Inflation fell to 3.8 per cent. Our model flipped to a rate rise anyway.
The June CPI released this morning brought the headline rate down for a third consecutive month. Underneath it, the domestic measures went the other way: services inflation rose, non-tradables inflation rose, and the trimmed mean did not move at all. On this framework’s inputs the pressure score crossed its hike boundary for the first time since we began publishing it — by 0.03, on an assumed threshold, with objections we set out in full below.
The pressure score is approximately +0.33, versus an assumed hike boundary of +0.31 — about 0.03 past it. Two weeks ago the same rule sat below that line. The boundary is a design choice, not an estimated RBA trigger.
The headline fell for a reason that should not reassure the Board
At 11:30 this morning the ABS reported that the Consumer Price Index rose 3.8 per cent over the year to June, down from 4.0 per cent in May, and that prices actually fell 0.1 per cent during the month. That is a third consecutive fall in the annual rate, from 4.6 per cent in March, and the lowest monthly reading since February.
It does not survive disaggregation. The decline came almost entirely from goods exposed to world prices and the exchange rate. Tradables inflation — imported goods, fuel, anything priced offshore — fell to 1.5 per cent, from 2.5 per cent in May and 4.5 per cent in March. Over the same months non-tradables inflation, which is the domestically generated part the Reserve Bank has the most influence over, went the other way: 4.6 per cent in March, 4.7 in April and May, then 4.9 in June. Services inflation rose from 3.7 to 4.0 per cent.
The two underlying measures the RBA cites most did not improve either. The trimmed mean was unchanged at 3.6 per cent. The weighted median rose to 3.7 per cent. So the composite this framework scores — half trimmed mean, a fifth weighted median, and 15 per cent each services and non-tradables — rose from 3.78 to 3.88 per cent even as the headline fell.
Put plainly: a cheaper Australian import bill is doing the disinflation, and domestic price setting is not co-operating. That distinction is the whole reason central banks look past headline CPI.
What the underlying measures actually did
View latest 18 plotted periods as table
| Period | Monthly CPI — headline | Monthly CPI — trimmed mean | Persistent-inflation composite |
|---|---|---|---|
| Jun 2026 | 3.8% | 3.6% | 3.9% |
| May 2026 | 4.0% | 3.6% | 3.8% |
| Apr 2026 | 4.2% | 3.4% | 3.6% |
| Mar 2026 | 4.6% | 3.3% | 3.6% |
| Feb 2026 | 3.7% | 3.3% | 3.7% |
| Jan 2026 | 3.8% | 3.3% | 3.7% |
| Dec 2025 | 3.8% | 3.3% | 3.7% |
| Nov 2025 | 3.4% | 3.2% | 3.5% |
| Oct 2025 | 3.8% | 3.3% | 3.7% |
| Sep 2025 | 3.6% | 3.2% | 3.5% |
| Aug 2025 | 3.2% | 3.1% | 3.3% |
| Jul 2025 | 3.0% | 3.0% | 3.3% |
| Jun 2025 | 1.9% | 2.8% | 2.9% |
| May 2025 | 2.1% | 3.0% | 3.2% |
| Apr 2025 | 2.4% | 3.2% | 3.4% |
What changed in the model, to three decimal places
The framework applies ten fixed weights to ten scored drivers. Nothing about those weights changed today; only the observations did. The composite moved from +0.249 to +0.331, and the movement decomposes cleanly:
- Inflation momentum contributed roughly 53 per cent of the shift. In May, three-month annualised momentum was running below the annual composite, and the driver scored −0.36 — an active argument for patience worth −0.043 of pressure. June monthly core came in at +0.37 per cent, momentum annualised to 3.9 per cent, and the driver is now +0.008. The offset did not shrink. It vanished.
- The inflation level added about 19 per cent, as the persistent composite rose to 3.88 per cent.
- The remainder came from labour, the two-year yield and the currency — the June labour force report, a two-year bond yield that rose to 4.56 per cent, and a trade-weighted Australian dollar whose six-month gain narrowed to +1.6 per cent, weakening its disinflationary pull.
