The Monetary Policy Board meets on 4 to 5 May. The cash rate sits at 4.10 per cent after two consecutive 25 basis point hikes in February and March, the first reversals of the easing cycle that ran through 2025. Markets are pricing roughly a 60 per cent chance of a third hike. The case for or against another move depends on what the data say about the underlying inflation problem, not on what borrowers would prefer to hear.
The framework set out in Inflation: Causes, Diagnosis and Cures is a useful discipline here. Diagnose first. Decide second. Skipping the diagnostic step is how public commentary about monetary policy generates more heat than light. So let me work through the diagnosis using the five drivers the framework identifies, then turn to the policy question.
The headline is misleading
Before getting to the substance, it is worth being clear about which headline number we are talking about. The monthly headline CPI series sits at 4.6 per cent annual on the most recent print.
The quarterly headline series sits at 4.05 per cent. Both are headline measures of the same underlying thing.
The gap between them is itself diagnostic information. The monthly series uses a smaller price collection sample and is more vulnerable to single-category swings showing up directly in the headline number. The quarterly series averages across three months of price collection and applies more comprehensive methodological adjustment. When the monthly print sits well above the quarterly, as it does now, it often indicates that the most recent month is being driven disproportionately by a small number of large category movements rather than by broad-based pressure across the basket.
Headline CPI on the quarterly basis is therefore the more reliable starting point at 4.05 per cent annual. Much of the gap above target looks like a supply shock. Monetary policy cannot directly offset supply shocks, and should look through their first-round effects while remaining alert to persistence and second-round transmission. The March monthly fuel print of plus 32.8 per cent captured the peak of a Hormuz-driven oil shock that has already substantially unwound. Petrol terminal gate prices have fallen from around 250 cents per litre to 179, partly through a 32 cent excise reduction and partly through underlying price softening. The April CPI fuel print will mechanically reverse a meaningful chunk of the March contribution.
Electricity at 25.4 per cent annual looks similarly alarming until you look at the monthly bars. They are bouncing between large positive and large negative values as government rebates cycle on and off. The annual rate is a base-effect artifact rather than a sign of ongoing pressure.
This is exactly the kind of noise the trimmed mean is designed to strip out. The quarterly trimmed mean sits at 3.5 per cent and the monthly trimmed mean at 3.3, both well below the headline measures and much closer to each other than the headline measures are. That ordering, with monthly headline highest, quarterly headline lower, monthly trimmed mean lower again, and quarterly trimmed mean close behind, is what you would expect when the recent shock is concentrated in a few volatile categories that the trimming process and the quarterly averaging both partly remove. Anyone using the monthly headline number as their primary signal is being misled by their preferred number.
Commodities more generally remain a supply shock. The Bloomberg Commodity Index, a broad-based benchmark covering energy, metals, and agricultural commodities, is up at the moment. In the past this index has been correlated with periods of higher inflation.
The core is the worry
The trimmed mean has gone from 2.7 per cent in mid-2025 to 3.5 per cent now. The latest three quarterly prints are 1.02, 0.93, and 0.81 per cent. None is large in isolation. The pattern is what matters. The quarterly pace has been above the 0.6 per cent rate consistent with target for three quarters running. That is a momentum statement, and momentum is what the trimmed mean exists to reveal.
The breadth picture says the same thing in a different way. The distribution of expenditure-class quarterly growth shows the median back above the target-consistent line and the box widening. More categories are drifting upward together than was the case six months ago. The framework's strict definition of inflation is a sustained increase in the general level of prices. A few categories doing extreme things is a relative price story. Many categories drifting upward together is an inflation story.
Non-tradables inflation, which captures domestically generated price pressure, sits at 4.63 per cent annual. Services inflation is 3.6 per cent. Both have stayed sticky for the entire post-pandemic period. The Q4 2025 national accounts showed annual GDP growth of 2.6 per cent against the RBA's own estimate of potential output growth of around 2 per cent. The economy is running roughly half a percentage point above its inflationary speed limit. The RBA's February SMP confirmed the diagnosis directly, noting that capacity pressures are judged to be greater than previously anticipated and that the strength and breadth of the recent inflation pick-up is itself evidence of those pressures intensifying.
Transmission risk through diesel
The framework treats institutional transmission as one of its five drivers but focuses on the wage-setting channel that converted 1970s shocks into entrenched inflation. Australia in 2026 has looser transmission than 1976 on that dimension because formal wage indexation is gone and product markets are more open. Wages at 3.44 per cent are elevated but stable, with no sign of acceleration. That is genuine good news.
But there is another transmission channel that the framework underplays and that the current episode brings to the foreground. The international shortage in diesel can be seen in the difference in price for a barrel of crude oil compared with the per-barrel prices for petrol and diesel.
