This paper offers a general equilibrium model that explains how the observed correlations of money and output fluctuations may come about through endogenously determined fluctuations in the money multiplier. The model is calibrated to meet long run features of the U.S. economy (including monetary features) and then subjected to shocks to the Solow residual following a random process like that observed in U.S. data. The model's predicted business-cycle frequency correlations, of both real and nominal variables, share the following features with U.S. data: i) M1 is positively correlated with real output; ii) the money multiplier and deposit-to-currency ratio are positively correlated with real output; iii) the price level is negatively correlated with output [in spite of (i) and (ii)]; iv) the correlation of M1 with contemporaneous prices is substantially weaker than the correlation of M1 with real output; v) correlations among real variables are essentially unchanged under different monetary policy regimes; and vi) real money balances are smoother than money demand equations would predict. Although features (i) and (iv) may have been considered support for a causal influence of money on output, the paper demonstrates that they are consistent with an economy in which money has no such causal influence.
