RBA Holds at 4.35% Unanimously (11 August 2026): The SMP Cut Its Assumed Path Only to 4.4% — and Marked 2027 Inflation Up

The Reserve Bank of Australia left the cash rate target unchanged at 4.35% on 11 August 2026, unanimously, for a second consecutive meeting — exactly as a 100% priced market expected. The information was in the document published in the same minute. The May Statement on Monetary Policy had conditioned its return-to-target forecast on 60 basis points of further tightening, to 4.70% by end-2026, a path markets had since priced out entirely. The August SMP closed most of that gap from its own side rather than the market's: the assumed cash rate is now 4.4% at December 2026 and 4.5% through 2027, which the Bank describes as "less than one full cash rate increase by the end of 2026, before a small reduction in the cash rate further out" and "around 25 basis points lower than the path underpinning the May forecasts." Its central forecast still contains a partial hike that market pricing does not. And with 25 basis points taken out of the assumed path, the 2027 inflation forecast went up: headline CPI at June 2027 moved from 2.4% to 2.8%.

That is the reconciliation this post was written to watch for, and it resolved in the direction that says the disinflation was borrowed rather than earned. Near-term headline forecasts fell with the barrel — December 2026 CPI from 4.0% to 3.6%. The 2027 numbers went the other way, and the trimmed mean path in 2027 barely moved at all. AUD/USD was flat at $0.7055 after the announcement; the repricing happened in swaps, where the odds of another increase this year fell to about 40% from roughly 50%.

What actually happened

At 2:30pm AEST on Tuesday 11 August the Board left the cash rate target at 4.35%, and the statement recorded that "today's policy decision was unanimous" (primary source: RBA media release, 11 August 2026). The quarterly Statement on Monetary Policy published in the same minute, with its forecasts finalised on 5 August; Governor Michele Bullock's press conference followed an hour later.

The statement's own words are the cleanest summary of where the Board has landed. Inflation "picked up materially in the second half of 2025," headline inflation "is still too high," and trimmed mean inflation "also remains elevated and is little changed from the March quarter." On the domestic side it acknowledged both of the things that argued for patience: "momentum in the housing market has shifted, with housing prices falling in some capital cities and new housing loans declining noticeably," and "labour market conditions have eased by a little more than expected in recent months."

Then the sentence that does the work, unchanged in substance from June: the Board "will continue to do what it considers necessary to bring inflation sustainably back to target, including increasing the cash rate target further if upside risks materialise." Policy is judged "somewhat restrictive" — not restrictive — and inflation "is not expected to return to around the midpoint of the target range until late 2027 and there are upside risks to this projection."

The market read that as a hold with the option kept open, and priced it accordingly rather than dramatically:

Source: Reuters. The pattern is what a fully priced decision normally produces — nothing in spot, a modest fade in the tail that pricing had assigned to a further increase.

What consensus expected: a hold nobody disputed

The pricing was about as one-sided as Australian rates pricing gets. The ASX 30-Day Interbank Cash Rate Futures contract for August was trading at 95.65 as at 6 August — an implied 4.35%, and therefore a 100% probability of no change. All 37 economists in a Reuters poll expected a hold, with 27 of 36 seeing the cash rate unchanged through the end of the year.

The bank forecasts converged on the way in. Westpac was the last of the big four to move, dropping a call for two further 25 basis point increases; ANZ, the Commonwealth Bank and NAB were already positioned for an extended pause. Westpac chief economist Luci Ellis gave the reason without decoration: "We no longer expect rate hikes by the RBA this year. Inflation has been more benign than we feared."

Some context on the level, because 4.35% is not a neutral setting. It is a twelve-year high, reached in May 2026 after three consecutive 25 basis point increases in February, March and May. The Board then held in June. Its own June statement described the situation in terms that have not been retracted: "Financial conditions have tightened this year in response to three increases in the cash rate target," and, at the end, the clause that has done the most work in Australian rates markets this year — the Board "will do what it considers necessary," including "increasing the cash rate target further if required." That decision was unanimous (primary source: RBA media release, 16 June 2026).

The forecast the August SMP had to reconcile

Here is the whole argument in one table: what the May Statement on Monetary Policy assumed, and what actually happened. The forecasts were finalised on 29 April 2026 (source: RBA, May 2026 SMP; outcomes from the Australian Bureau of Statistics).