The result is a framework in which, of ten drivers, exactly one now argues against a rise — the Australian dollar, at −0.004. In July there were two, and the larger of them was worth ten times that.
We published four numbered cases in July. June landed exactly on one of them.
On 11 July this framework published four possible June-quarter outcomes, computed from the March-quarter index before the data existed. One of them read: a 0.6 per cent quarterly rise, giving 3.9 per cent annual inflation.
The June quarter came in at 0.6 per cent quarterly and 3.9 per cent annual. The projected number was exact.
The reading attached to that number was not. We labelled the 0.6 per cent case “clear cooling toward the band — hold, tightening bias likely softened”. The quarterly headline did cool exactly as projected, and the framework still moved to HIKE, because the call is driven by the persistent-core composite rather than the quarterly headline, and those two series separated this month. Getting the arithmetic right and the interpretation wrong is a specific, recorded failure, and it is more informative than the hit.
My read: the honest lesson here is that I attached a policy inference to a headline number, and headline numbers are exactly what this framework is built not to trust. The projection engine did its job. The sentence I wrapped around it imported an assumption — that a cooling headline means cooling underlying inflation — that the June data has now falsified. I have changed how the quarter-end cases are labelled rather than quietly dropping the record, and the previous edition stays online, unedited, at its original URL.
The four strongest arguments against this reading
A call that crosses a threshold by 0.03 owes its readers the case against it, made properly rather than as a disclaimer. These are the objections a competent critic would raise, in the order they would raise them.
1. The composite double-counts domestic prices, so it was always going to rise
The persistent-core composite is 50 per cent trimmed mean, 20 per cent weighted median, 15 per cent services and 15 per cent non-tradables. The first two are two different central-tendency estimators of the same price basket. Services and non-tradables overlap each other heavily — most services are non-tradable by construction — and both are already inside the trimmed mean. So roughly 30 per cent of the composite is a deliberate second helping of domestically generated prices.
That means the composite mechanically rises when domestic inflation rises. Reporting that it climbed from 3.78 to 3.88 per cent while the headline fell is therefore partly definitional, not a discovery. The weighting is a stated design choice — it exists because the Board has said repeatedly that it is watching domestic services inflation — but a reader is entitled to know that the measure was built to be sensitive to exactly the thing it just detected.
2. The measure the RBA actually targets did not move
The trimmed mean is the Reserve Bank’s preferred underlying measure. In June it was 3.6 per cent, unchanged. The weighted median rose 0.1 percentage point, which is inside rounding. Weight only what the RBA itself publishes as underlying inflation and June was flat, not hawkish — and flat, after a headline fall and a negative monthly print, is a reasonable person’s definition of progress. This is the single strongest argument for a hold, and it does not depend on doubting anything in this article.
3. The momentum signal is contaminated by unadjusted data
This is the objection that most damages the specific claim made earlier, so it gets the most space. The trimmed mean and weighted median monthly series are seasonally adjusted. The services and non-tradables monthly series are not. The composite blends all four, then annualises three months of it.
Decompose the May-to-June change and the problem is plain: the seasonally adjusted block fell 0.07 percentage points, while the unadjusted block rose 0.315. Unadjusted monthly services over the last seven months run +2.1, −0.7, −0.2, 0.0, +1.0, −0.7, +0.8 — a sawtooth, not a trend. The momentum driver credited with 53 per cent of the shift is being moved by the least reliable series in the input set. Annualising that is close to annualising noise.
The robustness check, and it is the reason this reading still stands. Recompute momentum on the trimmed mean alone — seasonally adjusted, published by the ABS, preferred by the RBA, none of this framework’s weighting involved. Three-month annualised trimmed-mean inflation is 4.07 per cent, against an annual trimmed-mean rate of 3.6 per cent. Underlying inflation is running above its own annual pace on the cleanest series available, which is what “momentum is not cooling” is supposed to mean. The direction survives the objection. The tidy 53-per-cent decomposition does not, and should be read as indicative rather than load-bearing.
