2026-09-29 → 2026-10-02

The Reserve Bank Board announces its cash rate target at 14:30 AEST, 06:30 CEST. The consensus is unusually tight for a decision that would be the RBA's first hike of this cycle: NAB, CBA and Westpac all expect a 25bp increase to 4.60%, the highest cash rate since 2011, and ANZ expects the same move plus a second in November that would take the rate to 4.85% by year end. The big four banks and market pricing together put the probability of a 25bp move at 80-90%.

The case for hiking is largely imported. Australian annual CPI eased to 3.5% in July from 3.8% in June, and the trimmed mean held at 3.6% - both still above the 2-3% target band, but both moving the right way. What changed is energy: Brent has gone from around US$80 a barrel to above US$100 and traded above US$105 on 28 September, with the Strait of Hormuz disruption showing no sign of resolving, and the RBA has been explicit that the risk it is weighing is petrol and diesel pass-through into an inflation rate that never fully returned to target.

Last Thursday's labour force report made the decision harder rather than easier. Unemployment rose to 4.6% from 4.5%, above the 4.5% consensus and the highest since late 2021, even though employers added 39,500 jobs against the 20,000 expected. The composition was worse than the headline: full-time employment fell 6,300 while part-time rose 45,800, and the participation rate climbed to 67.1% from 66.9%. The ABS also flagged that a new Supplementary Survey collection model was implemented in August and recommended awareness of its potential effect on the seasonally adjusted series, which gives the Board a defensible reason to discount the tick up.

The scenarios divide on the statement rather than the number. A 25bp hike is so heavily priced that delivering it changes little by itself; what moves AUD is whether the Board leaves November open. A hike with an explicit door open - ANZ's path - should lift the front of the Australian curve and the currency against a US dollar that is itself strong. A hike framed as sufficient, with the energy shock characterised as first-round and temporary, would be sold almost immediately. A hold, at 80-90% pricing, would be a genuine shock: AUD would fall hard and the whole of the tightening premium would move to November.

The sequencing is the detail worth carrying. The Board decides on Tuesday afternoon and the August CPI lands on Wednesday morning, which means it acts without the freshest inflation read and with a labour market that has just softened. Whatever it does, the statement almost has to be hedged - and a hedged statement into a currency positioned for a clean hawkish outcome is the asymmetry to watch.

The statement is published simultaneously with the decision and, given that the decision itself is 80-90% priced, it carries essentially all of the event's information content. The market is not asking whether the RBA moves on Tuesday; it is asking whether Tuesday is the start of something or the end of it.

Three things are worth reading for. First, how the Board characterises the oil shock - as a first-round effect on headline inflation that monetary policy should look through, or as something with second-round momentum into wages and services. Second, whether the language shifts from data-dependence toward explicit conditional guidance about further tightening; ANZ's forecast of a second hike in November only works if the statement leaves that door visibly open. Third, how it treats a 4.6% unemployment rate, since a rising jobless rate is normally the sort of thing that stays a central bank's hand and the Board will have to explain why it did not.

The mechanical consequence for markets is straightforward. With the hike priced, a statement that reads as one-and-done takes the November premium out of the Australian front end and takes AUD down with it, most cleanly against a US dollar carrying a 4.810% 2-year and a Fed with a live meeting on 27-28 October. A statement that keeps November alive does the opposite. The statement was also written before Wednesday's CPI, which limits how firmly the Board can commit in either direction and argues for the cautious reading being the right base case.

The ABS publishes the August CPI at 11:30 AEST on Wednesday 30 September, the morning after the RBA decision. The monthly headline is forecast at 0.5% against 1.0% in July - a halving of the monthly pace, and on its face the most benign line in the release.

It is worth doing the annualisation before calling it benign. A 0.5% monthly print compounds to a little over 6% a year; July's 1.0% compounds to roughly 12.7%. Both are far above anything consistent with a 2-3% target band, and a deceleration from one to the other is a deceleration within a range that should not exist at all. The monthly series is noisy enough that no single print settles anything, but two consecutive months at or above 0.5% would mark the fastest back-to-back pace of the year.