Diesel terminal gate prices are still at 225 cents per litre, well above the 155 they sat at before Hormuz. The excise reduction applies to diesel too, but the unwinding has been much less complete than for petrol because the global supply situation is still pressuring diesel markets. And diesel is not a final consumer good. It is a widespread input cost. Trucks, freight rail, agricultural production, mining, construction equipment, and intercity logistics all run on it.
When diesel goes from 155 to 225 cents and stays there, every business that depends on diesel-intensive supply chains faces a higher cost base. That cost shows up in their pricing decisions over the following months as contracts roll over, freight surcharges get renegotiated, and supply chain costs get passed through. The transmission is mechanical in direction rather than institutional, though variable in magnitude depending on margin compression, competitive intensity, and demand conditions. It does not require anyone to be making decisions in an inflationary frame of mind. It just requires the input cost to be real and the firms paying it to have at least some pricing power.
This transmission is also harder for the Bank to look through than the petrol equivalent. Petrol shows up as a single CPI line item that the Bank can identify, attribute, and discount. Diesel transmission is diffuse and lagged. It shows up as broad-based goods and services price drift across many categories. Food prices rise because farm input costs are higher and freight to distribution centres costs more. Building materials cost more because trucks delivering them and equipment moving them on site cost more to run. Services with mobile components face higher operating costs. None of these will look obviously diesel-driven in the CPI breakdown. They will look exactly like the kind of broad-based drift the trimmed mean is supposed to capture.
Expectations are the swing factor
Long-run 10-year breakeven expectations sit at 2.28 per cent, comfortably anchored within the band. The Bayesian composite estimate of inflation expectations sits at 2.72 per cent, near the top of the band but still inside it. The short-run 1-year measure has drifted up to 3.02 per cent, just outside the band. This is exactly the configuration the framework warned about. Long-run expectations are holding because the market still believes the RBA will act before things get out of hand. Short-run expectations are drifting because the recent inflation reads are giving people reason to expect more of the same.
A drifting core is what causes short-run expectations to drift, because firms and households extrapolate from the prices they actually pay. The credibility built up since the 1990s reform is a stock that gets drawn down if the Bank is seen to be tolerating drift. The case for acting in May is partly a case for demonstrating that the drift is being taken seriously while it is still small enough to be addressed cheaply.
Putting it together
The current episode is not a pure supply shock that should be looked through. It is not a pure demand episode requiring textbook Taylor rule tightening. It is not yet an expectations problem requiring Volcker-style aggression. It is a combination, and it has a specific character. The headline noise is going to make diagnosis harder over the next two or three CPI prints, not easier. The fuel reversal will look like vindication for holding rates. The underlying transmission of the diesel shock through input costs will continue working through, mostly invisibly in the headline number but showing up in the trimmed mean and in non-tradables.
The asymmetry of error argument from the framework still applies but it is doing work at the margin rather than dominating the question. The cost of being wrong by hiking is a slightly deeper slowdown in interest-sensitive sectors. The cost of being wrong by holding is letting the expectations situation drift further. Neither is catastrophic in the May timeframe specifically. Both become more serious if the wrong call is repeated meeting after meeting.
The lag structure adds genuine uncertainty in both directions. The two hikes already delivered are barely in the system yet, which argues for waiting to see how they land. But monetary policy operates with twelve to eighteen month lags, which means whatever the Bank does in May will not fully affect inflation until mid-2027, and waiting for the diagnosis to become unambiguous means acting a year too late. The Bank has more information than any outside commentator, and the genuine 5 to 4 split on the March vote shows the Board itself is divided.
My read is that a hike is more justified than not, but only just. The trimmed mean drift over three quarters, the breadth of price increases across the consumption basket, the demand component running above capacity, the short-run expectations already outside the band, and the transmission risk from diesel input costs collectively tilt the balance toward acting. The cumulative tightening to date does not look sufficient against this combination. But the case is not overwhelming. The fuel reversal is real, the existing hikes are still flowing through, and a Board that paused in May to assess the next round of data would not be making a serious mistake. The market price of around 60 per cent reflects this balance, and that is probably the right read.
The bigger point is that direction matters more than timing. The Bank probably has not done enough yet, and the question is mostly about when the next move comes rather than whether. May, July, or August would all be reasonable times to add the next 25 basis points if the data continue to point the way they are pointing. The Bank moved to eight meetings a year partly to give itself this kind of optionality. The thing that would be wrong is reading the trimmed mean at 3.5 per cent as “still in single digits, no problem” and letting another two or three quarters drift past while the diesel transmission continues its work. That is the failure mode the framework was written to help us avoid.
My call
A 25 basis point hike in May is slightly more justified than holding. Cutting is not on the table. Moving by more than 25 would be overcorrection. The marginal decision is whether to act now or wait one meeting, and on balance acting now is the better choice.
I wouls raise rates to 4.5%. you just cannot allow a suppkly sside shock leads to companies raise prices to recover ccosts rather than becoming more efficient.
ReplyDeleteAnother great article.