4. The margin is smaller than the framework’s own uncertainty
The score clears the boundary by 0.03. That boundary is asserted, not estimated from Board behaviour. Perturbing any single weight by ±25 per cent moves the score across a 0.29–0.37 range, which straddles the threshold. Removing the inflation-level driver alone returns the reading to HOLD. A result that survives 19 of 20 weight perturbations is not fragile, but a 0.03 margin on an assumed line is not a strong claim either, and it should not be reported as one.
What would change this reading
Stated now, with the arithmetic done, so it cannot be quietly revised later.
- The composite clears the boundary by 0.025. The domestic-activity driver currently contributes +0.030. A household spending indicator on 4 August weak enough to cut that contribution to +0.005 or below — a domestic-activity score falling from 0.30 to about 0.05 — returns this framework to HOLD, with nothing else changing.
- A July trimmed-mean monthly print below roughly 0.19 per cent (due 26 August, after the meeting) would pull three-month annualised trimmed-mean inflation beneath its own annual rate and remove the momentum argument entirely. At 0.20 per cent it stays above, at 0.15 per cent it falls clearly below.
- If the Board holds on 11 August, the most likely reason is objection two: the measure it actually targets did not move.
This framework and the market now disagree
No single reading — including this one — should be taken in isolation. The standard discipline is triangulation: compare a model reading against market pricing and mechanical benchmark rules, and be explicit when they diverge.
They diverge now. ASX 30 Day Interbank Cash Rate Futures implied about a 78 per cent chance of no change at the August meeting as of 28 July 2026 — the most recent settlement available, and one that predates this morning’s CPI release. That timing caveat matters and cuts against the significance of the disagreement: the market has not yet had a session to respond to the June data. The comparison is published because it is checkable on 11 August, not because a small framework should be assumed right and a deep futures market wrong.
Implied from 30 Day Interbank Cash Rate Futures as of 28 July 2026: 78% no change, 22% a change at the next meeting. This is a genuine market-implied probability (the feed does not separate hike from cut). It currently disagrees with our HIKE lean — the clearest disagreement on this page, and the one most easily settled by the outcome.
A fixed-coefficient Taylor benchmark mechanically prescribes a higher rate while inflation sits this far above target. We report the gap rather than hide it: the ten-driver framework reads HIKE because momentum, expectations and transmission evidence are scored alongside the raw inflation gap rather than being subsumed by it. Experimental benchmark only.
A directional proxy built from RBA government-bond yields, not an extracted cash-rate path. It leans HOLD on shares of 43% hike / 55% hold / 2% cut. The 2-year yield has risen, which is why this proxy contributes hike-side pressure to the composite even while its own lean stays HOLD — a lean is the largest share, not a verdict.
Uncalibrated pressure score +0.33 against an assumed +0.31 hike boundary, implying one 0.25 percentage point move from 4.35%. The score shares behind the benchmark leans above are normalised heuristics, not probabilities; only the ASX row is market-implied.
Market pricing source: ASX RBA Rate Tracker. Benchmark leans are computed by the same pipeline and carry the statuses shown in the data file.
Sixteen scenarios test both sides of the decision
The same weights and thresholds are applied to every stress test: 6 HOLD, 7 HIKE and 3 CUT. The nearest hike state is Oil shock, round two; the nearest cut state is Household balance-sheet squeeze, which sits 11th on current similarity. Each scenario is a deterministic sensitivity, not a probability-weighted forecast; the live baseline beneath the library shows today’s observed state on the same axes for comparison.
One ranking result deserves flagging rather than burying, because read quickly it looks like it contradicts this article. The state today’s data sits closest to is High inflation, momentum cooling — which is not an endorsement of cooling momentum. The similarity index is 55 per cent distance from the current levels and 45 per cent agreement with the observed direction of travel, and this state matches on levels while explicitly failing on direction: its own evidence line records that current persistent-core momentum is re-accelerating rather than cooling. It is the nearest neighbour by position, and the framework marks it inactive on the evidence. That divergence between a close level-match and a failing direction-match is the whole story of this month.
