The Liquefaction Model
The 2010 M7.1 Darfield earthquake in
New Zealand started a sequence of events – the Canterbury Earthquake Sequence
(CES), that propagated eastward in the Canterbury region over several years. Since
the City of Christchurch is built on alluvial sediments where the water table
is very shallow, several of the larger events created widespread liquefaction
within the city and surrounding areas. Such ground deformations caused a
significant number of buildings with shallow foundations to settle, tilt and
Prior to these New Zealand earthquakes, liquefaction was observed but not on this scale in a built-up area in a developed country. As in previous well-studied liquefaction events (e.g. 1964 Niigata) this was a unique opportunity to examine liquefaction severity and building responses. Christchurch was referred to as a “liquefaction laboratory” with the multiple events causing different levels of shaking across the city. However, we had not previously seen suburbs of insured buildings damaged by liquefaction.
The Source Model
The 2010 M7.1 Darfield earthquake in
New Zealand started a sequence of events that propagated eastward in the
Canterbury region over several years, collectively causing upward of 15
individual loss-causing events for the insurance industry. The Insurance
Council of New Zealand state that the total insured loss was more than NZ$31
billion (US$19.4 billion).
With such a significant sequence of events, a lot had to be learned and reflected into earthquake risk modeling, both to be scientifically robust and to answer the new regulatory needs. GNS Science – the New Zealand Crown Research Institute, had issued its National Seismic Hazard Model (NSHM) in 2010, before the Canterbury Earthquake Sequence (CES) and before Tōhoku. The model release was a major project, and at the time, in response to the CES, GNS only had the bandwidth for a mini-update to the 2010 models, to allow M9 events on the Hikurangi Subduction Interface, New Zealand’s largest plate boundary fault, and to get a working group started on Canterbury earthquake rates.
But given the high penetration rate of earthquake insurance in New Zealand and the magnitude of the damage in the Canterbury region, the (re)insurance and regulatory position was in transition. Rather than wait for a new National Seismic Hazard Map (NSHMP) update (which is still in not available), RMS joined the national effort and started a collaboration with GNS Science as well as our own research, to build a model that would help during this difficult time, when many rebuild decisions had to be made. The RMS® New Zealand Earthquake High Definition (HD) model was released in mid-2016.
Across the global risk management community, we are bombarded by new information every day. As risk professionals we have to prioritize how we give our attention to new information. From an RMS perspective, when we release new model insights, we know there is a need to be concise and boil down huge research projects into just the important details. But there is a concern that the top-level results get taken as a uniform value that can be applied across the board, losing vital nuance.
When RMS released its New Zealand Earthquake High-Definition (HD) model in mid-2016, an important message was that the annual average loss (AAL) had increased by 30 percent. The ground-up, all-lines, countrywide AAL increased 30 percent relative to the previous version of the model released in 2007. An increase in loss came as no surprise after the Canterbury Earthquake Sequence of 2010/11 – see our New Zealand earthquake blogs.
The HD model was launched at two industry seminars in Wellington and Auckland and came with online documentation: some 44 pages of Understanding Changes in Results and 114 pages of model methodology, supplementary materials on our RMS OWL client portal and a team of modelers happy to talk about their work.
Faced with this information, one approach is to note that the New Zealand market is very consolidated so industry figures should be useful guides for actual portfolios. Let’s just use the old model and scale it by 30 percent. “She’ll be right”, as they like to say in New Zealand. But with two models being so different, this scaling-up would not make sense. Why are they so different?
There has always been a balance between cross-subsidy and property-specific, risk-based underwriting and pricing in insurance, particularly for homeowners’ policies. While an actuary can easily quantify differences in fire risk for houses constructed from wood versus concrete based on claims, this becomes much more difficult when the peril concerned is infrequent, such as for earthquake or flood. Clearly risk models help to bridge this gap, but facilitating a move from cross-subsidy to risk-based pricing is more complex than simply using risk analytics. Factors such as regulation, market conditions, distribution channels and insurer IT systems all determine whether individual insurers and markets will move towards greater differentiation of risk. This is not to mention the political dimension of insurance affordability and social equity.