After tumbling last week on concerns that between damage from Harvey and Irma, losses for the P&C space would be devastating, today the broader insurer space is breathing a sigh of relief after the Hurricane's damage reportedly underwhelmed, especially following some especially dire observations over the weekend from the likes of Torsten Jeworrek, member of the board of the German reinsurance giant Munich Re, who on Sunday said that Hurricane Irma is proving to be a “major event” for Florida and the insurance industry.
As Reuters reported yesterday, Jeworrek and other insurance executives gathered in Monaco for an annual conference to haggle over reinsurance prices and strike underwriting deals. And even though Irma eventually skirted densely populated Miami, the Munich Re board member said that “Irma is still a major event for Florida and also a major event for the insurance industry.” When asked by journalists, Jeworrek also hazarded a rough guess for the insured losses for the global industry of Hurricane Harvey: he said that losses were estimated at between $20 billion and $30 billion, which would put the storm on a scale of Sandy.
That number will likely prove to be overly optimistic, because as Goldman showed yesterday, many estimates of Harvey damages rose sharply during the first week after landfall; however, most have now settled in the $70-100bn range (Goldman itself assumes $85bn). While the uncertainty around these figures remains high, it seems clear that Harvey’s aftermath will be particularly severe.
So with Harvey more or less quantified, the question then is what will Irma's damage be? And it is here that some initial commentary from Credit Suisse explains (and may have sparked) a broad rally across the insurer sector.
In an overnight note from CS' Ryan Tunis, the Swiss bank analyst writes that even as Hurricane Irma barrels up Florida’s Gulf coast, "insured losses are now expected to be less than many feared with modeling agencies in general agreement of a sub $60b number."
As a result, Credit Suisse's takeaway is that while Irma may be costly to reinsurer book values, the hurricane alone should not trigger the need for additional capital, and further expects tangible book value hits below 15% from Irma and Harvey combined. Of course, Tunis concedes that the capital raise conversation would likely be more pertinent should there be a "next catastrophe this year" or even another bad wind season in 2018. While projections for Hurricane Jose aren’t showing anywhere close to the amount of landfall risk that Irma had, but there are still a few projections of a potential threat to the US Northeast.
The key thing that saved insurers, however, and which resulted in lower insured losses is the last minute western shift in Irma’s FL path:
As destructive as Irma's path continues to be, we think that the path up the west coast of Florida is better from an insured loss standpoint than projections from mid last week and even Friday that involved a more direct hit of Miami. As of 5pm today (Sunday), the storm seems to have weakened to a cat 2 as it has moved over Fort Myers, FL. AIR Worldwide estimates that there is $1 trillion of coastal exposure along west coast up to Tampa, 30% less than the $1.5 trillion of coastal exposure in the Miami Tri-County area. A west coast landfall should also bring more of the insured loss burden onto the personal lines companies. For a Miami-Dade hit, Lloyds industry losses assume 49% are from residential property and 49% from commercial property (rest auto and marine). For the Pinellas county (next to Tampa), residential property losses make up 70% of losses, while commercial property are 29% of losses.
Another question: how much flood is privately insured? According to Credit Suisse, it's unclear at this point what portion of the damage will be caused by storm surge flooding versus wind, but Tunis believes that it's a higher percentage than what the market was expecting on Friday.
"The private insurance industry should be far less exposed to flood damage than wind, especially on the commercial side where we understand that private flood insurance is either rarely offered or heavily sublimited, which we would think would be even more true in flood prone spots. The vulnerability to storm surge to Fort Myers, Tampa and St. Petersberg has also been very well understood by the cat modeling services (Karen Clark cited Tampa as #1 most vulnerable city to storm surge in 2005).
In this context, one of the debates in the coming weeks will center around how much of the homeowners flood damage is privately insured, especially since private flood insurance losses from Irma are likely to exceed that of Harvey.
We understand that though standard home policies do not include flood insurance, high net worth home policies are often inclusive of flood protection. Because of the high insured values of homes in FL (see Figure 1), there are more homes with high net worth home policies which could include private flood insurance.
Once the policyholder goes to the higher net insured value policy market – and we fear that many may in this case given high average property values of affected areas – we understand that policyholders have the option to purchase flood insurance. We have heard estimates that 50% of high net worth policy homes buy private flood insurance. Companies that may be exposed to this include AIG, CB, and Pure.
What about reinsurers? It is here that much of the hit will likely be, as Barclays warned two weeks ago:
[T]he reinsurers will be on the hook for the vast majority of privately insured losses. The most recent RMS report (Sunday afternoon) indicated that there is a 90% chance that the insured wind loss will be less than $60b which is down from the figure estimated Friday of 90% chance of sub $75b losses. That number does not include insured loss damage from flooding, the $10-15b of Irma damage in the Caribbean, business interruption losses in commercial insurance, or post event loss amplification issues such as demand surge. Offsetting this is our understanding that the Florida hurricane catastrophe fund (FHCF) may pay out around $15b of the total in a $40-60b loss scenario.
Putting these revised estimates in context, Credit Suisse writes that at this point, this does not appear to be a 1-in-100 Atlantic Hurricane event.
As a result, we are comfortable using company disclosed 1-in-100 year average loss disclosures as a worst case ceiling for individual company loss estimates. This number for the reinsurers (1-in-100 year US wind PMLs) appears to be 5-10% of tangible equity. If we assume 2-3% book value hits from Harvey, then this implies 7-13% hits to book value from both hurricanes. Reinsurers are believed to be relatively unscathed from Harvey, but one other incremental negative data point from this weekend was a new RMS estimate for Harvey losses which implies $18-25b of privately insured losses Harvey, which may be 30% higher than preliminary estimates perhaps leaving reinsurers more vulnerable.
What this means for insurers is basically good news, as today's stock price reaction confirms:
As we have noted, net exposures should be relatively limited for PGR, ALL, TRV and HIG either through avoidance of Florida property or heavy reinsurance purchase, which is still the case. The path of the storm and expected storm surge does bring a higher level auto losses into play, though likely not at the scope of Harvey with plenty of advanced warning for all Florida residents and businesses. Katrina may be the best comparison at this point, which resulted in $2.8b of industry auto losses in today's dollars and around $328mm for PGR. PGR reported $258.5mm ($328mm in today's dollar) of personal auto from Katrina, Rita, and Wilma. ALL reported $283mm ($359mm in today's dollars) of personal auto losses from Katrina, Rita, and Wilma. For ALL, the auto losses were about 8% of total Katrina losses. We note that PGR and ALL have 14%/10% market share of Florida auto physical damage (See Figure 4). PGR also has 16% of the commercial auto market in Florida.
* * *
Finally, adding fuel to the P&C surge was a similar analysis by Citi, according to which the cost of Hurricane Irma-related total estimated losses has been slashed to $50BN from $150BN.
"While the hurricane season is only half way through, so that number could increase but, for now, P&C insurers and reinsurers have dodged a bullet" Citi analyst James Naklicki wrote. As a result, Citi projects U.S. insured loses of $20 billion. Furthermore, citi also expects less selling pressure on reinsurers as the risk of capital raising eases, and remains overweight Travelers and Chubb; underweight Axis, RenRe and XL, while upgrading Axis Capital Holdings to Neutral from Sell.
End result: a surge in insurer stocks, more than undoing all of last week's Irma-related losses.