August, 17th 1998

Letter to the Editor of

The Economist


 Dear Sir,

In a recent Economic Focus article your author deals with the "Return of Monetarism". In particular he addresses the issue of the proper strategy for the future European Central Bank. Using an argument with respect to the stability of the European demand and supply for money, your author advises against the use of money supply targets. This argument I find most unconvincing for the following reasons.

 Firstly, your author believes that the observed higher stability of the European money demand function as compared to single country money demand functions will not continue after the switch to a single currency since the reason for this phenomenon ("currency substitution", though the author is avoiding the technical term) will be bound to disappear with the introduction of the single currency. However, with the transition to EMU currency substitutability within EMU rises to infinity and not to zero. In addition, empirically, currency substitution has been insignificant and does not explain the stability differences. Therefore, the falsely assumed disappearance of currency substitution cannot be turned into a convincing argument against a strategy whose usefulness does not even depend on it.

 Secondly, empirical research demonstrates that the observable higher relative stability of the European money demand function is not a matter of currency substitution but is due to an effect of risk reduction by diversification. This effect is known from portfolio theory (portfolio diversification effect) and will not stop working after the introduction of the single currency, quite to the contrary. (Hint: The European Central Bank will hold a portfolio of national money demand functions. The variance of the error term of the European money demand function is subject to the principle of variance reduction by portfolio diversification.)


Thirdly, your author believes that the transition to the single currency "combined with the liberalisation of financial markets" is a structural shift which is likely to cause money supply to become unstable. However, while liberalisation has already been completed and therefore cannot coincide in time with the introduction of the single currency, the structural shift creates a problem for any strategy of monetary policy and not only for a strategy that targets the money supply.

 Fourthly, your author's statement "The money supply is useful as a policy target only if the relationship between money and nominal GDP—and hence inflation—is stable and predictable." is incorrect though it is frequently made (in a fashion reminiscent of the use of Tibetanian prayer mills) and has even entered academic textbooks on money and banking. However, in contrast to your author, I do not consider the number of economists who hold a view as evidence in favor of its correctness. More specifically, the usefulness of the money supply target does not depend on absolute stability but on its stability relative to other relations. If we have a structural shift (change in parameters) why should we observe a decrease only in the stability of the relation between money stock and GNP but not in the relation between interest rates and GNP? This relevant question your author prefers to ignore.

 Fifthly, the Bundesbank chooses the money supply as an intermediary target and combines the intermediary target strategy with rule behaviour. Rule behaviour as opposed to discretionary behaviour can be justified with game theoretic arguments which your author also ignores.

 Sixthly, the increase in parameter uncertainty due to the switch to EMU serves as an additional argument in favor of rule behaviour. Why? Since a policy maker is ignorant about the parametric implications of a structural change when it takes place in an economy he is wise to become less active and, in the extreme, to do nothing because his actions would more likely than not increase the already existing instability. The expected increase in uncertainty therefore strengthens rather than weakens the case for rule behviour.

 Seventhly, inflation, a long run phenomenon, is determined by long run or average money supply gowth. If you want a stable rate of inflation why should you adjust monetary policy according to observed short run changes in the price level? Since your author likes analogies let me ask him the following question: Suppose you have to take pills, daily, against high blood pressure, would you change the dosis, daily, according to your daily changing measurements of your blood pressure? (If you don't know the answer, please ask your doctor).

 Eighthly, your author acknowledges that the Bundesbank was very successfull in its policy.

Nevertheless, he recommends the policy of less successful central banks. This sounds like an odd preference for second best solutions. Upon closer examination the recommendation looks less irrational. Mind, controlling the money stock by following a money supply rule implies a higher volatility of interest rates, something bankers don't like. Obviously, your author's argument may be reduced to a famous but nevertheless fallacious statement : What is good for General Motors (the City) is good for America (Europe).

 As a consequence and as far as the interests of Europe are concerned, I find your author is clearly out of focus.


 Nikolaus K.A. Läufer

University of Konstanz


Zur Homepage des Verfassers