Corner of Berkshire & Fairfax Message Board

General Category => Fairfax Financial => Topic started by: FairFacts on March 19, 2018, 09:06:57 AM

Title: Climate Change and Cat Losses
Post by: FairFacts on March 19, 2018, 09:06:57 AM
I have a nagging concern in the back of my mind that climate change will perpetuate the weather related cat. losses. What impact will this have on longer-term underwriting? Will it hasten hard insurance markets?
Title: Re: Climate Change and Cat Losses
Post by: johnpane on March 19, 2018, 09:25:08 AM
Interesting you should mention this. I have noticed an increasing trend of my neighbors reporting they have lost their cats (and dogs), but it has not seemed to be related bad or good weather.
Title: Re: Climate Change and Cat Losses
Post by: Cigarbutt on March 19, 2018, 10:16:47 AM
Climate change is a tough topic but Ill give you my present thinking concerning the potential impact on catastrophe insurers and reinsurers. I have references if you need them.

The long term trends in costs (inflation-adjusted) are going up but there are many variables. The obvious ones are: increasing population density and urbanization, population migration and concentration in coastal areas and generally increasing economic activity. At this point, the consensus is that climate change might be an independent contributor to the trend (more frequent and more severe events). 2017 events might trigger this line of thinking but one has to be careful about recency bias.

But the climate change effect (whatever it is) will likely manifest over decades. Catastrophe modelling is historical in nature and will tend to naturally   incorporate more recent trends. For the insurance industry with cat exposure, IMO this is worth watching but more as a long term threat, especially if the trend becomes clear and accelerates.

So, to answer your question, climate change related events are unlikely to trigger a hard market soon. Natural variability in catastrophe activity and human volatility in financial markets could however. Still looking for a model to explain the latter.
Title: Re: Climate Change and Cat Losses
Post by: walkie518 on March 19, 2018, 10:34:23 AM
From what I gather, over the last couple decades, underwriting has taken into account the increasingly intensity and frequency of cat events.  While the last decade has certainly been bad, it's been getting worse for a while. 

What one might want to consider, however, is if flood is privatized in the US. 

Insurers want the business but the impact on the housing market could be substantive, especially in poorer areas with flood risk.

Title: Re: Climate Change and Cat Losses
Post by: sleepydragon on March 19, 2018, 10:35:22 AM
One way to hedge against global warming is to make sure your house is located in high elevation neighborhoods. When the once per 100/50-year flood comes and destroy all those houses in low places, your house will worth more. This is what happened in Texas I think.
Title: Re: Climate Change and Cat Losses
Post by: sleepydragon on March 19, 2018, 10:37:18 AM
From what I gather, over the last couple decades, underwriting has taken into account the increasingly intensity and frequency of cat events.  While the last decade has certainly been bad, it's been getting worse for a while. 

What one might want to consider, however, is if flood is privatized in the US. 

Insurers want the business but the impact on the housing market could be substantive, especially in poorer areas with flood risk.

You cant diversify the flood risk and only people in flood zones will buy. So no private company want to sell insurance without govt help
Title: Re: Climate Change and Cat Losses
Post by: Cardboard on March 19, 2018, 10:38:16 AM
2006-2016: no major hurricane hit the U.S. The only noticeable one is Sandy which was a large storm. So we must have some people who did not pay attention regarding frequency.

This is highly contrary to what was thought to happen following devasting hurricanes in 2004 and 2005. This is an incredibly long period of time without such event while we have been told that some of the highest temperatures were recorded in some of these years.

If you want to consider a period of 11 consecutive years to be an outlier, no point in continuing this discussion. However, I would be very careful to simply use 2017 and project forward.

Cardboard 
Title: Re: Climate Change and Cat Losses
Post by: RichardGibbons on March 19, 2018, 11:54:25 AM
2006-2016: no major hurricane hit the U.S. The only noticeable one is Sandy which was a large storm. So we must have some people who did not pay attention regarding frequency.

This is highly contrary to what was thought to happen following devasting hurricanes in 2004 and 2005. This is an incredibly long period of time without such event while we have been told that some of the highest temperatures were recorded in some of these years.

If you want to consider a period of 11 consecutive years to be an outlier, no point in continuing this discussion. However, I would be very careful to simply use 2017 and project forward.

Yeah, I think this is a great example of the unpredictability of the effects of climate change.  I mean, someone who's ignorant about meteorology might believe that high temperatures are the only factor that impacts the strength of hurricanes, when other factors like wind shear are just as important.  Similarly, changes in wind patterns and changing ocean currents can dramatically affect where hurricanes will appear.

So, I think you're right, Cardboard. Climate change results in unpredictable changes around the world--hurricanes, flooding, and droughts. Since it's hard to do prediction accurately, there's value in reducing CO2 emissions to slow the rate of change to reduce the massive negative impacts of climate change and allow the models to catch up to the impacts of climate change on our world.
Title: Re: Climate Change and Cat Losses
Post by: maybe4less on March 19, 2018, 12:47:11 PM
2006-2016: no major hurricane hit the U.S. The only noticeable one is Sandy which was a large storm. So we must have some people who did not pay attention regarding frequency.

This is highly contrary to what was thought to happen following devasting hurricanes in 2004 and 2005. This is an incredibly long period of time without such event while we have been told that some of the highest temperatures were recorded in some of these years.

