Current conditions in the money and credit markets, along with the memory of the “crunch” of late 1966, have caused both the managers of financial institutions and their regulators to reconsider their concepts of “liquidity.” Both Minsky and Ritter have argued forcefully that traditional attention paid to balance sheet proportions should be abandoned in favor of an intertemporal analysis of cash flows [6], [7], and [8]. Ritter states that: “With a multidimensional cash flow forecast extending several years into the future, probability estimates can be made regarding potential liquidity needs over time” [8]. The purpose of this paper is to suggest a forecasting method which explicitly deals with the problem of uncertainty. The method is specifically designed for a mutual savings bank, but it could easily be adapted for a savings and loan association and, with perhaps greater effort, a commercial bank.2 In Section I, components of cash inflows and outflows are examined with the general objective of classification according to: (1) degree of uncertainty and (2) degree of management control. A risk analysis simulation model which permits explicit consideration of uncertainty is presented in Section II. Problems of implementing the model are discussed in Section III, while the implications of the analysis for future research are considered in Section IV.