Should central banks target anything other than inflation (such as money supply)?

By: Patrick Minford, Cardiff Business School

 

Before answering this question it is useful to take a look at the two main environments for monetary policy in the past 30 years.

 

The first is one- I have in mind the 1970s and early 1980s- when there is high and variable inflation and there is general scepticism among the general public backed up by hostility among the well-informed (the ‘chattering classes’) to the idea that monetary policy can control inflation. Instead political opinion holds it that controls on wages and prices should be used to hold down inflation and monetary and fiscal policy should accommodate to that resulting inflation rate, at pressures that maintain full employment.

The first environment is one where monetary tightening has ‘no credibility’ as regards to its capacity to reduce inflation more than temporarily. People in the markets- be they workers, investors or consumers- will expect the tightness to be ended as soon as unemployment starts to rise beyond some low ‘pain barrier’. Of course we know that wage/price controls cannot bring down inflation; only permanent reduction in money supply growth, backed up by fiscal commitment not to print money faster for reasons of public finance, can do the job. So in this environment there has to be a change in regime whereby the government commits to such policies and continues with them until the scepticism is defeated. This process is painful and can be called the ‘battle against inflation’ which we saw played out in the early and middle 1980s in most OECD countries.

 

The second environment is one where this battle to bring down inflation has succeeded and inflation has settled at a reasonably low rate, say in a range of 2-5%. By this point public opinion and well-informed chattering opinion has learnt that monetary policy can do and has done the job of reducing inflation and has not done so at the expense of unemployment. A new cross-politics consensus forms that inflation should be dealt with by the technical means of monetary policy- just as crime should be dealt with by police and law courts. It follows naturally that it should be delegated out of politics to a quasi-independent ‘monetary judiciary’. This body is given a clear mandate to control inflation and this is framed often in the form of an explicit inflation target. In practice the central bank varies interest rates to hit this target. This of course is modern ‘inflation targeting’.

It has caused some people surprise that it has been so successful. It should not have. The big surprise was the conquest of inflation in the first environment. Many do not remember how difficult this was for the politicians and their advisers for the simple reason that it involved allowing unemployment to rise in the short term to high rates previously regarded as politically intolerable. Thus these people faced not merely scepticism but disgust and even hatred.

 

That battle once won and opinion educated, the process of containing inflation is both simple and uncontroversial. Furthermore everyone understands it so that they will naturally assume it will be successful. Since no political constituency any longer wants inflation to be put up or taken chances with, everyone will assume that it will not in the longer term vary from the target.

In the current environment monetary policy works through expectations predominantly. This is not the place to solve equations for a model of the economy but it is easily shown that the central bank’s monetary policy rule to vary interest rates in order to ensure an inflation target it met will produce an inflation rate that varies from that target only temporarily; this is because expectation pin down the long term rate of inflation. In concrete terms it means that wages are driven by expectations of inflation around the target, similarly long term bond rates of interest, share pries, house prices and so on.  The economy’s behaviour will consist of temporary, mutually consistent variations in output, inflation and interest rates. This consistency is governed by the equations of aggregate demand, aggregate supply and of course the central bank’s rule. To ensure this all the central bank has to do is stick to its rule- to raise interest rates when inflation and output rise and vice versa to lower them.

 

Now consider whether the central bank should target anything else than inflation if it is content with this behaviour. The answer is clearly no! Any other target will simply cause confusion and inconsistency if it would mean that the inflation target would be overridden. I will consider in particular the money supply – at the centre of the ECB’s second pillar.

 

So what about the money supply?

A central bank with an inflation target spends its time forecasting inflation, using its models and all the other information available. This includes such things as wages, prices themselves, growth, unemployment and much else. The idea is to produce a coherent view of the whole picture, with the inflation forecast emerging from it naturally. Imagine that it forecasts inflation will be 2% and growth 2%, but that the money supply (say a ‘broad money’ like M3 or M4) will be 12%. Then it is saying that the ‘velocity’ (the ratio of GDP in money terms to the supply of money) of this money will fall by 8%. This is not an absurd forecast; indeed this has been happening for several years now in the UK and to a more modest extent also in the euro-zone. It could have lots of reasons, to do with the structure of lending, borrowing and cash holding by firms and people, but particularly financial firms who make their money from these operations. Forecasting exactly what will change the balance sheet operations of these firms is a nightmare in fact because of the huge opportunities in a deregulated ‘money market’ to switch between different financial channels.

