WEEKEND ECONOMIST: Outliers

Last week's data releases were dominated by the CPI for the March quarter. The headline number of 0.1 per cent was well below our forecast of 0.8 per cent but this discrepancy was entirely due to the item "deposit and loan facilities". That component fell by 14.1 per cent – a staggering move in the quarter for any component of the CPI, contributing 0.66ppts of the 0.7ppts by which we missed with our forecast.

Naturally that component did not figure in the Trimmed Mean CPI (one of the RBA's two preferred measures of underlying inflation) which gives the measure after eliminating those items that explained the top and bottom 15 per cent of moves after adjusting for weights. The Trimmed Mean actually registered higher than our forecast (1.0 per cent vs 0.8 per cent) and the fair conclusion from the release is that underlying price pressures were a little stronger than we had anticipated despite the plunge in the headline measure.

The combination of the lagged effects of the 30 per cent fall in the AUD/USD through the second half of 2008 and reduced discounting, possibly in anticipation of the spending of fiscal stimulus packages, seems to explain most of the slippage.

However, we have maintained our view on monetary policy for the remainder of the year.

True, with a 1.1 per cent read for the March quarter (average of the trimmed mean and weighted median CPI) it will now be difficult for the RBA to achieve its 3 per cent annual underlying inflation forecast for 2009. That will require an average of around 0.6 per cent per quarter (down from 1.1 per cent) for the remainder of the year.

Certainly the lagged effect of the currency depreciation will fade and the slowdown in wage and demand pressures will see the underlying inflation rate moving in the right direction. Certainly housing costs (the largest component of the CPI) and holidays (down 4.5 per cent) are pointing to powerful downward pressures associated with demand that can be expected to remain a big drag on inflation through 2009.

Probably the most important aspect of the CPI result from a policy perspective was the plunge in the "deposit and loan facilities" component. That component was incorporated in the CPI in 2005 to try to capture the interest rate effect on inflation without the interest rate feeding directly into the index.

This is an unusual measure which the Bureau of Statistics has developed to measure the 'price' of retail financial services. Essentially the methodology is to sample banks deposit and loan rates and take a mid point. The price of deposit services is assessed as the difference between the mid point and deposit rates.

If the spread between loan and deposit rates narrows it means that either the price of deposit services has fallen or the price of loan services has risen or, as was the case in the March quarter, a substantial change in the relativities (spread contraction).

The staggering 14.1 per cent fall in this item in the March quarter indicates that the extent of the fall in inflation (annual inflation fell from 3.7 per cent to 2.5 per cent) was largely due to the narrowing of the loan/deposit spread. The key dynamic would have been the full pass through of the 100bp RBA rate cut in February to mortgages and other personal loan products which was not matched by a comparable cut in retail deposit rates.

This is not the first time this series has behaved strangely (up by 9.5 per cent in Q2 2008) although the average move prior to Q1 had been 1.2 per cent and the biggest fall since it was included in the CPI was –2.1 per cent. The large move, however, does emphasise the challenge banks are dealing with by having to compete for retail deposits while keeping families happy on the mortgage rate side in a period of falling RBA rates.

This damaging collapse in the retail spreads for banks provides a persuasive explanation as to why the pass through of the latest 25bp rate cut by the RBA was so muted – banks' retail spreads are under pressure.

Readers will recall that a key rationale for our choice of 2 per cent as the low point for the cash rate in this cycle was the level of rates below which the banks would not pass on any further reductions to mortgage rates. It was predicated on the view that banks would want to be competitive in the retail savings market and would be reluctant to push deposit rates below a certain level – the level of deposit rates consistent with a 2 per cent cash rate seemed a reasonable floor. However the evidence from the CPI indicates that banks may already have reached the point where margins need to be protected and retail deposit rates must be held up.

The bank pass through should NOT be the determinant of how far the RBA can cut. Around 50 per cent of banks' assets are direct housing loans while around 30 per cent of loans would be linked to the bank bill rate. Even with no adjustment to mortgage rates, 'bank bill' borrowers will derive a benefit from lowering the cash rate. Banks themselves fund around 45 per cent of liabilities out of the bank bill/floating rate swap market and any fall in bank bill rates will benefit funding costs.

With bank funding costs currently rising sharply as 'cheap' pre-crisis term funding has to be replaced with 'expensive' post-crisis term funding, relief in the bank bill/floating rate swap market might avert actual increases in other loan rates, including mortgage rates. In short, that rate level below which banks are unable to pass rate cuts through to mortgage rates may not define the low point in the cycle.

That leaves open an even lower rate level to define the lowpoint in the cycle. We are happy to retain our target rate of 2 per cent with downside risks, with the flexibility to fall further not being restricted by the pass through capacity of the banks.

Our target of 2 per cent by Q4 2009 is consistent with our growth profile of a contraction of 1 per cent in GDP in 2009. Thursday, the IMF produced its latest forecast that GDP would contract by 1.4 per cent in 2009 – more pessimistic than our view and weaker than the lowpoint in the previous recession in Australia when the economy contracted by 1.3 per cent in 1991.

A likely quarterly profile that would be consistent with the IMF's forecast would see the Australian economy contracting in the second half of 2009. In contrast our forecast envisages a return to anaemic but nevertheless positive growth in the second half of 2009.

With business investment, employment, inventories and exports still contracting, despite some early stability in new housing construction and consumer spending, there would be ample justification for the RBA to cut rates through the end of 2009. A profile envisaged by the IMF with the overall economy continuing to contract in the second half of 2009 would probably see the RBA cutting below our 2 per cent target.

Bill Evans is Westpac's global head of economics