Has the Re/Insurance Industry Missed a Major North American Risk?

June 30, 2022


By Tom Johansmeyer and Ted Gregory | June 6, 2022

The United States remains the largest peak peril risk area in the global re/insurance market and accounts for the bulk of the reinsurance that’s traded. Over the past few years, though, we’ve seen more risk transfer transactions that account for cross-border catastrophe events, focusing on the United States and Canada. Nearly 20 percent of all PCS Canada catastrophe events are cross-border events with PCS U.S. events. While PCS has seen several such events occur, none has been large in both countries. The potential for large catastrophe events on both sides of the border exists, and most re/insurers have begun to manage their risk and capital to account for it. The broadening perspective shows progress, but it also highlights a missed opportunity

Catastrophes along the southern border of the United States could pose a greater threat to re/insurer capital than those along the north. The past few years have shown how susceptible Mexico is to catastrophe activity. PCS Mexico data shows four new events since our launch in 2019, not to mention at least another 20 going back to 2010. The country has tropical storm exposure on both coasts and is impacted regularly. This year alone, PCS designated three events in Mexico in 2021, with another one in 2020.

North American Cross-Border Hurricanes Since 2010
EventYearBorderU.S. %MX %CA %

Source: PCS, a Verisk solution (including PCS Canada and PCS Mexico)

For smaller events, such as Hurricane Dorian, the loss between the United States and a neighbor country can be more balanced, as we saw with 28 percent of the aggregate two-country loss coming from Canada. For larger catastrophes, though, the portion going to the United States tends to be far larger. In 2020, more than 3 percent of the Hurricane Delta loss went to Mexico. In 2012, only 1 percent of the Hurricane Sandy loss went to Canada. If you go back a little further, though, you’ll find some instances of cross-border catastrophe events that show more balance. Hurricane Wilma, for example, was a multibillion-dollar event on both sides of the border. Fourteen percent of the aggregate two-country loss went to Mexico (using data for Mexico gathered on an ad hoc basis). And with three tropical events designated as PCS Mexico catastrophe events in 2020 and 2021, the risk of another such two-country loss persists.

The main reason why we have yet to see a U.S./Mexico catastrophe bond is the belief that there isn’t enough exposure in Mexico, according to client conversations. That said, it seems likely that the global re/insurance market will begin to take closer looks at the risks they face around the world, given the surprises that have arisen over the past decade. Since 2016, when we took over the leadership of PCS, we’ve seen several catastrophe events and trends that were virtually without precedent:

Wildfire (2016, 2020): In 2016, the Fort McMurray event in Canada became the largest industrywide insured loss event from wildfire in PCS history for both Canada and the United States. An outlier at the time, the increase in wildfires since then has pushed what was once the largest such event out of the top five. Further illustrating how much has changed, 2020’s 17 wildfire events made it the most active wildfire year in PCS’s 72-year history. Four states – California, Oregon, Washington, and Colorado – sustained more than US$12 billion in industry-wide insured losses.

Istanbul hail (2017): When we launched PCS Turkey, we were told by many reinsurers that the only risk they worried about was earthquake, although they’d concede that flood could be problematic. After designating our first new event since launching in 2016 (which was related to political violence), we designated two hailstorms in 2017 as our first natural catastrophe events since inception. The larger of the two is the second largest industrywide insured losses we have for Turkey since 1999.

Multibillion dollar U.S. hail (ongoing): Large severe convective events were considered a rarity until an uptick in the frequency arose around 2016. During the first half decade between 2011 and 2021 (2011 – 2015) there were 25 wind and thunderstorm events exceeding US$1 billion. During the second half of that decade (2016-2021) there were 35 wind and thunderstorm events above that threshold, representing a 29 percent increase from the first five-year period to the second.

Significant meteorological events during uncommon months (2020): Historically, December tends to have fewer, and lower severity catastrophe events than most other months. However, in December 2020, the insurance industry experienced two major, multi-billion-dollar wind and thunderstorm events, along with a Midwest derecho and a tornado outbreak in the quad-states area, producing uncharacteristic levels of losses in a month not typically known for major cat events.

Multibillion dollar political violence (ongoing): The Los Angeles riot of 1992 was the largest industrywide insured loss for riot and civil disorder on record with PCS until 2019. In fact, it was significantly larger than all prior riot and civil disorder losses combined. That changed with the Chilean riot of 2019, which became the first multibillion dollar riot and civil disorder event on which PCS reported. The second and third came in the United States (2020) and South Africa (2021, with informal PCS reporting only). PCS projected ultimate losses for several specialty classes of business show that the conflict in Ukraine has made 2022 the fourth consecutive year with aggregate industry-wide insured losses from political violence over US$2 billion, with the potential to go much higher.

There are more examples, as well. Many. The notion of the “unprecedented” or “surprise” catastrophe event means less and less every time we see one. And that’s what brings us back to Mexico. We’ve heard enough client discussions about the region since launching PCS Mexico in 2019 to know that industry concerns are growing. Hurricane Wilma showed us that multibillion-dollar losses in Mexico are indeed possible. With this in mind, it seems the age of the cross-border U.S./Mexico catastrophe bond or industry loss warranty may be upon us.