Fairfax underwriting track record

[giofranchi]

I hope Andy Barnard is the answer but he's a reinsurance guy managing insurance operations and it's not quite the same skill set.

 

Tommm50,

could you please elaborate a bit more on this one? I believe that insurance, reinsurance, and even investing require the same skill set.

 

For us the underwriting of our specialty insurance risk is really the same process as the fundamental analysis necessary to make sound investment decisions.

--Steven Markel, 2008

 

Thank you,

 

giofranchi

[Tommm50]

I hope Andy Barnard is the answer but he's a reinsurance guy managing insurance operations and it's not quite the same skill set.

 

Tommm50,

could you please elaborate a bit more on this one? I believe that insurance, reinsurance, and even investing require the same skill set.

 

For us the underwriting of our specialty insurance risk is really the same process as the fundamental analysis necessary to make sound investment decisions.

--Steven Markel, 2008

 

Thank you,

 

giofranchi

 

Insurance underwriters underwrite individual risks, their managers organize and quality control that process. Reinsurance underwriters underwrite the insurance companies they reinsure. That's a much different perspective. The best reinsurance underwriters are those that have insurance underwriting experience (optimally in the business they're reinsuring) but you'd be surprised at how uncommon that is. Most reinsurance underwriters I've known grew up on the reinsurance side. Their underwriting knowledge is based more on what they're told vs actually done themselves.

 

I agree underwriting insurance or reinsurance or investing should all use a careful analytical approach. The process in general is similar.  When you get down to the detail level though the questions you're asking to underwrite Long Haul Truckers, Legal Professional, Environmental Liability, or D&O are much different, requiring specific expertise and experience. I'm sure Andy is a very good manager and at his level that's the most important quality. My point was more attuned to the first paragraph.

[vinod1]

To answer your question about my view of the market cycle I'm attaching a speech I gave to an insurance company attorneys' group on just that issue. Understand I'm just a guy who's been in this business a while and trying to give the attorneys a broader view of the environment in which they operate. It's not designed for this forum so you'll have to jump back and forth between the text and the slides. Bottom line: I don't see insurance companies being able to rely on "hard markets" to help them. They've got to manage themselves as it the past three years (and I mean Zenith) will go one forever.

 

Tommm50,

 

Thank you! I really liked your presentation. Very informative and something I have not seen other industry executives talk about.  Only Buffett had mentioned in the past that the cycle of hard and soft markets might not repeat as regularly as it had in the past but the rest of the industry seems to be always seeing a hard market just around the corner.

 

Vinod

[lookingstill]

Tommm50, I join other board members in thanking you very much for being so generous with sharing your knowledge and expertise. Would you mind my asking what is your opinion of Lancashire? I know that company has its own thread, but I'm concerned that you may never visit it.

[giofranchi]

Tommm50, I join other board members in thanking you very much for being so generous with sharing your knowledge and expertise. Would you mind my asking what is your opinion of Lancashire? I know that company has its own thread, but I'm concerned that you may never visit it.

 

Hey! I was thinking about exactly the same thing! You have just posted this question some minutes before I could do it!  ;D

It is very clear, from what he has written and from the presentation he has attached to one post of his, that Tommm50’s knowledge about the insurance and reinsurance industry is deep (by the way, thank you for answering to my question!) and even a match for twacowfca’s! It would be a real pleasure, and a true learning experience, to spark a conversation among them about Lancashire!  ;)

 

Cheers!

 

giofranchi

 

[LakesideB]

 

1. I didn't see much analysis of the Zenith expense ratios in the annual report or their loss ratios for that matter. On page 11 of the annual report they note Zenith's combined ratios as 136.4 in 2010, 127.5 in 2010, and 115.6 in 2012. I managed uw operations writing WC for the Hartford for about 20 years. We always sought to hit 60% loss ratio or better. We didn't always manage it. The annual report describes Zenith as the best in the business but doesn't provide any backup for the contention. Let's say they think 70% is a good loss ratio. That means they're dragging 63.4%, 57.5%, and 45.6% as expense ratios over the past three years. That's outrageous on a book of $600 million GWP. If you look at the notes on page 134 they break out the expense components of Fairfax Asia. I didn't see anything comparable for Zenith (I'd be happy if someone pointed it out to me).

 

Hi Tom, Thats a phenominal set of slides and notes you have attached. Many thanks.

 

As pertaining to loss ratios for Zenith. Here are the details I found from the annual report (pg 130)

 

                                        2012          2011

Zenith loss ratio              77.9%        78%

Commissions                    9.8%          10.1%

Und. Expense                  28.2%        34.5%

 

The loss ratios don't look that impressive to me. Especially when you contrast it with C&F at 73% and 72% as of 2012 and 2011 respectively. Perhaps that it not the right comparison as the former is doing workers comp and the latter commercial line insurance! It could be that he meant that among the group of worker comp insurers, Zenith has the best loss ratios, which may make sense as workers comp pricing is much harder to get right.

 

 

Also the OP of this thread wants to know how to get the accident yr numbers for different subs. These are detailed quiet lucidly in the annual report for each sub. For instance, for northbridge accident yr loss triangle is on page 157. You can see from the loss triangle the development based on accident year which should reconsile with average numbers Prem gives out in his letter.

 

Thanks

[Tommm50]

Thank you Lakeside.

 

If that's the best loss ratio in the business you have to ask yourself why are we in it?

