Trans-Tasman Stress Atlas

Trans-Tasman Insurance and Financial Stress Atlas

The first public map of where the cost of home insurance and household financial stress already overlap — by area, across Australia and New Zealand, built entirely from government open data.

On 11 November 2026, Westpac NZ began telling borrowers the natural-hazard risk of a property at the point of origination. That is the right first step. The gap it leaves is everything after settlement — the years in which an insurer reprices a policy, raises an excess, or withdraws cover. The borrower sees the renewal notice; the lender does not. This atlas shows, area by area, where the cost of insurance and the signs of household financial stress already sit on top of each other. It does not claim one causes the other.

What the index is

It is a composite of administrative stress signals, published at Australian SA3 and, provisionally, at New Zealand SA2. For Australia it combines four pillars — personal insolvencies, electronic-gaming cashflow loss, small-business contraction, and a home-insurance affordability gap — each rank-transformed to a percentile and weighted. For New Zealand it combines two: socioeconomic deprivation and rental-bond turnover. Every pillar is drawn from a government or regulatory open dataset, so each number traces back to a published table. The full construction, weights and caveats are in the methodology.

Australia: where the two pressures meet

The sharpest signal is the overlap — the areas sitting in the top fifth of the country on both the insurance-affordability gap and personal insolvency. There are 22 such SA3s, in three clusters. The largest is inland agricultural country across the NSW–Queensland border: Lachlan Valley, Moree - Narrabri, Bourke - Cobar - Coonamble, Tamworth - Gunnedah, Inverell - Tenterfield, Burnett, Granite Belt and Darling Downs (West) - Maranoa. The second is south-east Queensland’s growth corridor — Southport, Surfers Paradise, Redcliffe, Caboolture and Beenleigh — where premiums have risen fastest against modest incomes. The third is the regional Queensland coast: Whitsunday, Bundaberg, Townsville and Hervey Bay. Upper Murray exc. Albury, Wyong, Campaspe, Wheat Belt - North and Albury complete the set.

What drives the very top of the table differs by place. Some areas are insurance-led, where the premium-to-income ratio does the work; some are insolvency-led, carrying a high bankruptcy and debt-agreement count; and a few are contraction-led, pulled up by business failure rather than household debt. Publishing all four pillar percentiles beside the composite — never the composite alone — is what lets you tell those cases apart on the map above.

New Zealand: provisional, and honest about it

The New Zealand composite is explicitly provisional: two pillars, not four. It can show where deep socioeconomic deprivation coincides with a churning rental market — Wiri West, Māngere Mascot, Mount Roskill Central South, Highbury East, Conifer Grove East and Gardens (Dunedin City) sit at the top, driven by deprivation. It cannot yet say anything about insolvency or insurance at SA2, because comparable statutory data is still being assessed. Of 2,379 SA2s, 207 have no bond-turnover reading and 421 have no composite at all — either too small to score or with only one pillar present. Those areas are drawn grey, “no composite”, never as low stress; a one-pillar score would just be deprivation relabelled.

What is not here

Three gaps matter, and hedging about them would be dishonest. First, gaming: of Australia’s 340 SA3s, 194 — 57.1% — carry no cashflow pillar, because sub-state gaming expenditure is open data only in Victoria and Queensland. Western Australia, South Australia, Tasmania, the Northern Territory and the ACT publish nothing usable at this grain, and New South Wales publishes machine counts, not dollars. Those areas are hatched, not scored low. Second, New Zealand is two pillars, as above. Third, the insurance pillar is a point-in-time baseline from the 2026 ACCC monitoring report, expressed as an average-premium band over Census income — a coarse ratio, not a modelled premium, and one that will not move between quarters until a recurring affordability source exists.

What comes next

The index refreshes each quarter with the AFSA insolvency release; each refresh is a new version and the old one is never overwritten. Release 2 is planned for early 2027; its subject will be announced with the February index refresh. Corrections and better sources are welcome — the code and data are public.