For markets the m/m line matters mainly as the input to the annual rate published alongside it, and as a cross-check on the trimmed mean. An upside surprise here - 0.7% or above - combined with a firm trimmed mean would make Tuesday's hike look like the first of several and would price November hard. A downside surprise, 0.3% or below, would suggest the July spike was a one-month energy artefact already fading, and would take the sting out of the annual number regardless of what that number prints at.

The annual rate is the headline the Australian press will lead with, and it is forecast to jump to 4.1% from 3.5% - 1.1 percentage points above the top of the RBA's 2-3% target band, and a 0.6-point acceleration in a single month.

The arithmetic deserves attention before the alarm does. The annual rate rises while the monthly rate halves, which can only happen through the base: replacing a weak August 2025 month with a stronger August 2026 one. Taking the forecasts at face value, 3.5% plus a 0.5% new month reaching 4.1% implies the month rolling out of the calculation was around -0.1%. That is a base effect, not fresh inflation, and it means a 4.1% print would overstate the current run-rate rather than reveal a new acceleration.

That distinction will not stop it mattering. A 4-handle on Australian inflation, published the morning after the RBA lifts the cash rate to its highest level since 2011, is a political and market event whatever its composition - it validates the hike in the public narrative and puts November squarely on the table. The risk for anyone positioned on that reading is that the RBA, which knows exactly where the base effect comes from, does not treat it the same way.

The scenario that would genuinely change the picture is an upside miss - 4.3% or above alongside a firm trimmed mean - because that could not be explained by the base alone and would imply the energy shock is broadening. A print at or below 3.9% would suggest the reverse, and would leave the RBA looking early rather than late.

The trimmed mean is the line the Board actually watches, because it strips the largest price moves in both directions and therefore removes most of what an oil shock does to a headline index. It is forecast at 0.3% m/m against 0.5% in July.

A 0.3% monthly trimmed mean compounds to roughly 3.7% annualised - still above the target band, but a long way from the 4.1% the headline is expected to print, and consistent with the annual trimmed mean that held at 3.6% in July. If the release comes in as forecast, the shape is unambiguous: headline accelerating on energy and base effects, core decelerating underneath. That is the textbook signature of a first-round supply shock, and it is the single best argument for the RBA treating Tuesday's hike as insurance rather than the start of a campaign.

The alternative is the one that would reprice the Australian curve. A trimmed mean holding at 0.5%, or rising, would say the oil shock has stopped being about petrol and started being about everything - second-round pass-through into services and, eventually, wages. Against a labour market that has just loosened to 4.6% unemployment, that combination would be the most uncomfortable outcome available to the Board: core inflation broadening while employment weakens. Of the three Australian inflation lines on Wednesday, this is the one to read first.

This is the Federal Reserve's preferred inflation gauge and the first reading of it since the FOMC raised the funds target to 3.75-4.00% on 16 September. The August core PCE price index is forecast at 0.3% m/m after 0.2% in July, when the annual core rate held steady at 3.3%.

The reference points are the Committee's own. September's Summary of Economic Projections has core PCE finishing 2026 at 3.4%, and the dot plot carries a 2026 median of 4.1% - one more hike this year - held at 4.1% through 2027. A 0.3% monthly print annualises to roughly 3.7%, which is above the projection path rather than on it. The Committee has told the market it expects one more move; the question this release answers is whether the data will let it stop there.

The scenarios are asymmetric because of what the bond market did last week. With the 10-year having cheapened 21bp to 5.170% while the 2-year added only 5bp, the curve is pricing inflation risk at the long end and leaving the policy path broadly alone. A 0.4% or higher print should therefore extend the bear steepening rather than flatten it - it would confirm the long end's read and put 27-28 October firmly in play. A 0.2% print would do the most damage to the move, because it would suggest the energy shock is not leaking into core and the 21bp of long-end cheapening was an overreaction.

The detail to look past the headline for is the split between goods and services core, and specifically whether anything in services is accelerating. Brent above US$105 shows up in headline PCE mechanically; it only justifies further tightening if it is showing up where the Fed cannot look through it. So far, at a core annual rate that has been stuck near 3.3% for months, it has not been.

The third estimate of second-quarter GDP, forecast unchanged at an annualised 1.5%. The second estimate already put Q2 growth at 1.5%, down from 2.1% in the first quarter, with increases in consumer spending, exports and investment partly offset by lower government spending and higher imports.

