In this article
Kooky
Builder of Shaka, the payment router that pays every agent their commission on closing date.
About Kooky and Shaka →A forecast of home prices usually reaches a Queensland household as one line: prices to fall by so much, or to rise by so much. What does not travel with it is everything that gives the number its meaning: which index it refers to, which cities, between which two dates, and what the forecaster assumed about interest rates on the day the note was written.
Those details decide how the number should be read. A fall "of 10 per cent" measured from the highest point of a cycle to its lowest is a different statement from a fall of 10 per cent in a calendar year, and a figure for the eight capitals combined says little about a three-bedroom house in Townsville or a unit in Nundah.
This guide covers who publishes these figures, what goes into them, what they measure, why they change, and what happened after three sets of published forecasts, in 2020, in 2022 and in Brisbane this year. It does not say where prices are going.
Commonwealth Bank scenario as reported by The New Daily on 13 May 2020; CoreLogic national Home Value Index for April to September 2020 and for calendar 2021.
Who publishes price forecasts
Three kinds of organisation put a number on future home prices in Australia, and each does it for a different reason.
Most of the percentages that reach a headline come from the economics teams of the major banks, whose notes are reported by the press within hours. Research houses and portals form the second group: SQM Research sells an annual outlook, Christopher's Housing Boom and Bust Report, and Domain publishes a forecast for each financial year. The Reserve Bank is the third, and the quietest.
Related readQueensland's mean dwelling price is now second only to NSW, ABS says| Publisher | What it publishes | Usual form | When it changes |
|---|---|---|---|
| Bank economics teams | Dwelling price forecasts by capital city, alongside a cash rate forecast | A percentage for a calendar year, or a peak-to-trough figure | Whenever the rate view or the monthly data moves |
| Research houses and portals | An annual or financial-year outlook, city by city | A range, often with scenarios | Once a year, sometimes with a mid-year update |
| Reserve Bank | Forecasts for inflation, jobs and growth, with housing as an input | A stated assumption, not a published price target | Quarterly, with each Statement on Monetary Policy |
Index providers are easy to mistake for forecasters. Cotality, the firm formerly known as CoreLogic, reports each month what its Home Value Index did. That is a measurement of the past, and it is the yardstick most forecasts are later held against. The Australian Bureau of Statistics likewise measures and does not forecast prices.
Where the Reserve Bank stands
The Reserve Bank publishes a set of forecasts four times a year in its Statement on Monetary Policy. The variables it puts numbers on are inflation, unemployment, employment and economic growth, along with components of spending such as dwelling investment, which is the money spent building and renovating homes.
Home prices appear in a different way. In the Outlook chapter of its August 2026 Statement, whose forecasts were finalised on 5 August, the Bank wrote that housing prices "are assumed to continue to decline gradually for a period". The verb matters. The Bank stated the path of prices it had fed into its forecasts for spending and growth, because falling prices reduce household wealth and the number of homes changing hands. It did not attach a percentage for the public to quote.
The same chapter's forecast table lists the cash rate among its assumptions, with a note that it is assumed to move in line with expectations derived from financial market pricing. The rate path is an input the forecasts are conditioned on.
The Bank's reticence follows from its job. Its published forecasts centre on inflation and the labour market. Westpac's chief economist Luci Ellis, a former Reserve Bank assistant governor, wrote in a Westpac note dated 7 August 2026 that "targeting housing prices is not part of the central bank's remit".
Related readQueensland's median house price posts its first quarterly fall since 2022What goes into the model
In March 2019 the Reserve Bank released Research Discussion Paper 2019-01, A Model of the Australian Housing Market, by Trent Saunders and Peter Tulip. Commercial models are rarely published. This one is.
The model takes four things as given: household income, interest rates, population and the rate at which the housing stock wears out. From those it works through a chain of equations. Building approvals respond to interest rates and to prices. Completed homes, set against the number of new households, move the rental vacancy rate. The vacancy rate moves rents. Prices then respond to the gap between rents and the cost of owning, to their own recent momentum and to variable mortgage rates.
The list of inputs behind a commercial forecast is much the same, with a few more practical ones added:
- The cash rate path. The forecaster's own view of where the Reserve Bank will move next, and when.
