This paper addresses empirically the issues of selecting the appropriate opportunity cost variables in the money demand function in the presence of alternative non-nested hypothesis. Adopting a conventional real partial adjustment hypothesis, the relative performance of two alternative competing specifications for the demand for narrowly defined money with domestic and foreign short-term deposits rates as alternative measures of the opportunity cost of holding money balances is compared utilizing formal significance tests of model specification in the presence of alternative non-nested hypothesis. The tests are carried out with quarterly data from the small open economy of Kuwait over the period 1983-90. The tests unequivocally discriminated in favor of a foreign interest rate specification. A brief discussion of the factors explaining our results is presented.