Read down the right-hand column and the errors do not all point the same way, which is exactly why this SMP is not a routine revision. Inflation undershot, which argues for a lower path. Unemployment overshot the forecast — 4.4% in June against 4.2% projected — which also argues for a lower path. But the assumed policy rate that was supposed to produce the return to target has been priced out, which argues the other way. And the oil price that drove the inflation undershoot is a global variable the RBA does not control and cannot bank on.

The May document was unusually explicit about the mechanism: its baseline assumed "an increase in interest rates of 60 basis points," and a Middle East resolution in which "shipping flows return to pre-conflict levels in Q4 2026." Both of its most important external inputs — the policy path and the barrel — had moved by August, in opposite directions.

Why the headline undershot: the oil assumption, not the domestic pulse

The June CPI release is where the mechanism becomes visible. Annual headline inflation eased to 3.8% from 4.0% in May, with the monthly CPI actually falling 0.1% and June-quarter momentum at +0.6%, less than half the March quarter's +1.4%. The driver was one line: automotive fuel fell 10.9% in June, after an 11.9% fall in May, which the ABS attributed to lower world oil prices. That dragged the entire transport group's annual rate to +0.1%, from +3.3% in May.

Pulling the other way, electricity was 22.4% higher than a year earlier — which the ABS attributes largely to the expiry of Commonwealth and State Government rebates rather than to any change in the underlying cost of power. And the domestic components did not cool: housing was the largest annual contributor at +6.8%, new dwelling prices rose 5.8%, and rents rose 3.6%.

Which is why the trimmed mean did what it did. Rachael McCririck, the ABS head of price statistics, described it directly: "When we look through some of the bigger price movements, underlying inflation is steady at 3.6 per cent in the 12 months to June 2026." A 10.9% monthly fall in petrol and a 22.4% annual jump in electricity are both textbook candidates for trimming — so the two most dramatic numbers in the release were removed from the measure the Board actually targets, and that measure did not move.

The distinction matters for the currency because imported disinflation is symmetric. Every other oil-importing economy got the same discount at the same time, so a fall in Australian headline inflation driven by the world oil price changes Australia's relative rate outlook far less than the number suggests. We worked through the same release in more detail in Australia's June CPI at 3.8%, and the general version of the point — that the Aussie is a commodity currency in both directions — is set out in what drives the Australian dollar.

Which resolution the Bank chose

Three resolutions were available. Mark the whole inflation path down and drop the conditional-hike clause, conceding that 4.35% plus a softer labour market does the work the assumed 4.70% was doing. Mark down the headline with oil, keep the clause, and leave the underlying path above 3% into 2027. Or keep a slower return to target outright and flag that the disinflation to date was imported.

The August SMP chose the second and leaned toward the third, and the assumption table is where it is visible. Comparing the two documents line by line (sources: May SMP and August SMP, detailed forecast tables):

Read the inflation rows as a pair and the mechanism falls out. The near term got better and the medium term got worse. December 2026 headline came down four-tenths; June 2027 went up four-tenths. That is the signature of a forecast that has banked a favourable fuel shock it did not expect while conceding that the domestic disinflation behind it has not arrived.

Two revisions explain the back end, and both are in the table. The assumed policy path came down 25 basis points, which by the Bank's own model means less restraint in 2027. And the trade-weighted index assumption fell from 66.6 to 65.3 — the SMP holds the exchange rate flat at its level when forecasts were finalised — which raises the price of imports. A lower assumed policy rate and a weaker assumed currency both push inflation up at the two-year horizon, and the published numbers duly went up.

Note what did not move: the trimmed mean, which is the measure the Board actually targets. May had it at 3.1% to June 2027 and 2.6% to December 2027; August has 3.0% and 2.6%. A full percentage point of headline relief bought one-tenth of underlying disinflation in 2027 and none at all by the end of it. The Board's own statement makes the same point in words — trimmed mean inflation "is little changed from the March quarter."

There is one more piece of arithmetic worth putting side by side, because the SMP publishes both halves of it. The wage price index is forecast at 3.3% through 2026 and 2027. Labour productivity is forecast at −0.5% across 2026. Wages growth above 3% against falling output per hour is a unit-cost path that does not sit comfortably under a 2.5% inflation target, and the statement names the constraint directly: "historically weak productivity growth continues to constrain potential growth." That is the domestic reason a Bank looking at rising unemployment forecasts still declined to close the door on tightening.

