Have Australian workers' pay packets recovered from the inflation of recent years? According to an RBA board member's speech, the answer depends on which wage measure you look at.

Have Australian wages caught up with prices? It is a question that sits beneath a lot of household frustration, and one Governor Michele Bullock acknowledged directly at her 29 September media conference: "Every household has seen how the price of everything has gone up in recent years," she said. "Pay packets don't go as far as they used to".

A week earlier, Monetary Policy Board member Iain Ross AO offered a more technical answer in a speech at the University of Melbourne. Real average earnings, measured through the national accounts (known as AENA), have recovered and remain above their pre-pandemic level. But real Wage Price Index measures have declined relative to pre-pandemic.

Both statements can be true at the same time. Understanding why says a lot about how wages are measured, and about what the RBA looks at when it thinks about the link between pay and inflation.

Two measures, two different questions

The two measures Ross referred to are designed to answer different questions.

The Wage Price Index, or WPI, is built to track the price of labour. Conceptually, it follows the pay for a fixed set of jobs over time, holding the quality and quantity of work constant. If the rate paid for a given job rises by a certain amount, the WPI captures that. What it deliberately leaves out are changes in the mix of jobs across the economy, promotions, and shifts in hours or bonuses.

Average earnings in the national accounts take a broader approach. In general terms, the measure divides total employee compensation by the number of employees. That means it picks up everything the WPI strips out: people moving into higher-paid roles or industries, changes in hours worked, bonuses, overtime and other shifts in the composition of the workforce.

"Real" in both cases means the figures are adjusted for inflation, so they indicate what earnings can buy rather than their dollar value.

Why the two could diverge

Once the differences are clear, the divergence becomes easier to interpret, although the headline comparison alone does not pin down a single cause.

If the pay rate for a typical fixed job has not kept pace with prices since the pandemic, real WPI measures would fall, which is what Ross reported. At the same time, if the workforce has shifted in ways that lift average pay, such as more people in better-paid roles or higher-paying sectors, or other compositional changes, average earnings per employee can rise faster than the rate for any given job. That would allow real AENA to recover and move above its pre-pandemic level even as real WPI lags.

For an individual worker, the practical experience may resemble either measure. Someone who has stayed in the same role may feel the WPI story more acutely. Someone who changed jobs, took on more responsibility or moved into a different industry may have had an experience closer to the average earnings measure. Bullock's observation that "there is some evidence that there are more multiple job holders than there used to be" is a further reminder that aggregate figures can mask very different individual situations.

The long-run anchor: productivity

Ross placed these measures within a longer frame. Over the long run, he noted, real earnings and labour productivity tend to move together, both rising at an average of around 1% a year.

That relationship is central to how economists think about sustainable real wage growth. If workers produce more per hour, businesses can pay more in real terms without having to raise prices to cover the cost. If productivity stalls, real wage gains become harder to sustain without adding to inflation.

That is why the Governor's comments on productivity carry weight in this debate. "Productivity growth remains weak," Bullock said, and more bluntly: "productivity is doing nothing." When productivity is flat, the room for broad real wage gains is limited, and attempts to catch up on lost purchasing power risk feeding into business costs and prices.

What it means for the wage-price spiral question

Ross's speech was titled "A Wage-price Spiral: What are the Chances?" The divergence between the two earnings measures is relevant because a spiral requires wages to respond to past inflation in a way that then drives further price increases.

The fact that real WPI measures remain below pre-pandemic levels suggests pay rates for given jobs have not been racing ahead of prices. Combined with features of today's system that Ross highlighted, including that only around 10% of enterprise-agreement employees renegotiate each year, an average nominal term of three years for agreements lodged over the year to March 2026, and union membership of about 13%, it supports his conclusion that there is little evidence of a wage-price spiral risk under current institutions.

Bullock was equally clear: "There is no wage price spiral going on in Australia."

That does not mean inflation is contained. The Monetary Policy Board lifted the cash rate target by 25 basis points to 4.60% on 29 September, citing stronger-than-expected inflation outcomes, cost pressures on firms, higher global energy prices and "pressure on domestic capacity". Headline inflation was 4.0% over the year to August on the monthly CPI indicator, and Bullock said underlying inflation has "been about 3.5 per cent for the past six months". Both sit above the 2–3% target.

The implication for households is uncomfortable. Inflation above target erodes real pay further, while higher interest rates add to the cost of servicing debt. Bullock said scheduled mortgage payments are "back up sort of around 10 per cent" of household disposable income. The RBA's argument is that bringing inflation down is what ultimately protects the purchasing power of wages, even if the path there is difficult. "I know this decision is difficult for households with a mortgage and businesses," she said.

The bottom line

Whether real wages have recovered depends on the measure. Real average earnings from the national accounts are above their pre-pandemic level, while real Wage Price Index measures are below it. The first captures shifts in the composition of the workforce; the second tracks the pay rate for fixed jobs.

For the RBA, the combination points in two directions. Weak real WPI outcomes suggest wages are not driving a spiral. But weak productivity limits how quickly real pay can sustainably improve. The next CPI release on 28 October, and the Board's decision at 2:30 pm AEDT on Tuesday 3 November 2026, will show how the inflation side of that equation is evolving.