He has read that house prices are falling 1.6 per cent, 14 per cent and more than at any point in forty years. All three were published in the same week, about the same market.

"I own two investment properties and I've read four things this week that can't all be true," he told us. "One says the market's down under two per cent. One says it's the worst crash in forty years. What do I do with that?"

He is a mid-forties owner with a house in Melbourne's outer ring bought in 2019 and one in mid-ring Brisbane bought in 2022. Both tenanted, both ahead on repayments. He is not in trouble, which is exactly why his question is the interesting one. He isn't asking whether to panic — he's asking what any of these numbers mean for the two houses he actually owns.

It is the most common question we're getting right now, and the answer starts somewhere unexpected: those four numbers aren't contradicting each other. They're answering four different questions — and none of them is the question he actually asked.

What the numbers actually say

Start with the central bank's own published assessment this month — the most measured document anyone has produced on this, and the one almost nobody has read past the headline written about it.

National housing prices are down 1.6 per cent from their peak, which was set in March this year. Prices are still around 5 per cent higher than a year ago. And they are still roughly 50 per cent higher than at the start of the decade.

All three are simultaneously true. A market can be down from its peak, up on the year and dramatically up over five years at once — because each figure is measured from a different starting line. The 1.6 per cent runs from March. The 5 per cent from last August. The 50 per cent from 2020. Choose your starting line and you can generate a crisis or a boom from identical data.

The same assessment puts the long-run effect of the announced investor tax changes at 0 to 5 per cent. Not 14. Not forty years of anything. Zero to five, over the long run.

So where do the enormous numbers come from? Forecasts — projections of where prices might go, mostly concentrated on the most expensive segments of Sydney and Melbourne. A forecast is not a measurement; it's an opinion about the future with a number attached. The sharper scenarios also appear under the heading of risks — things that could happen — not central projections. That distinction dies the moment a headline is written.

The part nobody is reporting

Here is what makes this genuinely worth his attention, and it sits in the same document as the fall.

Dwelling investment is still growing — up 3.5 per cent over the year. Residential building approvals were still increasing. And developers report that demand for greenfield land continues to exceed supply in some regions.

In the middle of what is being reported as the worst decline in four decades, there are places where buyers still outnumber available land.

At the same time, construction is expected to slow, because lower sales volumes and higher costs make new projects harder to justify. Fewer homes get built. Demand doesn't disappear when supply does — it goes somewhere else, usually into rents.

That is not a market in collapse. That is a market re-sorting — falls concentrated in Sydney and Melbourne, conditions weaker in Victoria than other states, and capacity pressure still a live concern in Queensland's construction sector.

Which brings us to his actual problem.

Why the national number can't answer his question

He owns two houses. He does not own the national market. There is no version of the national index that he can sell.

Even the state-level splits above are too coarse. His two properties sit in different states, price segments, rental markets and supply pipelines — and the number he keeps reading averages all of that with several million dwellings he has nothing to do with.

This is where our research work lives. When we run street-level data across a single suburb — achieved sale prices rather than asking prices, actual days on market, real vacancy duration, genuine supply pipeline — the spread between the best- and worst-performing street inside one suburb routinely runs to 20 to 30 per cent on effective yield, and wider on capital performance. Same postcode. Same median. Same council. Same catchment. Same interest rate, same tax changes, same headlines.

A national figure of −1.6 per cent is the average of streets that rose and streets that fell hard. In a rising market that dispersion is a missed opportunity — it is the same force that means a $210,000 renovation can add almost nothing when the street won't carry it. In a correcting market it is the difference between an asset that dips and recovers and one that doesn't.

The correction is not falling evenly, and never does. It lands hardest on properties that were only working while everything went right — thin buyer depth, weak underlying rent, no structural scarcity. Well-selected assets in genuinely supply-constrained locations barely register the same event. That held in every previous cycle, and nothing in this month's data suggests otherwise.

What we told him to do

Three things, in order.

Stop underwriting off any number that covers more than a suburb. National, capital-city and state figures are useful for reading sentiment. They are useless for deciding anything about a specific address.

Get achieved data on each of his two properties individually — what comparable homes on those actual streets have transacted for, how long they took, what they genuinely rent for and how long they sat empty between tenants. Not the median, and not the advertised rent — the difference between what a forecast says rents will do and what they actually did is routinely enormous. The same trap catches owners who assume a national count tells them something about their own property, which is exactly what happens when more than a million homes get recorded as unoccupied.

Separate measurement from forecast. Down 1.6 per cent from March happened. Everything above that number is somebody's view of the future, and views change monthly.

He came to us braced to be told which of the four headlines was right. The honest answer was that none of them were about him.

The bigger point

A market still up 5 per cent on the year and 50 per cent over five years, where land demand outstrips supply in places and fewer homes will be built into persistent demand, is not a market to abandon. It has simply stopped rewarding buying anything and started rewarding buying the right thing — which is a considerably better market for anyone willing to do the work.

The mistake was never owning property through a correction. It's owning property while trying to steer by a number that was never measuring your house.

This article is general information only and does not take into account your personal circumstances, financial situation or needs. It is not financial, taxation or legal advice. Consider seeking advice from a licensed professional before making any investment decision.