Tuesday, 26 February 2008

Inflation and the RBNZ

Frederic Sautet has another provocative piece on New Zealand over at the Austrian Economists blog. This time he has a posting on inflation in New Zealand and the role of the Reserve Bank. Sautet opens by making the point some people argue that,
As the prices of some traded goods have gone up and house prices are relatively very high, the New Zealand economy is experiencing "inflationary pressures."
I argue that this kind of thinking can be a problem in that it seems to assume relative price changes are "inflation". If the (relative) price of traded goods or housing are changing why should we worry? Such changes may affect the CPI, but are they really "inflation"? Or at least are they "inflation" we want to worry about. If they are just relative prices changes then we don't want to suppress them. Relative prices are the signals that the economy needs to allocate resources efficiently. If in a mistaken effort to control "inflation" the RBNZ distorts these signals they will do the economy more harm than good. This does not mean we can just forget about inflation and stop worrying. It just means we should try to distinguish between "pure inflation" and relative price changes. An idea discussed here.

Sautet goes on to say
Originally, the government and the RBNZ signed policy agreements to specify the inflation target (something that the US Federal Reserve doesn't have). In 1990, the inflation target was 0-2%; it is now 1-3%.
The thing here is that the target is written in terms of changes in the CPI and so anything that increases the CPI endangers the RBNZ target and thus the bank must react to it. But as I argue above that reaction could do more harm than good, depending on the source of the CPI increase. I doubt the RBNZ can really tell the difference between "pure inflation" and relative price changes and thus may well act when it shouldn't or over react. An obvious counter to this would be, What is a better inflation measure? ... and I don't know.

I have to agree with Sautet over the bad effects that the addition of "broader economic goals" to the RBNZ's mission has had and with his assessment of Bollard's tenure as head of the RBNZ along with Brash's recent statements on using taxation as a means to control inflation.

My real issue here is what in practice does "price stability" mean, how do we measure it and what effects would different definitions and measures have? The Reserve Bank Act type approach to fighting inflation may be the best we can get, in the real world, but I would like to be a bit more convinced of that.

1 comment:

Ivan Kitov said...

I do not have data on CPI and its components for NZ, but I studied this problem for the USA.
One can easily understand and control mid- and long-term changes in relative prices of goods and services. The reason for that is that we observe linear trends in the differences between CPI itself and its various components such as housing, food, energy, transportation, education, medcare, etc. So to say, the differnce between core CPI and headline CPI in the USA is described by the following function
diff=core CPI-CPI= -1.55*t+3118, where t is time (calendar year) with the goodness-of-fit R**2=0.9

Therefore, relative prices, i.e. deviations of individual prices from overall CPI, you are discussing are linear finctions of time only. Fluctuations around the trends are canceled out at very short time as you can see from the figures in my posts:

Thus, the problem is only in description of the CPI evolution over time, not its components.