Monday 23 February 2015

Should a Central Bank Target Domestic Inflation Only?

Inflation targeting and cholesterol...

Central banks around the world generally have a mandate to target inflation. Whilst this makes sense, I would argue that it is not specific enough and that this target by its nature leads to policy errors.

I think of this as similar to the question of cholesterol. Whilst it is widely considered fact now that there is a 'good' and a 'bad' cholesterol (HDL and LDL respectively), for years the two were linked together in one number and labelled 'bad'. Since the 1950s, this has led to poor dietary recommendations as a diet that increased the 'good' cholesterol was shunned.

With inflation targeting, I would argue that there similarly are two components. Domestic and foreign inflation. And these two components require opposite action regarding interest rates.

Inflation on domestically produced goods and services

When demand goes up and supply is limited, the result is inflation. This is the sign of a healthy economy. Too much demand can lead to overheating and therefore should be controlled by raising interest rates to increase savings. Too little inflation is a sign of too little demand and this demand should be stimulated by a reduction in interest rates. This is a rational justification for inflation targeting. And I would argue that it is completely true for inflation on domestically produced goods and services. This is 'good' inflation targeting.

 Inflation on imported goods and services 

There are a number of reasons why imported goods may go up or down relative to domestic goods. Commodity prices can go up and down. Exchange rates can change. Internal wages relative to output can change in other countries. I will go through each of these and explain why I think that this, if anything, should have an opposite impact on the desired interest rates. I call this 'bad' inflation targeting.

Commodity prices: When commodity prices go up, inflation in commodity importing nations goes up. This leads to a reduction in buying power for consumers and, in turn, to a reduction in demand for home produced goods. So in this case inflation (deflation) should ideally be combated by a fall (rise) in interest rates. However, mandated central bank policy suggests that the opposite should be done. In terms of bad policy created by this, one could, for example, look at the ECB raising rates in 2011 at a time when inflation was high due largely to high commodity prices.

Exchange rate changes: A stronger domestic currency leads to lower inflation on imported goods. But it increases the buying power of domestic consumers. Therefore it is likely to increase overall demand but at the same time export some demand to the countries with weaker currencies. Effectively this is an increase in living quality for the domestic consumers. Here an increase in interest rates would help increase savings and keep demand in check. Vice versa for a weakening currency. Once again, mandated central bank policy suggests the opposite should be done.

Wage inflation in other countries: Supposing another government, let's call it China, follows policies which keep currency undervalued, wages low, and interest rates negative in real terms. Production of goods in that country is going to be cheaper as it is heavily subsidised. This is going to lead to deflation being exported. This can mask the effects of inflation in the domestic country. I believe that this should be treated as the boom that it is - a large increase in the buying power of domestic consumers. Therefore interest rates should be increased to quell demand and counteract the build-up of external debt. Instead the opposite has been done. Central bankers have seen the low inflation and kept interest rates low. For examples of policy errors, see the UK and US (to name but two) with interest rates too low in the years leading up to 2008.

To put this more generally, imported inflation/deflation is like a tax/subsidy on the people of the country. Monetary policy should be used as a counterbalance to this, rather than reinforcing the effects further.


'Good' and 'bad' inflation targeting should be separated into the two parts - the domestic and the imported. At the very least, the imported inflation/deflation should be ignored in the calculation. Even better, the two constituents of inflation should be studied separately and an optimal interest rate should target a composite figure that could actually (if empirical evidence supports it) have a negative coefficient for imported inflation.

No comments:

Post a Comment

Sorry, I have had to moderate comments because of an annoying pest spammer who keeps posting American football matches links.