If you want to consider a period of 11 consecutive years to be an outlier, no point in continuing this discussion. However, I would be very careful to simply use 2017 and project forward.

Cardboard

Why are you discounting Ike in 2008?

Also, a little disingenuous to exclude Sandy. It hit the US mainland with hurricane force winds and is apparently the fourth costliest storm on record.
Title: Re: Climate Change and Cat Losses
Post by: Cardboard on March 19, 2018, 01:05:16 PM
My bad, forgot about Ike.

However, it has still been a much more quiet period than what was predicted by the non-ignorant people or scientists.

I can tell you that it was pretty scary to invest in Fairfax in 2006. A somewhat repeat of the 2005 hurricane season and many of the people who made a lot of money with these options would not have fared so well. Count myself included.

More storms due to global warming (which is really the proper way to call it since we are talking about greenhouse effect) was really on the back of people's mind investing/trading in Fairfax at the time.

Now, will this change and will we see category 4 and 5 hurricanes now every year going forward? Could be.

Cardboard

Title: Re: Climate Change and Cat Losses
Post by: rb on March 19, 2018, 06:40:47 PM
I don't think increases in CAT frequency or severity will impact insurance companies that much. Policies have a finite life. As the cat losses change so will premiums. So the underwriters will likely be ok. The insureds? Who knows?
Title: Re: Climate Change and Cat Losses
Post by: FairFacts on March 21, 2018, 05:35:25 AM
I found the following report that goes into some detail about one re-insurers (Swiss RE) approach to climate change risk. Its worth a read for anyone interested...

http://reports.swissre.com/2016/financial-report/responsibility/natural-catastrophes-and-climate-change.html# (http://reports.swissre.com/2016/financial-report/responsibility/natural-catastrophes-and-climate-change.html#)

also, for anyone doubting climate change, this is NASA's position, the graph on CO2 levels is scary....

https://climate.nasa.gov/evidence/ (https://climate.nasa.gov/evidence/)

Title: Re: Climate Change and Cat Losses
Post by: Cardboard on March 21, 2018, 05:51:47 AM
Like I tried to explain, so far zero correlation. That is not to say that it will not change.

However, if you are convinced that global warming will lead to more/larger disasters in the not so distant future (if not this discussion is useless in terms of investing if happening only a few decades from now): hurricanes, sand storms, drought, etc. then you should stay away from insurers and reinsurers because these guys are always reactive and then adjust their policies accordingly based on recent past experience.

Even Buffett who has been scared to death about nuclear weapons since the 50's or 60's did not think it was necessary or missed to include an exclusion in his policies on nuclear terrorism until after 9-11. Should tell you a lot.

Cardboard
Title: Re: Climate Change and Cat Losses
Post by: Cigarbutt on March 21, 2018, 07:06:18 AM
I would say that there has been a strong tendency to make an association between costs related to "weather" events and "climate" change. Association is not automatically cause and effect. But it sells well.

Probably a long term issue but worth keeping an eye on.

Some interesting "facts":

-Studies have built a relatively convincing case that wind storm activity has increased in the North American Basin since the 1970's BUT this has not been replicated in other areas in the world and results may still be within statistical noise.

-Storm activity as measured in many studies has increased in the last 20 or 30 years but these levels are comparable to what happened about 100 years ago.

-Studies clearly show an increasing trend in storm activity of lower intensity and this is "associated" with the notion of higher water temperatures but this trend can be essentially explained by improved measurement equipment and technology.

However, this needs to be watched because, if there is a trend, the trend may not be linear and factors may compound.

Example:

With the Superstorm Sandy, it has been estimated that a significant part of the insured losses were related to higher sea levels. The trend in sea level change is significant and it is hard to see how this trend will change. This is a long term trend and it is not essential for your investment decison to establish if this is related to climate warming or not. Sea level rising levels is just a fact. If you think of what is happening in the Netherlands for instance, you may see that a rising sea level will not have a linear effect in terms of the consequences as the measures taken to control floods may spectacularly fail at some point. It may be possible at this point to model how a 1mm level of sea level rise could result in a certain amount of insured damages to exposed areas such as New York or Florida. Potential black swans.

FWIW, I think that this "factor" (climate change) should not be a major variable in your decision. The underwriting cycle is much more of an issue. Historically, after earthquakes, people have kept coming back progressively closer to the volcano as the quality of the arable lands (lava) was inversely proportional to the center of the eruption center. People (like insurers) tend to forget how things can get nasty after quiet periods. It is just human nature.

As always, it is likely a good idea to look at insurers and reinsurers that keep a reasonable cap on losses (think Allied World), that have a diversified book of business and that have the capital buffers in place to deal with "surprises".




Title: Re: Climate Change and Cat Losses
Post by: Cigarbutt on November 05, 2018, 06:25:13 AM
Here's a more recent reference:
http://www.chrisleithner.ca/newsletter/2016-2018/nov18_newsletter.pdf
The author seems to be on a mission and there are red flags vs objectivity but the part covering climate change and catastrophe loss exposure is relevant and useful IMO.

The underlying fundamental point is that insured losses have come up over time and this increase can be explained, at least partly, by urbanization and growing concentration of property value but, as a ratio, insured losses to GDP has come down. So, despite headlines which tend to link catastrophes to "climate change" and tend to be sensational (as they should be to a certain degree), the "losses" profile related to catastrophes (human and financial) has been improving.

The major risk remains unpredicted, and sometimes unaccounted for, variability. Work in progress.