Sixteen scenario models, side by side
Every column uses the same ten weights and the same decision thresholds.
- 01+0.07Soft landing on trackInflation↓ 3.3%Unemployment↑ 4.5%Wages↓ 3.2%Cash-rate reaction→ 4.35% HOLDCash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHOLD4.35%
- 02-0.21Disinflation + labour cracksInflation↓ 2.9%Unemployment↑ 4.8%Wages↓ 2.9%Cash-rate reaction→ 4.35% HOLDCash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHOLD4.35%
- 03+0.14High inflation, momentum coolingInflation↑ 4.0%Unemployment→ 4.4%Wages↓ 3.2%Cash-rate reaction→ 4.35% HOLDCash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHOLD4.35%
- 04+0.20False dawn: lower level, hotter momentumInflation↓ 2.9%Unemployment↓ 4.3%Wages↑ 3.4%Cash-rate reaction→ 4.35% HOLDCash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHOLD4.35%
- 05+0.04Supply-side disinflation, jobs intactInflation↓ 2.9%Unemployment→ 4.4%Wages↑ 3.4%Cash-rate reaction→ 4.35% HOLDCash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHOLD4.35%
- 06+0.24Inflation sticky, activity weakeningInflation↑ 4.0%Unemployment↑ 4.8%Wages↑ 3.8%Cash-rate reaction→ 4.35% HOLDCash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHOLD4.35%
- 01+0.69Overheating boomInflation↑ 4.0%Unemployment↓ 4.1%Wages↑ 4.4%Cash-rate reaction↑ 4.60% HIKECash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHIKE4.60%
- 02+0.43Oil shock, round twoInflation↑ 4.3%Unemployment↑ 4.5%Wages↑ 3.4%Cash-rate reaction↑ 4.60% HIKECash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHIKE4.60%
- 03+0.32StagflationInflation↑ 4.2%Unemployment↑ 4.9%Wages↑ 4.0%Cash-rate reaction↑ 4.60% HIKECash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHIKE4.60%
- 04+0.52Wage breakout, unemployment steadyInflation↓ 3.8%Unemployment↓ 4.3%Wages↑ 4.5%Cash-rate reaction↑ 4.60% HIKECash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHIKE4.60%
- 05+0.39Terms-of-trade boom with domestic spilloverInflation↓ 3.4%Unemployment↓ 4.2%Wages↑ 3.8%Cash-rate reaction↑ 4.60% HIKECash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHIKE4.60%
- 06+0.56Domestic activity reaccelerationInflation→ 3.9%Unemployment↓ 4.2%Wages↑ 3.9%Cash-rate reaction↑ 4.60% HIKECash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHIKE4.60%
- 07+0.38AUD depreciation shockInflation→ 3.9%Unemployment↑ 4.5%Wages↑ 3.5%Cash-rate reaction↑ 4.60% HIKECash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointHIKE4.60%
- 01-0.49Recession scare / global risk-offInflation↓ 2.4%Unemployment↑ 5.3%Wages↓ 2.2%Cash-rate reaction↓ 4.10% CUTCash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointCUT4.10%
- 02-0.33Household balance-sheet squeezeInflation↓ 2.8%Unemployment↑ 4.9%Wages↓ 2.8%Cash-rate reaction↓ 4.10% CUTCash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointCUT4.10%
- 03-0.37External slowdown and commodity slumpInflation↓ 2.6%Unemployment↑ 5.0%Wages↓ 2.7%Cash-rate reaction↓ 4.10% CUTCash ratePersistent-inflation compositeUnemploymentWagesLine-end values: current → assumed · dashed = hypothetical endpointCUT4.10%
Reproduction confirms arithmetic and recorded-source integrity only. It does not establish predictive accuracy. These are deterministic sensitivities, not probabilities, causal estimates or an RBA forecast.