This is the lowest-information release on the week's calendar and should be treated as such. The quarter it describes ended three months ago, it has already been estimated twice, and it lands at the same minute as core PCE - which means that in practice it will be read, if at all, as a footnote to the inflation print. Third estimates revise materially only rarely.

What can occasionally matter is what travels with it. This release carries corporate profits, industry-level GDP and state personal income, and the profits data is the one component that occasionally moves equity-sector expectations. There is also a standing question of whether gross domestic income tells the same story as gross domestic product; a wide gap between the two is the kind of thing that gets cited later when a slowdown is being dated, even if it moves nothing on the day.

Published inside the employment report, average hourly earnings are forecast at 0.3% m/m, matching August, when earnings rose 0.3% to US$37.75 and were up 3.1% over the year.

In an energy shock this is the most important second-round indicator available, and more important than the payroll count beside it. The mechanism by which an oil price becomes persistent inflation runs through wages: workers whose real incomes are being eroded bid for compensation, firms pass it on, and a one-off price level shift becomes a rate of change. At 3.1% annual wage growth against a core PCE rate of 3.3%, real wages are currently flat to slightly negative - which is precisely the configuration in which a supply shock stays first-round.

So the number to fear is an acceleration. A 0.4% or 0.5% monthly print, particularly if the annual rate breaks back above 3.5%, would be the clearest evidence yet that the Hormuz disruption is being translated into domestic cost pressure, and it would do more to price the 27-28 October meeting than a strong payroll headline. A 0.2% print, conversely, would let the Fed argue that the energy shock is doing its own disinflating by squeezing real incomes, and would support the case for stopping after one more move.

September payrolls are forecast at 98K after August's 162K. August was a substantial upside surprise - economists had penciled in roughly 53K, making it the strongest monthly gain since March - and it came with 55K of combined upward revisions to June and July that erased July's previously reported job losses.

That report changed the policy path directly. Market-implied odds of a September hike rose from around 52% to 59% on the print, and the Committee hiked on 16 September by a 12-0 vote. This Friday's release is therefore the first clean read on whether the labour market that justified the hike is still intact after it, and it arrives two days after core PCE and roughly three weeks before the 27-28 October meeting.

The consensus at 98K is itself the story: economists are forecasting the pace to fall by nearly 40% in one month, which is a bet that August was an outlier rather than a trend. Given that they under-forecast August by more than 100K, the forecast distribution around 98K should be treated as wide. A repeat beat - 150K or above - would put October firmly in play and, coming after last week's 21bp of long-end cheapening, should extend the bear steepening in 2s10s rather than reverse it.

The more interesting scenario is the weak one. A print below 50K would be the first real evidence that a funds rate at 3.75-4.00% is biting, and it would flatten the curve from the front rather than the back - the opposite of every move of the past fortnight. It would also put the Fed in the position every central bank in this brief is edging toward: tightening into an energy shock while its own labour market turns. Note that the US front end already leaned slightly this way, richening 6bp on Friday into this release.

The unemployment rate is forecast to hold at 4.1%, where it sat in August, unchanged from July and 0.2 percentage points lower than a year earlier.

A stable 4.1% is, for the Fed's purposes, permission. The Committee's dual mandate only binds in one direction when employment is deteriorating, and an unemployment rate that has been flat for two months while falling year-on-year gives the Committee no labour-side reason to stop tightening into an inflation rate that remains above target. As long as this line prints 4.1%, the argument inside the FOMC is an inflation argument and nothing else.

The contrast with Australia is the reason to read the two together this week. Australia's unemployment rate rose to 4.6% in August, the highest since late 2021, and the RBA is expected to hike into it on Tuesday. The US rate is flat at 4.1% and the Fed has already hiked. Two labour markets moving in opposite directions, two central banks tightening into the same oil shock - the divergence is in the data, not in the policy, and it is what makes AUD/USD the cleanest expression of this week's calendar.

The threshold that would change the US conversation is 4.3%. A two-tenths jump in a single month, particularly alongside a sub-50K payroll print, would move the debate from how much more tightening to whether the last hike was a mistake, and would do more to the front end of the US curve than anything else on this week's calendar.