He is 58, owns two rental houses in Melbourne's outer east, and until May he had never once thought about commercial property.

Then the tax rules changed, his accountant said a sentence he can't stop thinking about, and now he has a spreadsheet open comparing his three-bedroom rental against a freestanding childcare centre in a growth corridor he's never driven through.

We've had this exact question three times in a fortnight. It's worth answering properly, because the arithmetic that makes it look obvious is the same arithmetic that hides what you'd actually be buying.

What changed, and why everyone suddenly wants a service station

On 12 May, negative gearing was limited to new residential builds, and the 50 per cent capital gains tax discount was replaced with cost base indexation plus a 30 per cent minimum tax rate. Anything bought before that date is grandfathered. Anything bought after it isn't.

The intent was explicit: discourage investors from buying established homes. It worked. National home values slipped 0.1 per cent in April, sat flat in May, then fell 0.3 per cent in June and 0.3 per cent again in July — on top of three interest rate rises. We've written before about whether those tax changes are a reason to sell, hold or buy more, and the short answer hasn't changed.

And the money went looking for somewhere else to be. Since the budget, one national agency alone has transacted $164.7 million across 29 freestanding commercial assets. Fast food transaction values are up 48 per cent year-on-year, childcare up 14.2 per cent, in a national freestanding market worth $2.68 billion. Brokers describe a clear shift from residential enquiry to commercial enquiry, dated specifically to the budget.

The headline sales are what's pulling people in. A Sydney childcare centre at $16.01 million on a 4.37 per cent yield — a national record per licensed place. A NSW fast food site at $11.65 million on 3.86 per cent. A regional Victorian fuel and convenience truck stop at $21.27 million on 5.96 per cent. A large-format liquor and childcare pairing at $18.05 million on 5.5 per cent, from ten first-round offers.

Set a 5.5 per cent net commercial yield against a suburban house grossing 3 per cent before rates, insurance, management, maintenance and vacancy, and the spreadsheet looks like it has already decided.

The part the yield number doesn't tell you

Three things.

The yield is the compensation, not the prize. A 3.86 per cent yield on a fast food site is not a bargain — it is the market pricing a very long lease to a very strong covenant. You aren't being paid more for taking more risk; you're being paid less because the income is safer. Meanwhile a 6 per cent regional fuel site is not "better value" than a 4.37 per cent metro childcare centre. The 1.6 per cent gap is the risk difference, quantified by people who do this professionally. When a yield looks generous, the first question is always: what does the market know about this asset that I don't?

You are buying one tenant, not a market. A house has a tenant pool. If one household leaves, dozens can replace them, and the vacancy is measured in weeks. A single-tenant commercial building has exactly one customer, and when that customer's lease expires — or their business model changes, or a regulator moves, or the operator restructures — your income doesn't fall by ten per cent. It goes to zero, and it stays there while you pay outgoings on an empty specialised building that was designed for one use. Suburban office vacancy just hit a 30-year high; one metro office market is sitting at 23.5 per cent. Commercial vacancy is not like residential vacancy. It is longer, deeper, and it arrives all at once.

The buyers arriving now are what compresses the return. The agencies selling these assets are openly forecasting that new entrants from a residential investing base will add competition and push yields down through the second half of FY27. Read that plainly: the people who arrive next pay more for the same income. If your reason for switching is the yield gap, you are entering at the point where the yield gap is being competed away — by people exactly like you, making exactly your decision, for exactly your reason.

The question underneath the question

Notice that nobody in this conversation has mentioned a location.

The whole debate is being run on asset class — houses versus childcare, residential versus commercial — as if those were the two options and the address were a detail. It isn't a detail. It's the entire result.

This is where the comparison quietly breaks down, and it breaks down in a way that favours neither side of the argument. A specialised commercial building is a bet on a catchment. A childcare centre's value is a function of how many under-fives live within a realistic drive, what the competing supply is on the surrounding roads, whether the corridor's household formation is still running, and how the traffic actually moves past the door. A fuel and convenience site lives or dies on traffic count, the side of the road it sits on, and the intersection two hundred metres away.

None of that is a suburb-level question. All of it is a street-level one.

We measure this every day on the residential side, and the spread is the thing most owners never see: two streets inside a single suburb — same postcode, same median, same school catchment, same council — routinely diverge 20 to 30 per cent on effective yield once you use achieved rent rather than advertised rent and account for vacancy duration and days on market. One street lets in four days to a queue. The one behind it sits five weeks and lets at a discount to a tenant who's gone in eleven months. The suburb median reports both of those as the same number.

If a suburb median can hide a 20 to 30 per cent performance gap on a house — an asset with dozens of possible tenants and a deep, liquid resale market — consider what an asset-class average hides on a purpose-built building with one tenant, a specialised layout, and a resale pool consisting of other people who want that exact use.

The honest version of the question isn't "residential or commercial?" It's: which specific asset, on which specific street, do I understand well enough to hold through a bad five years? For most owners with two rental houses and no commercial experience, the answer to that question is not a childcare centre in a corridor they've never driven through.

What we'd actually tell him

Three things, in order.

Separate the tax problem from the asset problem. He's grandfathered. His two existing houses keep their treatment. Selling them to escape a rule that doesn't apply to him — and crystallising the capital gains to do it — is not tax planning; it's paying a large, certain, immediate cost to avoid a hypothetical one. Grandfathering is also more fragile than most owners assume — it can be disturbed by events that have nothing to do with a sale, as owners of jointly held property discovered this year. Get it modelled by an adviser before anything else.

If commercial genuinely suits you, arrive as a professional. Lease term, remaining term, option structure, who pays outgoings, the tenant's actual financial strength, land value underneath the improvement, alternate use if the tenant leaves. A residential investor's instincts are close to useless here, and the gap gets expensive quickly.

Then apply the same standard to both sides. If you're prepared to interrogate a commercial catchment street by street, interrogate your residential holdings the same way — because most owners have never done it, and the underperformance they're blaming on "the market" or "the tax changes" is very often specific to the street they bought on. It's the same failure that catches owners who assume a property that won't sell can simply be rented out instead.

The reframe

Here's what makes this moment interesting rather than alarming.

Investors are leaving established residential stock. Victoria alone lost more than 640 rental homes in a single month as owners sold up. Meanwhile the national vacancy rate is 1.3 per cent — up a tenth of a point and still, by any historical standard, extraordinarily tight, with most capital cities below one per cent. Asking rents are up 8.1 per cent over twelve months. Governments have promised hundreds of thousands of new homes and have not come close to delivering.

So the supply of rental housing is shrinking, demand isn't, and a meaningful share of your competition for well-selected established stock has just decided to go and buy a service station instead.

That is not a description of a broken asset class. That is a description of a market with less competition in it than it had six months ago — which is historically when the best acquisitions get made, by the people who prepared in advance rather than reacted.

The thing that breaks a property investor is almost never the asset class. It's a financial structure that couldn't survive a rough patch, or an asset chosen on an average. Sentiment shifts every few months. Structure — scarce land, a supply deficit no government has solved, and population growth — moves on a decade.

The mistake was never owning residential property. It's owning it, or leaving it, without ever having found out what your particular street was actually doing.

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