[Grenville]

Haven't had too much time to follow the discussion here, but I checked the 2009 Zenith National AR and it has details on historical loss ratios that might be helpful. There is detail in Stanley Zax's letter on the loss ratio.

 

Morningstar has the 2009 AR archived. The link is below.

http://quicktake.morningstar.com/stocknet/secdocuments.aspx?symbol=znt

[ap1234]

Since Fairfax has owned Zenith, the loss ratio has been flat (80.2% - 2010, 78% - 2011, 77.9% - 2012). The expense ratio (excluding commissions) was 37% in 2010, 35% in 2011 and 28% in 2012.

 

I looked at the 2009 Zenith annual report. The net premiums earned in 2005 were $1.1 billion. in 2011, the net premiums earned were $500 million. This should explain why the expense ratio is so elevated as premiums shrunk by 50% and expenses did not shrink proportionally.

 

In terms of comparable loss ratios, I'm not sure it makes sense to compare the loss ratio of Zenith and Crum as they are very different business lines. Overall, the workers comp business does not appear to be a good business (although I imagine there are benefits given the float generated from very long-tailed business). However, Zenith has one of the best long-term track records amongst the peer group and have been able to generate a 30 year average CR of 95% despite the difficult industry conditions.

 

When Zenith says they have the best loss ratio in the industry, they are referring to the workers comp industry peers (specifically California workers comp). In the 2009 annual report, there is a table comparing the loss ratios of Zenith vs. the industry average dating back to 1978. Zenith has had a lower loss ratio vs. the industry average every single year.

 

At the recent AGM, Faifax noted that prices are increasing at Zenith (they are price leaders). When the market turns, they do not expect to generate a mid 90s combined ratio. They expect to generate a much lower CR. If their 30 year average CR is 95%, they need much lower results given the past few years elevated underwriting losses to return to their long-term average.

 

Does anyone follow the trends in workers comp? It's interesting to look at the loss ratios before Fairfax acquired Zenith (see below). Since Fairfax has owned Zenith, loss ratios have been at 80%. I assume this is typical for the entire industry but would be interested to hear people's thoughts?

 

2002: 55%

2003: 37%

2004: 27%

2005: 26%

2006: 32%

2007: 38%

2008: 43%

2009: 50%

 

 

 

 

 

 

 

 

[ap1234]

Lakside B said:

 

Also the OP of this thread wants to know how to get the accident yr numbers for different subs. These are detailed quiet lucidly in the annual report for each sub. For instance, for northbridge accident yr loss triangle is on page 157. You can see from the loss triangle the development based on accident year which should reconsile with average numbers Prem gives out in his letter.

 

Lakeside, thanks for your comment. Can you please elaborate? I have looked at the reserve development triangles on an accident year basis but I'm not sure how to get the data for the accident year CRs. For example, where would I find the 2004 accident year combined ratio for Northbridge?

 

Thanks!

[racemize]

Lakside B said:

 

Also the OP of this thread wants to know how to get the accident yr numbers for different subs. These are detailed quiet lucidly in the annual report for each sub. For instance, for northbridge accident yr loss triangle is on page 157. You can see from the loss triangle the development based on accident year which should reconsile with average numbers Prem gives out in his letter.

 

Lakeside, thanks for your comment. Can you please elaborate? I have looked at the reserve development triangles on an accident year basis but I'm not sure how to get the data for the accident year CRs. For example, where would I find the 2004 accident year combined ratio for Northbridge?

 

Thanks!

 

Page 157 shows the reserve development for the accident years--as you can see for 2004, they had reserve redundancies of 24%, which is really good.  Now, as to how to convert that to an actual combined ratio, I'm not too sure, but it would seem clear that a redundancy of 24% should yield a fairly good CR.  Perhaps someone on the board can enlighten us on how to do the conversion or how possible that is (seems like we would need to know the expenses and total policies written to do it).

 

Incidentally, I have a spreadsheet which allows one to convert cumulative triangles into accident year triangles, if you would be interested in it.

[LakesideB]

Appologies OP. I misunderstook and realize now that you want to be able to derive full accident year combined ratios for the subs. I wasn't careful enough to read your question carefully and thought you wanted to know the accident year development.

 

However as racemize suggested, perhaps there is a way to get a reasonable approximation of what the accident year ratios would be. For instance, from the 2003 annual reports one could get the loss ratio, the cat loss ratio, calendar yr reserve development, commissions and expenses. Now loss ratio numbers would already have cat loss ratio and calendar yr reserve development embedded in it.

 

If the cat loss number is not provided for that year in the AR, Prem has mentioned that typical cat loss ratio is roughly 6 combined points. The loss triangles will give the 2003 calendar year reserve development which you will subtract from loss ratio and apply the 2003 accident year reserve development instead.

 

Add in the commish and expenses and you should be able to get a good enough estimate of what the accident yr combined would have been. Its a bit of work to do this for all the years as you would have to open individual annual reports to get the breakdown of combined ratios etc.

 

racemize : would loveto get hold of your spreadsheet. Could you kindly attach it in your msg??

 

Also OP thanks for the info on Zenith. I have always wondered by Prem has always thought so highly of Stanley Zax .... especially when I see the combined ratios numbers since fairfax acquired them. I can finally realize what a terrific job they have done over the last three decades.

[racemize]

Here is a spreadsheet for AIG's reserve triangles, just change the numbers/years in the initial sheet and everything else automatically updates:

 

http://dl.dropbox.com/u/14968/reserve%20triangle%20analysis.xls

[LakesideB]

Thank you racemize.