2.3. Que faut-il retenir du monétarisme ?

Opérer une distinction entre deux régimes : la haute inflation et la basse inflation

http://www.econ.kuleuven.ac.be/ew/academic/intecon/Degrauwe/PDG-papers/quant.pdf

Is inflation always and everywhere a monetary phenomenon? There exists a strong consensus among economists today that when analysed over a sufficiently long period of time inflation is indeed everywhere a monetary phenomenon. This consensus has not always existed. (…) The view that inflation is always and everywhere a monetary phenomenon has a long tradition based on the quantity theory of money (QTM). In its simplest form, the QTM says that changes in money supply growth are followed by equal changes in the inflation rate and, by the force of the Fisher effect, in the nominal interest rate. The QTM is a measure of the extent to which the inflation movements can be explained by purely monetary forces. The one-to-one relation between inflation and money growth is a characteristic of long-run average behaviour of the model economy. These conclusions are now widely and firmly held by economists. (…) The essence of the quantity theory of money is that it consists of two elements. First, the theory predicts that in the long run there is a proportionality relation between inflation and the growth rate of money, i.e. in a regression of inflation on money growth the coefficient of money is estimated to be 1. Second, it assumes that over a sufficiently long period of time output and velocity changes are orthogonal to the growth rate of the money stock. The main prediction follows logically from this assumption. Thus, there are two aspects to the quantity theory. The proportionality prediction says that a permanent increase in money growth leads to an equal increase in the rate of inflation in the long run, while the orthogonality assumption says that a permanent increase in the growth rate of money leaves output and velocity unaffected in the long run. (…)

Over the long term (thirty years) the orthogonality assumption of the quantity theory is confirmed, i.e. money growth has no permanent effect on output growth. The prediction of proportionality, however, is not maintained. For the sample as a whole we find that the coefficient of money is systematically higher than 1. When we split the sample into subsamples according to the level of the rate of inflation, we find a very low and insignificant coefficient of money in the class of low inflation countries. Thus for low inflation countries the quantity theory prediction that inflation is a monetary phenomenon is not confirmed. Things are very different in the class of high inflation countries. There we find a coefficient of money growth significantly higher than 1. Thus, in this group of countries, money growth has a more than proportional effect on inflation. The picture that emerges from this analysis is the following. In the class of low inflation countries a higher growth rate of money does not lead to an increase in inflation in the long run, nor does it affect the rate of growth of output. This suggests that there must be a negative correlation between money growth and velocity growth. This conclusion follows from the fact that m+v=p+y is an identity. (…)

Conclusions

(…) The relation between inflation and money growth for low inflation countries (on average less than 10% per year over 30 years) is weak, if not non-existing. From our panel data analysis we conclude that there is no evidence for a long-term proportional relationship between money growth and inflation, as predicted by the quantity theory, for low inflation counties (i.e. yearly inflation of less than 10 %). We also find, however, that this lack of proportionality between money growth and inflation is not due to a systematic relationship between money growth and output growth. We find that in accordance to the QTM assumption money growth and output growth are orthogonal in the long run, i.e. higher growth rates of money do not lead to higher growth rates of output. This finding is consistent with the large number of econometric analysis using time series of single countries. Most of these studies have found that money is neutral in the long run, i.e. does not have permanent effects on output. A third finding (obtained from a panel data analysis) indicates that country specific effects become increasingly important when the rate of inflation increases. We interpret this to mean that velocity accelerates with increasing inflation; thereby leading to inflation rates that exceed the growth rates of the money stock. This also explains why in cross-section regressions inflation rates increase more than proportionately to money growth in high inflation countries.
(…)

Our results have some implications for the question of the use of the money stock as an intermediate target in monetary policy. As is well known, the European Central Bank continues to give a prominent role to the growth rate of the money stock in its monetary policy strategy. The ECB bases this strategy on the view that “inflation is always and everywhere a monetary phenomenon”. This may be true for the high inflation countries. Our results, however, indicate that there is no evidence for this statement in relatively low inflation environments, which happens to be a characteristic of the EMU countries. In these environments money growth is not a useful signal of inflationary conditions. It also follows that the use of the money stock as a guide for steering policies towards price stability is not going to be useful for countries with a history of low inflation.