A pragmatic view of money

By: Stefan Gerlach, University of Frankfurt

 

The role of money in monetary policy remains a controversial issue, as evidenced by a seemingly never-ending stream of newspaper articles, conferences and speeches by central bank governors. This is a remarkably high profile for an issue that would strike many as less important than several other monetary policy questions – such as whether central banks should reveal their interest rate projections or whether it is important to have “outsiders” on monetary policy committees – that have attracted little or no public attention.

 

To my mind, the role of money in monetary policy is straightforward. Monetary policy must be forward looking and thus be set on the basis of inflation forecasts since it takes some time for interest rate changes to impact on the economy. If money growth is useful for forecasting inflation, it makes sense to look at it; if not, it doesn’t. My reading of the vast body of research that has studied this question is that in many countries, in particular in the US, money growth does not help forecast inflation so one need not pay particular attention to it when setting interest rates.

 

However, that same literature also indicates that in other economies, including the euro area, money growth has in the past – and that’s all we have data on – been useful for forecasting inflation, which suggests that money should not be disregarded in setting policy. Of course, other variables such as exchange rate changes and oil price changes may also help forecast inflation and should be considered too, with the relative weights attached to money and the other variables depending on their usefulness in predicting inflation – and there is no a priori reason to assume that money will be more (or less) important than the other variables.

 

While I think that this view should be rather uncontroversial, many find it naïve or worse. The counter arguments are far too many to be considered here but let me touch on a few.

 

Some argue that money should play no role in monetary policy because the relationship between money growth and inflation is unstable over time, and assert that although money growth may have been useful in forecasting inflation in the past there is no reason to believe that it will be so also in the future. These arguments are of course correct but, since they apply with equal force to most if not all other variables used to forecast inflation, seem a bit off the point. Certainly the output gap, regarded by many as a much better variable to rely on, together with current inflation, when setting interest rates, is also subject to these critiques. Output gaps must moreover be estimated, something that the evidence squarely shows we are very poor at. To paraphrase Winston Churchill’s comment on democracy, I tend to believe “money growth is not a good variable to forecast inflation with, except when you consider the alternatives.”

 

A version of this critique asserts that money growth has recently become less helpful in forecasting inflation in the euro area as inflation rates have declined to, and fluctuated around, two percent per annum, and can therefore safely be abandoned. The first part of this argument seems to be correct but whether that suggests that the ECB should downplay or disregard money or not depends critically on whether the relationship between money growth and inflation has shifted.

 

One explanation for the recent inability of money growth to forecast inflation is that large and recurring shocks to velocity – that is, to the demand for money – have destroyed the close correlation between these variables. If this is so and if this situation is permanent, then it does indeed seem sensible to reduce the weight attached to money in forecasting inflation and in setting interest rates.

 

But there is another explanation for why money growth and inflation may have become less correlated: better monetary policy. Suppose that money growth in truth matters for inflation but that monetary policy is perfect in the sense that the ECB perceives all economic developments that could impact on inflation as they occur and moves interest rates swiftly to offset them. In this best of all possible worlds, inflation would always be at the ECB’s objective, that is, a little below two percent. Moreover, it would be uncorrelated with money growth which would fluctuate with interest rate movements while inflation remains constant! Indeed, all variables that play a role in the inflation process would spuriously appear as being irrelevant for inflation.

 

Thus, without knowing whether money growth has become less closely linked to inflation because of velocity shocks or better policy it is difficult to draw conclusions about what that implies for the role of money in monetary policy.

 

It seems hard to deny that monetary policy improved in the euro area as a consequence of the establishment of the ECB so that the “better monetary policy hypothesis” deserves some further consideration. Two aspects of this hypothesis deserve to be noted.

 

First, it accounts for the fact that the link between money growth and inflation (and indeed between output gaps and inflation) was particularly obvious in periods in which monetary policy has been conducted staggeringly poorly, such as in the 1970s.

 

Second, it explains why, as is commonly argued, money growth, domestic output gaps and exchange rate changes all appear to have lost significance for inflation in recent years.

 

But let me end by returning to my starting point. Too much air time has been devoted to the role of money in the conduct of monetary policy. Let us move on to other, arguably more important, matters.

 

 

Stefan Gerlach is professor of monetary economics at the Institute for Monetary and Financial Stability at the Goethe University in Frankfurt.

---