In this paper, we extend earlier work on hedging models so that uncertainty about both deposit supply and loan demand is incorporated as well as random rates of return on loans and CD's. Our model suggests that the optimal forward position is the sum of three ratios that should be estimated simultaneously. Using bank-specific data, the optimal hedge ratios are estimated in both the pre-deregulation and deregulation subperiods. Our results show that previous studies of bank hedging with interest rate futures have greatly overstated (a) the volume of short futures positions that banks should take and (b) the degree of homogeneity of optimal hedge ratios across the banking system. Similarly, deregulation has not uniformly affected the interest rate risk borne by different institutions.

In this paper, we extend earlier work on hedging models so that uncertainty about both deposit supply and loan demand is incorporated as well as random rates of return on loans and CD's. Our model suggests that the optimal forward position is the sum of three ratios that should be estimated simultaneously. Using bank‐specific data, the optimal hedge ratios are estimated in both the pre‐deregulation and deregulation subperiods. Our results show that previous studies of bank hedging with interest rate futures have greatly overstated (a) the volume of short futures positions that banks should take and (b) the degree of homogeneity of optimal hedge ratios across the banking system. Similarly, deregulation has not uniformly affected the interest rate risk borne by different institutions.

While it has been known for some time that, under uncertainty, the original version of the Fisher hypothesis is not precisely correct, empirical researchers have largely ignored this fact. Such an omission has possibly resulted in erroneous conclusions concerning other hypotheses; most notably the impact of prices on the real economy. This paper clarifies some of the previous interpretations of the existing empirical literature and provides a theoretical version of the relation between prices and interest rates. Empirical tests based on both the Livingston survey data and data from time-series forecasting models provide support for the Fisher effect and the hypothesis that only covariance risk is priced in the Treasury bill market.

While it has been known for some time that, under uncertainty, the original version of the Fisher hypothesis is not precisely correct, empirical researchers have largely ignored this fact. Such an omission has possibly resulted in erroneous conclusions concerning other hypotheses; most notably the impact of prices on the real economy. This paper clarifies some of the previous interpretations of the existing empirical literature and provides a theoretical version of the relation between prices and interest rates. Empirical tests based on both the Livingston survey data and data from time‐series forecasting models provide support for the Fisher effect and the hypothesis that only covariance risk is priced in the Treasury bill market.