- Borrowing capacity. How much a household on a given income can borrow at that rate. Canstar estimated in February 2026, as reported by Australian Property Update, that a single rate rise cut the borrowing capacity of an average income earner by about $12,000.
- Listings. How many homes are for sale compared with normal. When ANZ forecast rising prices in February 2026, part of its reasoning was that listings in the smaller capitals were more than 50 per cent below normal.
- Population. The ABS counted 5,739,500 people in Queensland at 31 March 2026, a rise of 90,100 or 1.6 per cent in a year, against 1.4 per cent nationally.
- Construction. ABS figures for the June quarter of 2026 show 11,671 dwellings commenced in Queensland and 9,549 completed.
- Policy. Tax and lending rules, which can move demand from investors and first home buyers without any change in rates.
The point for reading a forecast is that every one of these inputs is itself a forecast or an estimate. The cash rate path is a guess about decisions not yet made. Even the measured inputs move after publication: the ABS noted that its June quarter building figures were preliminary and subject to revision.
The Reserve Bank paper also shows how sensitive the result is to one input. Saunders and Tulip estimated that a cut of one percentage point in interest rates, if households expected it to last indefinitely, would lift prices in the long run by about 17 per cent on one setting of the model and 28 per cent on another. If the same cut was expected to last three years, the estimate was about 8 per cent after two years. The rate change is identical; only the belief about how long it lasts is different.
Related readReading a suburb median price in Queensland: what it shows and hidesWhat the number actually measures
A percentage needs four labels before it means anything: the index, the area, the type of dwelling and the two dates it runs between. Headlines usually keep one of the four.
One forecast published this year shows how much the labels matter. On 11 August 2026 ANZ economists Adam Boyton and Madeline Dunk released a revised outlook for capital city dwelling prices. Press reports of the same document, in The Nightly on 11 August and Smart Property Investment on 12 August, carried three different figures for the combined capitals: a fall of 4.3 per cent in 2026, a fall of 3.4 per cent in 2027, and a fall of 10.6 per cent from this year's peak to a low point expected in mid-2027.
None of the three contradicts the others. A figure for a single year compares its end with its start, so a year that rises for some months and then falls can show a modest net change. The peak-to-trough figure runs from the highest month to the lowest, wherever they sit in the calendar, and is always the largest number available. Two headlines with different numbers may be describing one forecast.
The area changes the number just as much. In the same ANZ outlook, as reported by The Nightly, the peak-to-trough fall expected for Sydney was 14.5 per cent, for Melbourne 12.8 per cent, for Adelaide 9.8 per cent, for Brisbane 7.9 per cent and for Perth 5.2 per cent. The combined capitals figure of 10.6 per cent is weighted towards Sydney and Melbourne because they hold the most housing value. Regional Queensland is in none of these figures.
Related readReserve Bank holds at 4.35% and says housing momentum has shiftedThe measure itself differs between publishers. A bank forecast normally refers to a value index of the kind Cotality produces. Domain's forecast refers to median prices over a financial year, and gives houses and units separately. An earlier guide explains why an index and a median can move differently in the same quarter; the consequence here is that a forecast of one cannot be marked against the other.
Ranges, scenarios and the base case
Careful forecasters do not publish a single number. They publish a central case and say what would have to happen for the result to land elsewhere.
Domain's Forecast Report for the 2027 financial year, released on 24 June 2026 according to ABC News, is an example. For Brisbane houses Domain gave a range, a rise of 3 to 7 per cent over the 12 months to June 2027, and Smart Property Investment reported on 1 July that the top of the range was worth as much as $87,000 on the city's median. The report listed what would push the result up, including earlier rate cuts, stronger first home buyer demand and high population growth, and what would push it down, including rates staying higher for longer, a weaker labour market and faster growth in supply.
Banks do the same thing with scenarios. In May 2020 Commonwealth Bank published a base case and a "prolonged downturn" case side by side, a pair examined below. The severe case was presented as a scenario beside the base case. It described what could happen under harsher assumptions, which is a different thing from what the bank expected.
