A single inflation figure can hide very different stories, which is why the RBA watches how widely price rises are spread across the basket, and in September it found that breadth was "well above average".

Imagine two economies that both report annual inflation of 4 per cent. In the first, the number is driven by a handful of items, say a jump in fuel and a spike in one or two volatile categories, while most prices are rising modestly. In the second, prices are climbing at a brisk pace across most of the basket, from services to everyday goods. The headline figure is identical, but the second economy has the harder inflation problem to solve.

That is the logic behind what economists call the breadth of inflation. It is one of the less publicised indicators the RBA considers, and it featured in the minutes of the Monetary Policy Board's 28–29 September meeting, at which the cash rate target was raised by 25 basis points to 4.60 per cent.

One number, many prices

The consumer price index (CPI) measures the change in the price of a large basket of goods and services. The headline, or all groups, figure is a weighted average of all those price changes. Australia's monthly headline CPI rose 4.0 per cent over the year to August 2026, and the quarterly headline measure was 3.9 per cent over the year to the June quarter 2026.

Averages are useful, but they can be distorted by large moves in a few items. A sharp rise in petrol prices, for instance, can lift the headline rate even if little else is changing. For a central bank trying to judge whether inflation is likely to persist, the more important question is often whether price rises are widespread.

Three ways to look past the noise

The RBA publishes several measures designed to reduce the influence of volatile items. For the June quarter 2026, they read:

  • Trimmed mean: 3.6 per cent. This measure excludes the items with the largest price rises and the largest price falls each period, then averages the rest. It is often described as underlying inflation.
  • Weighted median: 3.6 per cent. In general terms, this takes the price change in the middle of the distribution, once items are ranked and weighted by their importance in the basket.
  • CPI excluding volatile items: 4.0 per cent. This removes categories that are known for large swings, leaving a broader set of prices.

When the trimmed mean and weighted median sit well above the target, as both did in the June quarter, it suggests the problem is not confined to a few outliers. Both measures were above the top of the 2–3 per cent target range, and well above the 2.5 per cent midpoint.

It is worth keeping the labels straight. The 4.0 per cent figure for August is headline monthly CPI. The 3.6 per cent figure is the trimmed mean, a measure of underlying inflation. The 4.0 per cent reading for CPI excluding volatile items is a separate June-quarter measure.

What the September minutes said about breadth

The September minutes went a step further, looking directly at how many items were seeing sizeable price increases. They recorded that the share of items with prices rising by more than 2½ per cent was "well above average".

One way to read the 2½ per cent threshold is that it matches the midpoint of the target range. If a large share of the basket is rising faster than that, the RBA cannot rely on a few items cooling to bring inflation back to target; many prices would need to slow.

Several other findings in the minutes pointed in the same direction:

  • July inflation data were stronger than expected, particularly for market services.
  • Staff judged the upside surprise more likely reflected domestic capacity pressures than pass-through of Middle East cost shocks.
  • Staff expected underlying inflation in the September quarter to be higher than forecast in August.
  • Unit labour cost growth was above what is consistent with the inflation target, driven mainly by subdued productivity growth rather than strong nominal wages.

Market services are typically driven by domestic costs such as labour, which is why strength there is often read as a sign of home-grown pressure rather than an imported shock.

Why breadth shapes the policy response

A narrow price shock may fade on its own once the affected prices stop rising. Broad-based inflation is harder to shift because it tends to reflect demand running ahead of the economy's capacity to supply goods and services, and it is more likely to influence expectations about future prices and wages.

Governor Michele Bullock made the domestic diagnosis plain at her 29 September media conference: "Inflation is too high and has been driven by domestic capacity pressures." The Board's statement set out the concern that motivated the increase: "The Board remains focused on ensuring that high inflation does not become embedded."

Bullock also gave a simple benchmark for what success looks like. She said quarterly inflation of about 0.6 per cent is consistent with roughly 2.5 per cent annual inflation, which she contrasted with quarterly rates of 0.8 per cent and 1 per cent. She noted that underlying inflation had been running at about 3.5 per cent over the past six months.

The next CPI update is scheduled for 28 October 2026, ahead of the next cash rate decision at 2.30 pm on 3 November 2026. When those figures arrive, the headline number will attract the most attention, but the underlying measures, and any sign of whether price rises are becoming more or less widespread, are likely to be just as relevant to how the RBA reads them.

The bottom line

Breadth is about how many prices are rising quickly, not just how fast the average is moving. In September, the RBA found that the share of items rising faster than 2½ per cent was well above average, underlying measures were running at 3.6 per cent, and staff attributed the upside surprise mainly to domestic pressures. Together, those signals help explain why the Board treated inflation as a broad problem rather than a temporary spike.