High inflation, momentum cooling
Policy pressure: hold zoneInflation ↑ · Unemployment → · Wages ↓ · AUD ↑
Persistent core stays at 4.0% while annualised momentum cools to 2.0% · unemployment 4.4% · wages 3.2% · AUD +4%.
Illustrative model reaction: HOLD under these assumptions. The level of underlying inflation remains too high, but the leading pace, wages and currency all improve. This tests whether the Board waits for the better flow data to appear in the annual rate.
The 4.6% comparison is a model assumption, not an official trigger. Source and RBA uncertainty caveat.
Live baseline · ten-driver policy model
Today’s ten inputs, unchanged
Every stress test above changes the displayed macro state and five explicit context drivers, then runs the same weights and hold thresholds used for this baseline.
31/31 source and derivation checks passed
The labour market did not weaken. More people turned up looking for work.
The June labour force release on 23 July is easy to misread, and the tempting misreading is the bearish one. Unemployment rose to 4.43 per cent, with 12,700 more people unemployed than in May; underemployment rose to 6.5 per cent and underutilisation to 10.9 per cent. Taken alone, that is a softening job market.
It is not, and the reason matters. Employment rose 76,300 in the month, the employment-to-population ratio reached 64.0 per cent, and the participation rate jumped 0.3 percentage points to 67.0 per cent. When participation climbs that fast, the unemployment rate can rise while the labour market is getting tighter, because the extra people entering are counted as unemployed until they are matched. A supply-led rise in unemployment is not the same signal as a demand-led one, and only the second would argue for easing.
The framework reads that combination as marginally hawkish rather than dovish — the labour block contributes +0.011, up from +0.001 in May, because strong employment growth and a rising employment-to-population ratio outweigh the higher unemployment rate inside the block. That is a small number either way, and it is deliberately not treated as an easing trigger.
Wage data has not moved at all: the Wage Price Index is still 3.3 per cent from the March quarter, and the June quarter figure does not arrive until 19 August, eight days after the decision. The Board will decide without it, and so does this framework — the wages driver is scored off data that is now four months old, which is a real weakness in any reading published this week, including ours.
Does the RBA really look at these ten inputs?
Broadly, yes. Mechanically, no. The RBA has no published rule saying these ten variables are required or that they receive fixed weights. Its mandate is price stability and full employment, and staff brief the Board on Australian and global conditions, financial markets, forecasts and risks. The June decision itself discussed inflation, expectations, labour, demand, investment, credit, bond yields, the exchange rate, commodities and global risks.
Inflation level, persistence and expectations
Direct mandate evidenceCentral to the 2-3% inflation objective and the outlook for inflation. The level, momentum and expectations inputs sit within one 45% inflation block to make the overlap explicit.
Labour market and wages
Direct mandate evidenceRelevant to full employment, capacity pressure and services inflation. The unemployment-gap input is only a proxy; the RBA uses a broad labour information set and says full employment is not directly observable.
Domestic demand and capacity
Direct outlook evidenceConsumption, investment, GDP and capacity conditions shape the inflation and employment outlook. The domestic-activity score compresses several releases into one channel.
Credit, market rates and the Australian dollar
Transmission evidenceThe RBA monitors credit availability, bond yields, funding and lending conditions, and the exchange rate. These indicators help assess whether policy is transmitting; none is decisive alone.
Commodities and global risks
Conditioning evidenceEnergy and commodity shocks affect inflation, income and activity. The commodity input is deliberately small and cannot represent the RBA’s full global forecast.
Forecasts, liaison and Board judgement
Not reproducedThe RBA also uses staff forecasts, business and community liaison, fiscal and public-demand analysis, productivity, detailed lending conditions, asset markets and judgement. These are material omissions from the score.
Assessment based on the RBA’s monetary-policy framework, June 2026 decision, Statement on Monetary Policy process and published description of Board briefings.
