, the new backing theory holds that the purchasing power of money is determined by government fiscal policy.' The new backing theorists claim that the price level will not be affected by changes in the quantity of money provided appropriate fiscal policies are followed. The theory has been used by Bruce Smith and Elmus Wicker to explain the purchasing power of American colonial currencies, and by Charles Calomiris, in a recent article in this JOURNAL, to explain the purchasing power of the continental currency issued during the American Revolution.2 The purpose of this comment is to address a new twist Calomiris has given to the backing theory, to review the evidence he presents, and to show some points of similarity between the colonial and revolutionary episodes. My position is that the backing theory has little or no explanatory value when applied to either the American colonial or revolutionary periods.3 Smith, Wicker, and Calomiris conclude that changes in prices were largely unrelated to changes in the money supply, but they overlook two important institutional facts. First, the measures of the money supply used by all three are seriously flawed. In the case of colonial America, the money supply data do not accurately measure even the amount of paper money in circulation