Related readUnder 1 per cent of borrowers owe more than their home is worth, RBA saysThe headline often quotes the edge of the range
A forecast with a base case and a severe case contains two numbers, and the larger one makes the stronger headline. Before reacting to a figure, check whether it is the forecaster's central expectation or the outer scenario, and whether it is a range with a lower end that went unreported.
The Reserve Bank's own research supports the habit of thinking in ranges. Research Discussion Paper 2012-07, by Peter Tulip and Stephanie Wallace, measured the Bank's past forecast errors and concluded that "uncertainty about forecasts is high". It found the forecasts had substantial explanatory power for inflation but not for economic growth.
Why forecasts get revised
A revision is not an admission of failure. It is what a conditional statement does when its conditions change.
The most common trigger is the cash rate. A price forecast is built on top of a rate forecast, so when the rate view moves, the price view follows. Luci Ellis's note of 7 August 2026 shows the sequence plainly. Westpac had published housing projections in late June. By early August, she wrote, prices were falling a little faster than those projections expected, the bank had removed two rate rises from its outlook and brought forward its first expected cut from February 2028 to August 2027, and the housing forecasts would be formally revised the following week, after the Reserve Bank's meeting.
The second trigger is the monthly data. Index results arrive at the start of each month, and a forecaster whose path is already off after two or three readings will usually move it.
The third is policy that was not in the original assumptions. The Reserve Bank's August 2026 Statement lists tax policy changes beside the tightening in monetary policy earlier in the year and the general economic environment as reasons for the price path it assumed.
Some publishers build the revision into the product. SQM Research's Boom and Bust Report for 2026 was published on 26 November 2025, and the firm's own page lists a revision update dated 10 March 2026 as part of the same purchase.
Related readSouth-east Queensland sellers adjust as homes take longer to sellA forecast therefore has a date, and the date is part of the number.
2020: the forecast and the year that followed
On 13 May 2020, The New Daily reported the scenarios in Commonwealth Bank's quarterly trading update. The bank's base case had home prices falling 11 per cent over three years, on the assumption that the economy shrank 6 per cent in 2020 and recovered the next year. Its prolonged downturn scenario had prices falling almost a third, 32 per cent, from their March 2020 peak by March 2023. The same article noted that all four major banks expected a double-digit decline over the following couple of years.
What the index recorded was very different. CoreLogic's national Home Value Index fell 2.1 per cent between April and September 2020, then rose for three months, and finished the calendar year 3.0 per cent higher. Regional values rose 6.9 per cent and the combined capitals 2.0 per cent. In 2021, according to CoreLogic figures reported by The Canberra Times on 4 January 2022, national values rose 22.1 per cent and Brisbane's rose 27.4 per cent.
The forecasts were not careless. They followed from their assumptions, beginning with an economy that shrank 6 per cent in a year. CoreLogic's own summary at the end of 2020 was that forecasts for economic conditions had been overly pessimistic and had not reflected how the economy went on to perform. The price forecasts rested on those economic forecasts. When the foundation turned out wrong, so did the figure built on it, and by a wide margin.
Related readSunshine Coast home values ease from their autumn peak after a long run2022: right direction, different depth
Two years later the direction was called correctly, which makes the episode more useful to study.
On 26 August 2022, Investor Daily reported a forecast from Commonwealth Bank's head of Australian economics, Gareth Aird. National dwelling prices were expected to fall about 15 per cent from their April 2022 peak by mid-2023, with Brisbane down 17 per cent, Sydney 18 per cent and Perth 8 per cent. The forecast stated its key assumption: a cash rate peaking at 2.60 per cent. It also stated its own condition, that a larger fall was likely if rates went above that level.
Rates did go above it. The Reserve Bank's cash rate table shows the target reached 2.60 per cent on 5 October 2022 and kept climbing to 4.35 per cent on 8 November 2023, which is 1.75 percentage points higher than the forecast assumed.
Prices did not fall further as a result. CoreLogic reported that national values fell 7.5 per cent from the April 2022 peak to a trough on 29 January 2023, half the forecast decline, then rose 8.1 per cent to a new record on 22 November 2023, with Brisbane also at a record.
Commonwealth Bank forecasts as reported by The New Daily (13 May 2020, base case over three years) and Investor Daily (26 August 2022, peak to trough); recorded falls from CoreLogic's national Home Value Index, April to September 2020 and April 2022 to January 2023.
