An alternative theoretical setting is presented to characterise the money demand and the monetary equilibrium. Two main hypotheses are stated that contradict the assumptions normally sustained by scholars and policy-makers: national output is assumed to be a random variable, and people are supposed to face borrowing restrictions in capital markets. After the model of James Tobin, 1958, the demand for balances is determined in order to maximise the expected return of a certain portfolio combining risk and cash holdings. Unlike the model of Tobin, the prices of the underlying exposures are established in actuarial terms. Then the efficacy of monetary policy is explicitly affected by the expected return and the volatility of the series of perc...