Japan’s quake insurance exposes deep fault lines in ours

2026-03-05 - 16:07

Comment: Another major earthquake in New Zealand isn’t a remote possibility, it’s a statistical certainty. Although seismic hazard itself is relatively stable, sharp increases in insurance premiums and escalating losses from other natural hazards raise a pressing question: does the country’s natural hazards insurance model remain fit for purpose? To do that, it is helpful to compare our nation with another having similar natural hazard exposure: Japan. Both New Zealand and Japan sit on major fault lines. Both have endured devastating earthquakes and other natural disasters. Both run hybrid systems that combine private insurers with government backing. Yet the two countries have adopted very different approaches to the scope of earthquake insurance cover and the way it operates, influencing the speed and extent of recovery for households and communities. How New Zealand’s system works Residential homeowners who buy private house insurance automatically receive statutory cover administered by the Natural Hazards Commission (formerly EQC). That cover applies to earthquakes and a range of other natural disasters. This basic scheme was established at the end of World War II, providing cover for war damage before being adapted to earthquake risk. The statutory scheme is funded through a flat levy on private insurance policies – homeowners pay the same rate (currently 16c per $100 of building cover), regardless of whether their home is in Auckland, Wellington or near the Alpine Fault. The statutory scheme provides a first amount of cover for residential buildings (typically, up to $300,000) and certain forms of land damage. Any building loss above the statutory cap is met by the homeowner’s private insurer (up to the amount of the sum insured). This system has strengths: It provides a degree of universal protection for insured homeowners It extends to land damage, which sometimes prevents or affects re-building It spreads risk across the population And it is backed by an uncapped government guarantee However, it also has weaknesses, exposed after the Canterbury earthquakes. Multiple events triggered multiple claims, compounding damage. Many policies operated on an uncapped ‘full replacement’ basis, requiring detailed case-by-case assessment. There were also significant problems in the interface between private insurers and EQC. Some 460,000 EQC claims and 168,000 private insurance claims overwhelmed capacity. Assessments generated prolonged disputes over policy entitlements (especially the scope and standard of repairs). Settlement timelines stretched into years. There was also much litigation, with some claims still unresolved one decade later. Legislation in 2023 clarified definitions, raised caps, improved governance, and codified operational practices. Yet the core architecture of the system– including uniform treatment of hazards with different risk profiles and levies insensitive to risk – remains unchanged, despite increasing natural hazard risks and fiscal strain. How Japan’s system works In Japan, earthquake insurance is optional, purchased as an add-on to standard fire insurance policies, or by a number of non-profit co-operative insurers. Private insurers underwrite policies but pass the risk on to a government-backed reinsurance pool known as Japan Earthquake Reinsurance Co Ltd. Losses are shared between the state and insurers, with an overall event-level payout cap designed to protect fiscal stability. Premiums are explicitly risk-based. They vary by location (Japan is divided into three main zones) and by building type (non-wooden vs wooden). Discounts are available for earthquake-resistant buildings and for newer buildings. Pricing therefore explicitly reflects hazard exposure more directly than in New Zealand. Coverage is also deliberately limited. It only applies to residential buildings and personal property, but not land Coverage percentages are capped at a maximum of 50 percent of the fire insurance sum insured Payouts are capped at $568,000 for a building and approximately $113,000 for personal property Damage is assessed by fixed categorisation – total loss, large half loss, small half loss and partial loss – each corresponding to a fixed percentage of the insured amount. The aim is not indemnity for actual loss but rapid payouts and recovery. Insurers do not undertake repairs; they’re up to the homeowner to arrange. The result is fast claim settlements. In the case of Japan’s most catastrophic earthquake (the Great East Japan Earthquake in 2011), 99.5 percent of the 900,000 claims made were settled within 14 months. Some lessons for New Zealand While social, political and economic differences exist, comparison raises important questions in the following areas: Coverage scope Japan’s earthquake insurance does not attempt to cover land damage and only partially covers building loss. This limits systemic exposure and helps to keep premiums within socially acceptable bounds. New Zealand’s broader cover reflects different historical choices. The scheme bears a larger share of catastrophic loss when major events occur. That increases the size and variability of potential payouts, placing greater pressure on levies, the Natural Hazards Fund, and ultimately the Crown guarantee. Risk signalling New Zealand’s flat-rate levy sends weak signals about risk exposure and incentives for reduction or mitigation. While solidarity is an understandable policy choice, the absence of risk differentiation may undermine long-term resilience. Japan’s risk-sensitive premiums and building-type discounts create stronger alignment between insurance cost and mitigation incentives. Fiscal control Fiscal limits can shape public expectations and political decision-making after disaster. Japan operates with an explicit event-level payout cap and layered reinsurance structure. This provides transparency about maximum fiscal exposure. New Zealand relies on the Crown guarantee behind the Natural Hazards Commission, but no limit is in place. At the same time, Treasury has indicated that, over its first five years, income is not expected to meet projected costs. Claims settlement New Zealand’s approach has traditionally been designed to restore the insured to their position before the loss. Japan’s model prioritises speed of settlement of claims, accepting that payments will not correspond to actual repair costs. The result is reduced delay and fewer disputes with insurers over the scope and cost of repairs. Conclusion Earthquake insurance is more than a financial instrument. It is a mechanism for protecting against catastrophic loss. Every design choice involves a trade-off between level of protection, efficiency, affordability and fiscal risk. The next major earthquake will test New Zealand’s system again. When that happens, the extent of entitlements, the speed of settlements, and the stability of public finances will be as crucial as the strength of individual buildings. Looking to another country, like Japan, does not mean importing its system wholesale. But it does encourage us to ask some hard questions now – before the next earthquake forces them upon us. Associate Professor Rohan Havelock presented at the Kanto Regional Meeting of the Japan Society of Insurance Science on earthquake insurance in New Zealand in December 2025.