Preparing for a 1-in-1,000 year loss: Insurance resilience 10 years after the Christchurch earthquake
By Laura Barksby
Not many countries have experienced a catastrophe that directly hit one of their largest cities, but for New Zealand, on February 22, 2011, an M6.2 earthquake struck less than 6 miles (10 kilometers) from the Christchurch central business district. The event generated significant ground motions and unprecedented and widespread liquefaction. Furthermore, it was the most destructive event of the Canterbury Earthquake Sequence (CES), a series of earthquakes across the region that started with the M7.1 Darfield Earthquake on September 4, 2010, and continued through 2011.
As of today, the CES is the world’s second-costliest insured earthquake loss in history and supports the New Zealand regulator’s 1-in-1,000-year capital requirement. Let that sink in. After 2011’s enormous M9 Tohoku Earthquake and Tsunami in Japan, the world’s next-largest earthquake insurance loss occurred in New Zealand – from a series of events with relatively moderate magnitudes, in a country with a population of around five million, and in a city with around 500,000 residents.
The considerable insurance losses arising from the CES resulted from both the significant damage in Christchurch, particularly the February 2011 earthquake, and the high insurance penetration in New Zealand. The latter is largely thanks to a partnership between the insurance industry and the country’s Earthquake Commission (EQC), which provides insurance for over 95 percent of residential properties. As such, three of the world’s top six reinsurance programs cover New Zealand.
The CES was a test for the insurance market, and a small number of more regional and specialized insurers did become insolvent. This reiterated the need for insurers to have adequate financial reserves in the event of a significant earthquake. To reinforce this, in September 2016, the Reserve Bank of New Zealand’s (RBNZ) capital charge was calibrated to a 1-in-1,000-year loss – one of the longest regulatory return periods globally.
If the CES Occurred Today
Let’s consider a repeat of the CES today on a nationwide portfolio – how often could you expect to experience such a loss? On today’s exposure, a modeled loss equivalent to the February 2011 Christchurch Earthquake could be anticipated on average every few hundred years. So, clearly, while the Christchurch event was significant, its loss today would be much lower than the RBNZ’s capital requirement. Even if the losses from the three most damaging earthquakes in the CES were combined, the overall loss today still does not equate to a 1-in-1,000-year event.
Clearly, while the CES was very damaging, in most cases the losses did not reach such high return periods. What if, as an insurer, you are required to have enough capital reserved for something even more devastating? Naturally, this prompts the following questions: What could a 1-in-1,000-year loss look like in New Zealand? How do you manage risk to such a return period? These are questions that the insurance industry in New Zealand is having to consider.
In Christchurch, for an earthquake to trigger the regulatory return period, the event would need to be much more severe than in February 2011. Looking to North Island, strong events under Auckland and Wellington could exceed the 1-in-1,000-year return period. Potential events in Wellington include an M7.7 event on the Wellington Fault that runs directly under the city, as well as an M8.4 event on the nearby Wairarapa Fault. The last significant event on the Wairarapa Fault was in 1855; while the return period for this earthquake is very long, a repeat of this event is considered by insurers to be a worst-case scenario.
Looking further afield, there are other faults capable of generating solvency-testing events. A full rupture of the Hikurangi Subduction Zone, to the east of North Island, could generate up to an M9.2 earthquake. Sophisticated modeling of this fault enables the risk in Wellington to be captured, which, despite its distance from parts of the subduction zone, could experience severe damage in such an event.
Other capital-triggering earthquakes include complex, multi-fault ruptures. These events rupture multiple faults together, often resulting in larger event footprints, and spreading damage over a wider area. Examples include earthquakes that rupture over hundreds of kilometers, crossing both North and South Islands as well through Wellington. Such earthquakes not only have the potential to cause severe damage, but they also increase the likelihood of geographical loss correlations. The RMS® New Zealand Earthquake HD Model is the first in the industry to explicitly model these complex ruptures.
The CES losses were equivalent to approximately 20 percent of New Zealand’s GDP. Imagine what this percentage could be from a 1-in-1,000-year event. Not only would these events have a significant economic impact, but they would have considerable social consequences. In addition, post-event loss amplification would be notable as restoration times would increase, communities would have to relocate, lifelines and communication services would be affected, infrastructure networks would go down – and the list goes on.
Continued Resilience After the CES
Of course, the likelihood of these events occurring is very low. In the past 10 years, we have seen notable seismicity across New Zealand, but prior to the 2011 Christchurch Earthquake the last major loss was the 1931 Hawkes Bay event.
The Christchurch Earthquake certainly tested the insurance market, but there was a determination to ensure continued resilience. Following the CES, RMS collaborated with the insurance market as well as the scientific and engineering communities to develop the New Zealand Earthquake HD Model. Our involvement continues today with participation in the next update to the national seismic hazard map.
Vivid memories of February 22, 2011 do remain. But – over the last 10 years and moving into the future – New Zealand’s combination of high insurance penetration, comprehensive regulation, and sophisticated modeling all help insurers to understand and manage the challenge posed by seismic risk.
Is this page useful?Yes No Report an issue on this page
Thank you. If you have 2 minutes, we would benefit from additional feedback (link opens in a new window).