On Thursday the Reserve Bank decided to leave interest rates unchanged but signalled that, due to the apparent resilience of the economy it was preparing to remove its Canterbury earthquake insurance.  The Reserve Bank decreased the official cash rate (OCR) in March from 3.0 to 2.5%.  The implication of Thursday's announcement is that the Bank is very likely to increase the OCR back to 3.0% at their next monetary policy statement on 15 September.

But once we get the "insurance" distraction out of the way what might give an indicator of what the Bank might do next?  A key line from the Bank's statement on Thursday gives us a big clue:

"Wage and price setters should focus on underlying inflation, which is currently estimated to be below 2.5 percent."

The Reserve Bank has recently refined its measure of underlying inflation (see Reserve Bank Discussion Paper DP 2010/13 by Michael Kirker).  Although difficult to observe and measure, core inflation is the critical concept that matters most for monetary policy.

The majority of price movements in an economy reflect changes in how scarce goods and services are.  Economists call these movements relative price changes.  For example, the large recent increases in tomato prices reflect that it is winter, so there are a lot fewer tomatoes available.  Addington this were the floods in Queensland, which destroyed a major source of the supply of fresh tomatoes for the New Zealand market.  It is now far harder to obtain fresh tomatoes so the price goes up.

This type of price rise is not inflation.  It is just an indication that tomatoes are harder to get hold of.  If your love of fresh tomatoes is so great that you must have them at any price, then this price rise will reduce your ability to spend money on other goods.  But for the majority of us the increase in tomato prices means that we simply do not buy fresh tomatoes at the present time “we put off making Greek salads until the summer and use tinned tomatoes for cooking.

This illustration highlights that the CPI is not a perfect measure of the general price level.  The CPI is based on the basket of goods purchased by the typical New Zealander.  But what are considered to be typical purchases is revised infrequently, say every five years.  This means that it will not account for people's natural responses to relative price changes.  It will measure an increase in prices due to the increase in tomato prices even though the vast majority of us do not experience the price increase, as we choose not to buy tomatoes until the price falls.

The measure developed by Michael Kirker is an attempt to obtain a more accurate measure of the underlying general price movements.  I reproduce this measure of core inflation in the graph.  So what does it show us?  First, the core measure is more stable with far fewer peaks and troughs than the official CPI inflation measure.  This would suggest that the Reserve Bank is probably right to "look through" the current spike up in the CPI.  The analysis of Michael Kirker indicates that the core measure is more closely aligned with domestic "non-tradeable" price pressures.  A lot of the volatility in the CPI stems from international movements in the prices of tradeable goods and services, like our tomatoes.  The implication for the operation of monetary policy is that interest rate settings should reflect and anticipate domestic supply and demand conditions and should generally ignore the initial impact of international price movements.

The graph also shows that the core measure seems to be quite closely aligned with inflation expectations, as measured by the National Bank from their Business Outlook Survey.  This close relationship between expectations with the core inflation measure will perhaps provide the Bank with some comfort that the core inflation measure provides a reasonably meaningful measure of underlying inflation.  If so, the core measure should also provide the Bank with some concerns about the effectiveness of its operation of monetary policy.

To begin with there appears to be an unnerving upward trending the core inflation measure.  In 1995 the local peak in underlying inflation was 2.3%.  The next local peak was at 2.7% in 2002.  The most recent peak in 2008 was at 3.4%.  Yes, inflation pressures have fallen a long way during the current downturn, but inflation expectations did not fall as far as the core measure, instead they have been on the rise for over a year, and now exceed 3%.  This concern is perhaps amplified by the finding of Kirker that initial estimates of the core inflation measure tend to require upward revisions in following quarters.  The implication is that, unless inflation expectations begin to subside, the upward trend in underlying inflation will continue and further interest rate rises are warranted.