What could still change this call before 11 August
The reading is conditional on data available at the model lock. The Labour Force (23 July 2026) and Consumer Price Index (29 July 2026) releases have already landed and are carried in the score above. 2 scheduled releases remain before the meeting:
- International Trade Price Indexes, AustraliaJune 2026 · external context only
- Monthly Household Spending IndicatorJune 2026 · domestic activity / consumption
The monthly household spending indicator on 4 August is the one that can still move a scored driver materially. A weak consumption print would pull the domestic-activity contribution down from its current +0.030 and could return the composite below the +0.31 boundary. The score sits 0.03 above that line, which is not a comfortable margin.
Working the other way: the Board receives a full Statement on Monetary Policy forecast round at this meeting, and it has now seen two consecutive months in which services and non-tradables inflation accelerated.
Method, reproducibility and limits
The reproduction panel beneath the scenario library is the audit trail: every scenario score recomputes from the published weights, and every input traces to a hash-verified official file — 31 of 31 integrity checks pass on this run. Those checks establish data lineage and arithmetic consistency. They do not establish forecast accuracy, which only the public ledger of meeting outcomes can.
Formally, each driver i is mapped to a bounded score si ∈ [−1, +1] from its official observations, and the composite is the weighted sum S = Σ wi·si with fixed weights that sum to one (inflation block 45%, labour and wages 33%, activity 10%, financial and external conditions 12%). The decision rule is: HIKE if S > +0.31, CUT if S < -0.31, otherwise HOLD. Today S = +0.33.
The framework is not a statistically calibrated probability model or a causal estimate of RBA behaviour. Its weights, transformations and ±0.31 hold band are design assumptions, stated before the outcome and held fixed between meetings. TimesFM is a separate appendix experiment and does not enter this score.
Each driver weight is changed by ±25% and all weights are renormalised. Scores range from 0.29 to 0.37.
The published hike boundary is not estimated from historical Board decisions. The score is 0.33; a boundary of +0.34 or higher would return this reading to HOLD.
Removing Inflation level returns the reading to HOLD, identifying a decisive modelling choice.
The August call is preregistered, but there is not yet an out-of-sample meeting result. Predictive skill is therefore unestablished.
Where this method sits in the literature
Scoring official data into a policy-pressure reading is a monetary-policy reaction function — a well-studied idea, not an invention of this site. Situating the framework honestly: it is a fixed-weight, preregistered variant of that tradition, far simpler than the models central banks actually run.
- Taylor, J.B. (1993), “Discretion versus policy rules in practice”, Carnegie-Rochester Conference Series on Public Policy — the origin of mechanical benchmark rules; our adapted Taylor comparison descends from this.
- de Brouwer, G. and Gilbert, J. (2005), “Monetary Policy Reaction Functions in Australia”, Economic Record — estimated RBA reaction functions; unlike that work, our weights are asserted, not estimated, which is why we label the score uncalibrated.
- Orphanides, A. (2001), “Monetary Policy Rules Based on Real-Time Data”, American Economic Review — why rules evaluated on final-vintage data flatter themselves; the reason our historical checks are labelled final-vintage and our live ledger starts at zero.
- Ballantyne et al. (2019), “MARTIN Has Its Place: A Macroeconometric Model of the Australian Economy”, RBA RDP 2019-07 — the full-system model the RBA actually maintains; a ten-driver score is not a substitute for it and does not try to be.
- CAMA RBA Shadow Board, Australian National University — the established precedent for publishing independent, before-the-meeting RBA calls and living with the outcomes.
Conclusion: a close call that resolves in 13 days
On the evidence published up to this morning, the framework reads HIKE to 4.60 per cent on 11 August. The case rests on one observation that is hard to dismiss: the disinflation Australia recorded in June came from tradable goods, while services and non-tradables both accelerated and the trimmed mean stopped falling. On the cleanest series available — seasonally adjusted trimmed mean — three-month annualised inflation is running above its annual rate.
The case against is set out at length above and is not weak. The measure the Board actually targets was unchanged. The composite that did move is weighted toward domestic prices by design. The margin is 0.03 on an assumed threshold. Market pricing disagrees, albeit from a settlement that predates the CPI. And this framework has never yet been scored against a completed meeting outcome.
