Rental listings vs vacancy rate: which to use and when

Two published “vacancy rates” can both be correctly calculated and still disagree. The difference is usually the method, not a typing error.

The Australian Bureau of Statistics Census records rent paid by households in occupied private dwellings (RNTD) and counts unoccupied private dwellings. Unoccupied on Census night includes holiday houses, dwellings between occupants, and dwellings empty for other reasons. It is not a rental vacancy rate.

Victoria’s Department of Families, Fairness and Housing Rental Report uses rental bond data as a proxy for the stock of private rental dwellings, and reports median weekly rent from new lettings in the quarter. That is a bond-based stock and a new-lettings rent. It is not the same as an agent-survey vacancy rate, and the department’s notes exclude student housing, rooming houses, caravans, share-house duplicate bonds and rooms.

State bond authorities in other jurisdictions publish lodgement counts and sometimes rents. Those series have their own exclusions. New South Wales requires a fixed advertised rent and lodges bonds with Fair Trading; that still does not publish a lot-level vacancy rate. Read the current official note for the series you opened.

A count of comparable rentals advertised now (same type and bedrooms, in a suburb or a radius) is a direct observation. It needs no denominator. Its weakness is the missing stock: it cannot tell you what share of rental dwellings is empty, and it cannot see lets that never appear online.

Track the count over several weeks if you want direction. One snapshot is a level. Districts uses comparable listing counts in the supply pressure check for that reason: the count can be shown for the property’s class with a source date.

Labelled illustration. A published vacancy rate of 1.1 per cent for a postcode, period and method stated by that publisher. Nineteen three-bedroom houses advertised for rent in the suburb, up from eight three months earlier. The rate says the wider rental stock, on that publisher’s denominator, is thin. The count says the house class you are buying is loosening. Neither number is a typing error. For a three-bedroom house, plan around the class count and the pipeline. Do not invent a “balanced” percentage that converts one into the other.

• Calling a listing count a vacancy rate.
• Comparing two publishers, or two periods after a method change, as if they were one series.
• Treating Census unoccupied dwellings as rental vacancy.
• Applying a metro rate to a class that barely exists in that suburb.
• Using an unsourced “2 per cent is balanced” rule. Districts does not use one, and this page does not either.

Districts does not know how many rental dwellings exist in a suburb, so it cannot turn a listing count into a rate. It cannot see private lets or withdrawn campaigns. Where a suburb profile quotes a published rental statistic, it is labelled with publisher and period. There is no Districts vacancy rate.

Open the publisher’s methodology note. Write definition, denominator, window and geography on your file. If the note is missing, treat the figure as unusable for comparison. Track comparable listings yourself. A manager who lets that class can describe time to let and incentives without converting them into a rate.