Shaky View

Despite the fact that parts of Canada are in earthquake country, it is easy to forget the country is exposed to this infrequent peril.

Recently, Hollywood provided a reminder of earthquake devastation via the movie San Andreas, wherein the Hoover Dam, Los Angeles and San Francisco are all shaken to the ground by a series of earthquakes. While the movie is a complete work of fiction, its depiction of what can happen during and after an earthquake offers food for thought, and may also help to promote risk awareness, perhaps even in a country where insurance uptake, at most, is about 60%.

Google reports that more people used the search phrase “earthquake insurance” in May, after the movie was released, than in April, after the Nepal earthquake. Certainly a disaster movie reaches a wide audience. Based on a US$53 million box office gross, more people watched San Andreas in its opening weekend than the approximately 1.1 million Californian homeowners who have an earthquake insurance policy.

FICTION TO FACT

The possibility of a damaging earthquake in Canada should not belong in the realm of fiction. The British Columbia coast runs parallel to a major tectonic boundary. Part of this was the source of a mega-earthquake in 1700, big enough to cause tsunami damage thousands of kilometres away in Japan.

Given the current scientific knowledge, such an earthquake should not be a surprise if it were to reoccur tomorrow. Smaller faults associated with the plate boundary extend into the earth’s crust beneath Victoria, Vancouver and Seattle – some of which may not even be known.

Smaller unknown faults like these can be the source of devastating earthquakes. Consider, for example, the 1994 Northridge earthquake in California, the 1995 Kobe quake in Japan and the 2011 Christchurch tremblor in New Zealand, with estimated economic losses of US$48 billion, US$128 billion and US$21 billion, respectively, notes information contained on Swiss Re’s sigma explore database.

Canada also has its own version of the United States’ New Madrid scenario, courtesy of the Charlevoix seismic zone that is close to Quebec City. Earthquakes in locations like this, far away from a plate boundary, can lead to damaging shaking over huge areas (for example, as far away as Toronto). Low-probability/high-severity events such as these can have a major influence on the financial stability of the insurance industry.

Awareness levels

At first glance, there is good evidence British Columbians have more awareness of earthquake risks than Californians have with their library of disaster movies. An October 2013 study commissioned by Insurance Bureau of Canada, conducted by AIR Worldwide, notes that residential earthquake insurance penetration is around 60% in B.C. and 85% for commercial property. This compares to a Californian residential take-up of just over 10%.

Relatively high insurance penetration is a good thing for the economy of B.C., since the greater part of the disaster cost will be borne by external funds.

While B.C. surpasses California at this first hurdle of getting people insured, the next hurdle is to establish whether or not those providing the insurance are risk-aware enough.

After Northridge in 1994, many insurers were surprised by their losses and effectively pulled out of the residential earthquake market. Such a change in the market had consequences for the policyholders and can be linked to the decrease in take-up rate from 30% to 10%.

It is hoped such surprise losses can be avoided in the event of a future Canadian earthquake. In the last 20 years, catastrophe models have come a long way in helping stakeholders to anticipate potential damage from shaking.

However, without market experience, there is always uncertainty about how insured losses may evolve following an earthquake. This aspect of loss is more difficult to capture in models.

Lessons learned

As a December 2011 article in Canadian Underwriter indicates, the Christchurch earthquakes of 2010 and 2011 are case studies with some relevance for Canada. Christchurch is New Zealand’s third largest city, with similarities to Vancouver or Victoria in its building stock, seismic hazard level, earthquake history and soil type.

The most damaging of the Christchurch earthquakes was a magnitude 6.3 event that hit the city on February 22, 2011. It left the downtown area looking now like an abandoned urban wasteland, caused an economic loss of around US$21.5 billion and an insurable loss of about US$17 billion.

Swiss Re estimates that a similar scenario of a magnitude 6.3 earthquake beneath Vancouver would cause $50 billion in economic damages, notes the company’s 2014 report, Small quakes, big impact: lessons learned from Christchurch.

Preparedness

Two aspects of the Christchurch earthquake and its aftermath are important to consider with respect to Vancouver’s preparedness.

The first is liquefaction, a phenomenon of earthquakes whereby shaking causes firm soils to lose their coherence and behave like a muddy liquid. Buildings that do not sustain any material damage from shaking can become total losses as a result of the sinking or tilting of their foundations into this soil.

In Christchurch, liquefaction was widespread and an unexpected loss contributor for insurers. The Vancouver area, especially the Fraser River delta, is a textbook example of somewhere prone to liquefaction. Insurers should be aware of this hazard and incorporate it into their risk view and post-event planning.

The second aspect of the Christchurch earthquake to highlight is the claims burden when insurance penetration is high. From one day to the next in Christchurch, the insurance industry found itself with hundreds of thousands of claims and only a few qualified adjustors.

The scale of structural damage from earthquakes can easily be underestimated during an initial inspection, leading to multiple revisions of some claims. As such, the claims resolution process has been long, complex and expensive in Christchurch, continuing into 2015.

The insurance penetration of Vancouver exposes the city to a similar challenge of resolving an unprecedented number of claims.

A robust plan for resolving those claims with limited resources will be valuable. Such a plan should contain enough redundancy to overcome realistic problems, including the airport being inaccessible because of liquefaction.

POLICY COVER

Moving beyond Christchurch experiences, a potential source of loss accumulation worth highlighting is strata loss assessment coverage. This type of policy covers individual condo owners or homeowners against assessments from their strata corporation on retained losses.

It is increasingly common to find more restrictive coverage on strata insurance policies – for example, high deductibles of 15% to 20% – that would increase the potential flow into loss assessment coverages.

Variable take-up of loss assessment coverage among a strata community could lead to a messy claims situation with potential for disputes about who is responsible for what.

Correctly modelling and quantifying the loss potential for an insurance portfolio with these coverages requires technical expertise and detailed exposure data.

It is likely that some of the market underestimates the potential accumulation of loss that this could generate from an earthquake.

LOOKING EASTWARD

While Vancouver can look to Christchurch to help better prepare for an earthquake, Ontario and Quebec may need to consider a scenario far worse. Damaging earthquakes have and will happen in the Quebec region.

The probability is lower, but the potential impact is worse than for an earthquake of the same size in B.C.

Older buildings combined with the lower attenuation of seismic energy in eastern North America means that the damage from an earthquake can extend over a very wide
region. A good proportion of the building stock was not constructed with earthquake safety in mind, and the footprint of damage could cover hundreds of kilometres.

A large part of economic loss would not be covered by insurance, since residential earthquake insurance penetration is only about 5%.

This can, hopefully, change by increasing risk awareness and careful underwriting judgment.

Wood-frame houses bolted to their foundations are insurable risks for earthquake. The California Earthquake Authority, for example, has a sub-limit on damage from masonry chimneys as a method for controlling risk of this common source of damage.

The year of construction and soil type also can have big influences on earthquake damage and can be incorporated into the underwriting process. The collection and management of detailed exposure data can be utilized to manage accumulation potential and limit reinsurance costs.

In 1663, Quebec was hit by a strong earthquake that some of the early French settlers considered to be a punishment for disobedience. Such harshness is not necessary in the event of a future earthquake, but it must be acknowledged that there is more to be done to prepare for a recovery.

Stakeholders can build on every opportunity to promote risk awareness and be ready for an earthquake. There may not be a hero to save everyone, as was the case in San Andreas.