The forecaster had the direction right and the main input wrong in the direction that should have made the fall deeper, and the fall was shallower all the same. Whatever the mix of causes, the rate alone did not decide the outcome. Knowing the cash rate in advance would not have been enough to know the price.
Related readUp 0.3% or down 0.2%? Two indexes split on Brisbane prices in JuneBrisbane in 2026: three forecasts in six months
Queensland readers have watched the same process this year. What follows records what was published and what the index has shown to date. It is not a verdict on any forecaster, because the periods forecast have not ended.
On 6 February 2026, according to Australian Property Update's report three days later, ANZ forecast that capital city prices would rise 4.8 per cent in 2026, with Brisbane up 9.5 per cent in 2026 and 3.9 per cent in 2027. The bank expected the cash rate to stay at 3.85 per cent after February's rise.
By June, Smart Property Investment reported, ANZ expected a fall of 2.1 per cent for the capitals in 2026. Domain's range for Brisbane houses followed on 24 June. On 11 August, ANZ revised again, to the figures set out earlier, including a peak-to-trough fall of 7.9 per cent for Brisbane. The bank noted that Brisbane and Perth had begun falling sooner than it expected, and its new outlook assumed the Reserve Bank had finished raising rates at 4.35 per cent.
| Published | Combined capitals, 2026 | Brisbane | Cash rate view |
|---|---|---|---|
| February | Rise of 4.8% | Rise of 9.5% in 2026 | On hold at 3.85% |
| June | Fall of 2.1% | Not reported | Not reported |
| August | Fall of 4.3% | Fall of 7.9%, peak to trough | No further rises from 4.35% |
Australian Property Update, 9 February 2026; Smart Property Investment, 12 August 2026; The Nightly, 11 August 2026. The Brisbane figures use different measures and cannot be subtracted from one another.
Then the conditions moved once more. At the end of September the Reserve Bank raised the cash rate to 4.60 per cent, the fourth rise of the year, according to Commonwealth Bank's newsroom on 30 September. On 1 October, Cotality reported that national values fell 1.1 per cent in September, a sixth monthly fall in a row, leaving them 5.2 per cent below the record set in March, and that Brisbane's fall of 1.5 per cent was the sharpest of any capital that month.
Between February and August, a published view on Brisbane from a single bank went from a strong rise to a fall, and the cash rate view beneath it moved from 3.85 to 4.35 per cent. Anyone who had fixed on February's number as a fact would have been working from an outlook its authors had already replaced.
How to read the next one
Five questions restore the labels a headline drops. The answers are usually in the original note or in a careful press report of it.
- When was it writtenFind the date of the note, not of the article quoting it. Check whether the cash rate has moved since.
- Between which datesA calendar year, a financial year, or peak to trough. The last is always the largest figure.
- Which area and dwellingCombined capitals, Brisbane alone, houses or units. Regional markets are rarely included.
- What rate path it assumesThe forecast holds only if rates follow the path its authors expected.
- Is it the base caseLook for the range or the scenarios, and for which of them the headline quoted.
Comparisons need the same care: a single-year figure from one bank belongs beside a single-year figure from another, never beside a peak-to-trough number.
What a forecast cannot say about one home
Even a forecast that proves accurate describes an average. The combined capitals index is a weighted mix of millions of dwellings, most of them in Sydney and Melbourne. Brisbane's own figure blends inner-city units with houses on the outer edge of Logan and Moreton Bay. The 2020 figures show how far the parts can sit from the whole: regional values rose 6.9 per cent that year, more than three times the 2.0 per cent recorded for the combined capitals. A buyer in a regional Queensland town and a buyer in an inner capital suburb were living through different markets under one headline.
A forecast also says nothing about what decides the price of one property: its condition, its street, how many similar homes are listed that month and how many buyers turn up. Those are observed, not forecast.
A forecast is a dated answer to a question that begins with "if". Read the "if" first, then the date, then the number.
Kooky, from Shaka
Kooky edits Queensland Estate and builds Shaka, the payment router he made for Queensland property professionals. One payment comes in, and every agent, agency and party in the deal receives their signed share on closing date.
