Title: Money Velocity and the Natural Rate of Interest
Since World War I, M1 velocity has been, to a close approximation, the permanent component of the short rate, so that the time-series relationship between the two series has been the same as that between consumption and GDP. This logically implies that talking of ‘money demand instability’, or real money balances as ‘being out of equilibrium’, makes no sense, as it is conceptually akin to talking of ‘instability’ of the relationship between GDP and potential GDP. A corollary is that disequilibria in the relationship between M1 velocity and interest rates (i.e., the cointegration residual being different from zero) do not signal future inflationary pressures: Rather, they signal future movements of the short rate towards its stochastic trend.
A further implication is that, under monetary regimes which cause inflation to be I(0), permanent fluctuations in M1 velocity uniquely reflect, to a close approximation, permanent shifts in the natural rate of interest. Evidence from the Euro area and several inflation-targeting countries is compatible with this notion, with velocity fluctuations being systematically strongly correlated with a Stock and Watson (1996, 1998) estimate of trend real GDP growth. I exploit this insight to estimate the natural rate of interest for the United Kingdom and Canada under inflation targeting: In either country, the natural rate has been consistently declining since the early 1990s.