Honestly stated: this is a lean, not a conviction, and a reasonable economist reading the same releases could publish a hold and be entirely defensible. What this exercise offers is not certainty but a fixed, dated, falsifiable position with its own strongest counterarguments printed alongside it.
That last point resolves on 11 August at 2:30pm. The call, its inputs, its weights and its failure conditions are all published in advance, which is the only property that makes the exercise worth anything.
Common questions about Australia’s next interest rate decision
When is the next RBA interest rate decision?
The Reserve Bank of Australia announces its next cash rate decision at 2:30pm AEST on Tuesday 11 August 2026, at the end of its 10-11 August Monetary Policy Board meeting, alongside a quarterly Statement on Monetary Policy with updated forecasts.
Will interest rates go up in Australia in August 2026?
After the June quarter CPI released on 29 July, our independent ten-driver framework reads the public data as HIKE — a 0.25 percentage point rise from 4.35% to 4.60% — with a pressure score of +0.33 against an assumed hike boundary of +0.31. ASX cash rate futures still priced about a 78% chance of no change as of 28 July 2026, before that CPI release. The framework and the market disagree. This is an uncalibrated framework reading, not a guarantee and not financial advice.
Is underlying inflation in Australia still too high?
The RBA's preferred underlying measure, the trimmed mean, was 3.6% in June 2026 — unchanged from May and still above the 2-3% target band. Its three-month annualised pace is running at about 4.1%, above its own annual rate, which is why this framework reads underlying inflation as not yet cooling. The weighted median was 3.7%.
Did Australian inflation go down in June 2026?
Headline inflation fell to 3.8% in the year to June 2026, down from 4.0% in May, and the CPI fell 0.1% in the month. But the fall was concentrated in imported goods: tradables inflation dropped to 1.5% while non-tradables rose to 4.9% and services inflation rose to 4.0%. Trimmed mean inflation was unchanged at 3.6%.
What is the current cash rate in Australia?
The RBA cash rate target is 4.35%. It was last raised on 6 May 2026, the third 0.25 percentage point increase of 2026, and the RBA held it unchanged at its June meeting.
Could the RBA cut interest rates instead?
A cut remains distant on current data. Of our sixteen stress tests, 3 produce a cut, and each requires unemployment rising toward 5% alongside inflation falling back inside the 2-3% target band. The nearest cut state is a household balance-sheet squeeze, which ranks 11th on the current-similarity list.
Primary sources
- Reserve Bank of Australia - Cash rate target overview
- Reserve Bank of Australia - Upcoming Monetary Policy Board meeting
- Australian Bureau of Statistics - Consumer Price Index, June 2026
- Australian Bureau of Statistics - Labour Force, June 2026
- Australian Bureau of Statistics - Wage Price Index, March quarter 2026
- Reserve Bank of Australia - Monetary policy objectives and decision process
- Reserve Bank of Australia - June 2026 monetary policy decision
- Reserve Bank of Australia - Statement on Monetary Policy process
- ASX - RBA Rate Tracker (30 Day Interbank Cash Rate Futures)
Previous edition, retained unedited as the preregistered July record: Three hikes, then a pause: the public data lean HOLD for the RBA’s August decision (11 July 2026).
Publication and correction policy
Status: independent personal research; not commissioned, sponsored or externally peer reviewed. No employer, RBA, ABS or Australian Government affiliation or endorsement is claimed.
Author and contact: written by Marcello Babbili and published by Not A Tech Guy. Questions, corrections and methodological challenges are welcome: notatechguy@agentmail.to. Substantive challenges that change a published number will be credited in the correction note.
Versioning: this page identifies the observation dates, generation time and model version. Material corrections are made in place with the modified timestamp updated. Historical preregistered meeting calls are retained rather than rewritten after the outcome — the July edition remains online unchanged.
Scheduled review: this reading incorporates the June labour force and June CPI releases. It will be reviewed once the 4 August household spending indicator is ingested, and scored against the actual outcome after 11 August.