 

Now consider the problem for this central bank when a ‘monetarist’ comes on the scene, proclaiming that the growth of some particular money supply aggregate should not exceed x%. This creates a problem. How does this x% fit into the picture? What other parts of the picture are wrong and why? Why x% in particular? There is a strong sense of arbitrariness as compared with the painstaking building-up of the whole picture.

 

This is the problem self-inflicted currently faced by the ECB. It has ‘two pillars’- an inflation target and a money supply growth target range. The two are in conflict. What should it do? Battle rages monthly within the ECB council.

 

Now the Bank of England appears to be suffering from the same problem. Some members of its Monetary Policy Committee are reported as starting to ‘worry about’ excessive M4 growth, running around 12 % currently.

 

The trouble is the pure inconsistency involved. Were a central bank to have a money supply growth target, as used to be the case in the 1980s with several, then that would be a defensible system. The Bank would forecast likely velocity change and normal output growth use these to decide on a good target growth for the money supply given that it wanted to use this to control inflation ultimately. Then if money supply exceeded this target it would raise interest rates and vice versa. Of course if someone came along and objected that they thought inflation was exceeding the implicit target, they would be told: ’we have taken this into account in our money growth target. If money growth is kept to target, then inflation will come into line in time.’ The  objector would have to justify himself by showing that the assumptions used in the money supply growth target were wrong, because the Bank was operating within this money targeting framework.

 

The mirror image is true today with those who object based on the money supply. It is incumbent on them to show that the forecast of inflation based on the various models of the economy used by the Bank and others is wrong and therefore that the velocity forecast of those models is wrong too. This these objectors do not do. They use primitive relationships which have no justification in today’s modelling approaches. They are refusing to subject their implied forecasts of inflation to the same framework that the central bank is using.

 

This is fairly obvious from a perusal of inflation forecasts available in the market place, be they published consensus forecasts from the forecast industry or financial market implied inflation components in bond prices for example. These forecasts have barely flickered out of the 1.5-3% range in major OECD economies for quite a few years. The reason I have argued here lies in the nature of the inflation target regime; the markets argue that any systematic departure from the target range will result in interest rate changes that would push it back. Hence there is a widespread belief that inflation will settle within the target range. This very belief conditions actual behaviour for example in wage settlements or price increases.

 

Our objectors retort that there is some basic relationship between money growth and inflation that overrides such mechanisms. But where can one find such regularity? As money growth of various definitions in the US, the UK and the euro-zone show, it just does not exist. Velocity growth is all over the place. It is for that reason that there is not a major central bank left in the world that operates an exclusive money growth target: the necessary adjustments that would need to be made for velocity growth look too daunting.

 

It may well be that bank officials dress their worries about inflation up in terms of money supply growth; that is understandable disingenuousness, bred of a desire to give concrete form to a poorly understood concern. But to argue that one can run an inflation targeting regime and also target the money supply is to lose all intellectual coherence.

 

Conclusion

 

In today’s world, which comes after a major battle was fought and won in the 1980s to effect the monetary policy conquest of inflation, monetary policy no longer has to prove itself. It is like a boxing champion with no opponents, or like a police force with no crime. Expectations that it will succeed do its work for it if it sticks boringly to an inflation targeting interest rate rule. It is natural for monetary policy-makers to be haunted by nightmares that all this might suddenly change; in the course of these nightmares they mutter about targeting extra variables- the money supply, asset prices perhaps- whose hitting will make them feel more comfortable that the slain dragon will not reawaken. But nightmares are just that. The rest of us should gently remind them that such thoughts are not within their remitted powers; talk to them like a psycho-therapist to calm their fears. Above all, we must stop them lashing out to do unnecessary and potentially damaging things to markets which are none of their business.

 

 

---