Vacant social housing sitting empty while a waiting list grows is the kind of thing that should be visible to the public and to the agency alike. Duration is the important variable: a dwelling empty for two weeks between tenancies is normal operation, and one empty for two years is a policy failure.

This entry carries a genuine privacy constraint and is recorded as such. At fine geography, vacancy data can identify particular dwellings and particular tenants, and identifiable is a legitimate blocker rather than an excuse.

The register keeps that distinction carefully. Treating a real privacy limit as obstruction would cost this project the credibility it needs when the blocker is format and there is no excuse at all.

The answerable version of this question is aggregated to a geography coarse enough to protect individuals and fine enough to be useful. That is a solvable design problem, and stating it that way is more useful than demanding raw data nobody should release.

Checked on 2026-09-02, and partly corrected. DCJ publishes a quarterly social housing delivery dashboard, and average time to fill a vacant property is reported and audited against target -- 33 days in June 2024 against a target of nine. The entry previously implied vacancy duration was simply not measured publicly, which was wrong at state level.

What survives is the geography. The published figure is a statewide average, and an average cannot distinguish routine turnaround from dwellings held empty for years in particular areas, which is the thing the question was asking about. aggregated therefore remains the leading blocker, and identifiable remains a genuine constraint on how fine any release could go.