Worth noting on the external side, too: the August SMP revised major trading partner growth higher in the near term, saying "strength in AI-related activity continues to drive growth in high-income east Asian economies," and the statement observed that "the boost from AI-related investment has outweighed the adverse effects of the Middle East conflict." For an economy whose export book runs through east Asia, that is a growth input arriving from an unusual direction.

The Aussie through the five factors after the decision

The meter scores eight currencies — USD, EUR, GBP, JPY, CHF, CAD, AUD and NZD — on interest rates, growth, positioning, risk sentiment and commodities. Around this meeting the Aussie's picture was unusually clean, and the most important thing about it is how much of it was not Australian.

On the interest-rate factor, Australia carries a twelve-year-high policy rate while the other side of its most-traded pair moved the wrong way for the dollar. US payrolls fell 23,000 in July, reported on 7 August, and traders cut the odds of a September Fed hike from about 57% to 44%; the dollar index dropped 0.43% to 99.498 and the 2-year Treasury yield fell about eight basis points to 4.166%. A differential is a two-sided quantity, so the Aussie's rate support improved without the RBA doing anything — the mechanics of that report are in US payrolls fell 23,000, and the live read is on the USD currency page.

On the commodity factor, the June trade data helped: Australia's goods balance swung to a A$1.9 billion surplus from a revised A$2.4 billion deficit in May, on a rebound in gold and iron ore exports. Brent at US$83.55 cuts both ways for Australia — it lowers measured inflation while leaving the LNG and coal side of the export book weaker than the May SMP assumed.

On growth, the August SMP has GDP at 1.4% through 2026 and 1.5% to mid-2027 — still below the Bank's estimates of potential — with the unemployment rate now drifting to 4.8% by 2028 rather than the 4.7% May had pencilled in. The June labour force report was genuinely two-handed: unemployment up 0.1 percentage point to 4.4%, but employment up 76,300 to a record 14.82 million and participation at 67.0%, its highest since July 2025. We took that report apart in Australia's June jobs surge.

On risk sentiment, the Aussie remains the cleanest pro-cyclical expression among the majors, which is why it spent early August tracking a global equity recovery rather than anything domestic. And the China channel — the single largest external input to Australian growth — is running two speeds at once, which we mapped in China's July PMI at 49.2.

The net of that was AUD/USD above $0.705 into the meeting, a seven-week high, having peaked at 0.7061 on 5 August — and flat at $0.7055 coming out of it. The point worth carrying: the rate factor did most of the recent work, and most of the rate factor's recent move was American.

What would change the picture

The wage price index for the June quarter lands at 11:30am AEST on 19 August. That is the release that decides whether above-band underlying inflation is self-sustaining, because wages are the mechanism by which a price shock becomes a wage-price loop rather than a base effect. The August SMP nudged the WPI up to 3.3% through 2026 and 2027, from 3.2% in May — a small revision, but one made in the same document that lowered the assumed policy path, and against productivity it puts at −0.5%.

Then the July labour force release, against a June benchmark of 4.4% unemployment and record employment; then the minutes two weeks after the decision, which are the only place a unanimous decision reveals its internal distribution; then 28–29 September, the next meeting.

And underneath all of it, the barrel. The disinflation that broke the headline was imported, which means it is reversible in a way a domestic demand slowdown is not. The August SMP assumes Brent easing to US$81.2 by December 2026 and US$75.9 by mid-2027, plus a refined-fuel-to-Brent spread returning to historical averages by early 2027 — and it says explicitly that risks around its projections are "skewed to the upside." If the barrel goes the other way, the fuel base effect turns from a subtraction into an addition, and the number left standing is a trimmed mean the Board itself calls "little changed."

Which leaves the gap that this post was written around, now measured rather than guessed. Going in, the cash rate sat 35 basis points below the path the Bank's own return-to-target forecast was built on. Coming out, the Bank has moved that path down to 4.4% at December 2026 and 4.5% through 2027 — closing most of the distance from its side, leaving 5 to 15 basis points of implied tightening that market pricing does not carry, and accepting a slower return to target as the price. The forecast and the pricing are closer than they were. They are still not the same document.

To learn how Pip Theory builds its fundamental currency-strength scores, see the methodology overview.

Educational macro context only — not investment advice.