Four Empty Weeks vs a 10% Rent Fall: What Each Costs an Investor

Four empty weeks on a $600-a-week rental is $2,400 of rent that never arrives. Spread across a year, Plintha's engine puts the after-tax cost at $29 a week for an investor who bought before the 2026–27 Budget, and $47 a week for one who bought the same established house after it, once the new loss rules start on 1 July 2027.

A vacancy is the shock most investors picture first. On this property it is also the smallest of the three the engine tests. A 10% rent fall costs more, and it keeps costing every week of every year until rents recover.

Every figure below comes from Plintha's deterministic finance engine, using its standard stress-test rows. The exact inputs are published alongside this post so anyone can re-run them.

The property

One established house in New South Wales, run for two owners who differ only in purchase date.

• Purchase price: $720,000
• Weekly rent: $600 ($31,200 a year, a 4.33% gross yield)
• Loan: 80% LVR, interest-only, $576,000 at 6.75%
• Marginal tax rate: 37%
• Stamp duty: $26,587 on the engine's NSW schedule, the same for both owners

The 6.75% rate is an illustrative investor rate, not a quote. It is the same "now" rate used in our rate-rise worked example, which covers interest-rate risk. This post holds the rate still and moves the rent.

The base case already includes a vacancy allowance of 3% of rent. Here that is $936 a year, about a week and a half of rent. The engine's vacancy shock is four weeks on top of that allowance.

What each shock does to the weekly cost

The right-hand column applies only from 1 July 2027; until then both owners sit in the left-hand column. Each bracketed change is the difference between two rounded weekly figures, so it can sit a dollar either side of the unrounded amount.

The rent fall beats the vacancy. A 10% fall takes weekly rent from $600 to $540 and removes $3,120 of rent a year. Some costs fall with it: the management fee and the vacancy allowance are percentages of rent, so the engine's operating costs drop by $369. The net hit is $2,751 a year before tax, against $2,400 for the four empty weeks. Put another way, a 10% rent fall costs about the same as four and a half empty weeks, repeated every year.

The two together cost slightly less than their sum. When rent has already fallen, each empty week loses $540, not $600.

A rate rise of one percentage point still costs more than either. On this loan it adds $70 or $111 a week, more than the vacancy and the rent fall combined for either owner. Rent shocks are the smaller risk here, but nothing stops them arriving in the same year as a rate rise.

Why the same vacancy costs one owner $29 and the other $47

Both owners lose the same $2,400 of rent. The difference is what the tax system does with the larger loss.

The grandfathered owner can still deduct a rental loss against salary. When rent falls, the loss grows and so does the refund. At a 37% marginal rate the refund covers 37% of the missing rent, and the owner bears the other 63%: $1,512 of the $2,400.

From 1 July 2027, an established property bought after 7:30pm AEST on 12 May 2026 loses that offset. Its losses are deductible only against other residential property income, including capital gains, and any excess carries forward to later years (Treasury's Budget explainer on negative gearing and CGT). For an investor with no other residential property income, as modelled here, none of the missing rent comes back at tax time that year, and they fund all of it in cash. The unused loss is carried forward and may reduce tax on later residential property income or a capital gain. The engine's weekly figure gives that future value no credit.

That rule is not yet in force. Until 30 June 2027, the post-Budget buyer still receives the salary offset, and on the engine's numbers their costs match the grandfathered column: $197 a week in the base case, not $384. The right-hand column is the position from 1 July 2027, which is where most of a multi-year hold will sit. We set out the class rules in negative gearing after the 2026–27 Budget.

One condition sits behind the refund in the left-hand column. The ATO allows rental deductions for the period a property is rented or genuinely available for rent, which it describes as advertised with broad exposure to potential tenants, on terms under which tenants are reasonably likely to rent it (ATO, rental expenses). The engine assumes an empty property meets that test. A property held off the market, or advertised well above market rent, may not.

A weekly average hides a monthly bill

The engine reports a vacancy the way it reports everything else: as a weekly figure averaged over the year. That is useful for comparing shocks, but it is not how a vacancy arrives.

The loan does not pause while the house is empty. Interest on $576,000 at 6.75% is $748 a week, or $2,991 over four weeks, and for those four weeks no rent arrives to meet it. The $29 or $47 a week is that one-month gap spread across twelve months. The cash is needed in the month the tenant leaves, and any refund arrives after the tax return is lodged.

A rent fall works the other way. It never produces a single large bill, so it is easy to underrate, but it has no end date.

Which shock hurts most

On this property, in order of weekly cost: the rate rise, then the rent fall, then the vacancy. That order holds for both owners; the scale differs. Every shock costs the post-Budget buyer about 59% more, because the grandfathered owner bears 63% of each lost dollar and, from 1 July 2027, the post-Budget buyer bears all of it.

It also matters where each owner starts. The grandfathered owner's cost stays in the engine's amber band through the vacancy, the rent fall and both together. From 1 July 2027, the post-Budget buyer's base case of $384 is already in the red band for an interest-only loan before any shock is applied.

The engine's worst-combined row puts a rate rise of three percentage points, the 10% rent fall and the four empty weeks together. It returns $466 a week for the grandfathered owner and, from 1 July 2027, $811 for the post-Budget buyer.

What these numbers don't cover

• How likely each shock is. The engine prices a shock; it does not predict one. Published vacancy rates describe whole cities or regions. A market-wide rate says little about one house in one street, or about a tenant who leaves in December.
• Costs of re-letting. A real vacancy often comes with a letting fee, advertising and repairs between tenants. The stress row removes rent only and leaves other costs unchanged.
• Tax bracket. The 63/37 split assumes a 37% marginal rate. On a lower rate, the refund covers less and the two columns move closer together.
• Other property income. A post-Budget buyer with another profitable rental can use the loss against that income, which softens the right-hand column.
• Landlord insurance. The engine assumes no insurance payout for lost rent. Policies differ in what they cover.

None of this says a property with vacancy risk is a poor one. It says the cost of an empty month or a softer rent is a number that can be worked out before buying, and that it now depends on when the property was bought as well as on the rent. To see the same stress rows for a specific property, with your own price, rent and loan, you can run it through Plintha.

This article is general information, not financial, tax, credit or legal advice. It doesn't take into account your objectives, financial situation or needs. Figures use Plintha's engine with the assumptions stated above (80% LVR, interest-only at an illustrative 6.75%, 37% marginal rate, NSW duty, standard operating-cost model) as at 19 October 2026, and will change if any of those inputs change. Check your own position with a licensed tax agent or financial adviser before acting.