Underwriting

[Crip1]

Guys,

 

Let's start with a premise or two. I look at investing as "hiring" the management of a given firm to manage the business for the benefit of the owners...a.k.a. me. This in effect makes the management of my holdings my employees. And just like having employees, one looks to evaluate their performance. If the performance of an employee is not up to par, one fires said employee. If the performance of a stock holding is not up to par, one "fires" the management by selling said stock holding. Make sense so far? Along that same vein, when evaluating an employee, one seeks to give an objective evaluation citing the good and the bad of aspects of each employee's performance. Now, obviously, I am not going to send Prem a note with my performance evaluation, as he would rightfully toss it in the garbage if it ever got to him. But, as we are discussing FFH in this forum, I think it vital that we be as objective as possible in our assessment.

 

Now, much has been discussed about Prem and co's character and integrity which, by all objective accounts, is top notch. Much has also been discussed in regards to Prem and co's investing performance which, by all objective accounts has been top notch for not only the past couple years (CDS positions, treasuries sold and  munis purchased, S&P Puts, etc), but for the past several years. I do not know of any manager of money anywhere who has outformed Prem and co in the past 10 years....no one, and for that they deserve the accolades which they have received, and then some.

 

Now, turning to underwriting, I feel that this is an area in which FFH needs to improve. I would also bet that this is an an area where Prem and co feel they need to improve as well. Much has been written regarding the Q4 2008 results and adjusting those results so they they look more in line with the Markels, BRKs and Chubbs out there. But to be honest, I do not want to hear any more of "Well, if you adjust for this that and the other thing, the combined ratio is...". FFH needs to take their underwriting performance to the next level, they need to strive to compete (from a performance perspective, not a market perspective) with the MKLs, BRKs and CBs of the world. Through the previously referenced character and investing acumen, FFH has positioned themselves beautifully for this. They have had their ratings upgraded and now have substantial capital with which to write business in what is expected to be a hardening market. An anology is like a team on a power play where a forward takes a cross-ice pass and has a wide open goal...FFH cannot fan on the shot or shank it wide...they need to bury it.

 

Let's not deny this. Of course they past two years were remarkable in terms of perforamnce, no denying this. But looking out 5-10 years, if they can invest half as well as they have but underwrite as well as the MKLs, BRKs and CBs of the world, then FFH will become an absolute powerhouse of a company. That and nothing short of that is the goal.

 

Lest anyone think that I am unfairly bashing Prem and Co, that is simply not true. FFH is my largest holding. I have not "fired" a single share of FFH and may end up "hiring" some more shares if the price dips a heckuva lot more, so I am not down on the company. Underwriting is the one area where they need to perform at a higher level, period.

 

Please, feel free to post any and all thoughts to the contrary.

 

-Crip

[Redskin212]

Crip,

 

I could not agree more.  Without question Prem and co's character and integrity is second to none and their investment record speaks for itself.  One of my pet peeves has been "the combined ratio "adjusted for" hurricane Ike, wind storms, currency rates...."  I just hate it.  I can see that it may help the reader understand the CR from year to year - but I just blow by it as I think excuse excuse.

 

Now, I may take this one step further than you regarding management (especially at the operating company level).  As you know, there is virtually no senior management turnover at Fairfax and its subs, which it thougth to be a very positive attribute of the company.  Is it?  If you are in senior management at one of the insurance companies; NB, ORH, C&F you are ONLY responsible for insurance.  Hamblim Wasta takes care of the investments, which is bascially the left hand side of the balance sheet.  I am guessing that the shareholder's equity (ie capital raising, share buyback) and debt are directed by Prem and the head office team as well.  So all you have to worry about is underwriting insurance and claim management.  I am not saying that this is dead easy, but compared to your peer group,  you have a much narrower focus and therefore should be above average and not mediorce.  If I were doing a performance evaluation of the operating management, I think I would set the bar a little higher than breakeven.  The operating subs really have no excuses why they are not writing at 95.

 

Disclosure - FFH is definitely my largest holding and has been for a while.  ORH is second.

 

Redskin

 

 

[valuecfa]

Just for reference and constructive criticism I thought some numbers might be helpful. In last years letter to shareholders one can see the historical, consolidated combined ratio figures for Fairfax Financial:

 

http://fairfax.ca/Assets/Downloads/080307ceo.pdf

 

 

It appears, from 1986 - 2007 the weighted average cost of float has been (2.5%)

 

~meaning the float has cost shareholders 2.5%. So while this is not a great number by any means, it is not terrible either. Something they can definitely improve on.

 

"Our long term goal is to increase the float at no cost to our shareholders. This, combined with our ability to invest the float well over the long term, is why we could achieve our objective of a 15% per annum compounding of book value per share over time."

[oldye]

I'm not really sure how exactly you want them to "improve" underwriting.  I mean you always want some things to improve but unless they are fundamentally doing something wrong with their underwriting there really isn't much to improve.

 

They write business in certain lines and the market price for those lines has been deteriorating badly...the only response they can have is to write less business...they can't stop hurricanes or change the way they do accounting so that the CR actually reflects the fact that they hedge against currency changes and that there is no net impact. 

 

You have to be patient, when the market hardens, the fixed costs as % of business will decrease and you'll be happy with the results. 

 

I mean common Markel CR was 99, Fairfax was at what 103-104 excluding the currency stuff...how many companies out there have a cost of capital of 4%-5% during the 3rd worst hurricane season on record?

 

 

[Guest ericopoly]

The operating subs really have no excuses why they are not writing at 95.

 

if they can invest half as well as they have but underwrite as well as the MKLs

 

MKL wrote a 99 in 2008.  Do they have any excuse for not writing a 95?  Yes they do: Ike was the 3rd costliest hurricane EVER!

 

 

If you are in senior management at one of the insurance companies; NB, ORH, C&F you are ONLY responsible for insurance. 

 

This is why folks need to fix up the C/R to remove the currency impact, and also to back out the $80m commutation charge on the adverse development cover (they were neither responsible for the reserving, nor it is their fault that Prem wanted to commute it early to get the money to invest for higher returns -- the protection was all used up anyhow).

 

 

It is exactly because we want to measure their performance that we back these things out -- else we don't get useful information on how they are really performing!

 

Fine, be upset that they are making "excuses" -- but how else are you going to understand how they will do in a normal year, and no, 3rd largest hurricane is not a normal year.

[Smazz]

My opinion is,

if Ratings agencies did their job properly, FFH and BRKs would be able to charge alot more for their services - for numerous reasons.

 

Like all things, its not usually a question of IF but WHEN.

 

[Guest misterstockwell]

Totally agree.

 

Allstate routinely underwrites in the 90's WITH catastrophes INCLUDED. When they exclude them, it's low 80's. HCC does low to mid 80's all the time. Why pay for float? Why write insurance at a loss? You might as well make another few hundred million on it. Let's see.....they did ~$4 billion in net premiums, and Prem says they can double that in a hard market. So let's use $8 billion, and let's make a 5% underwriting profit. That's $400 million. That's a gross profit of almost $23/share. That's significant!

[Guest ericopoly]

Allstate routinely underwrites in the 90's WITH catastrophes INCLUDED. When they exclude them, it's low 80's.

 

 

Allstate wrote 99.4 in 2008

Allstate wrote 89.8 in 2007

 

Fairfax wrote 103.4 in 2008 (104.1 including the asbestos settlement)

Fairfax wrote 94 in 2007

 

So Allstate was about 4 points better in both years.

 

Question:

Isn't Allstate heavily into the car insurance business?  I want to make sure that you are comparing similar lines of business -- my perception of automobile insurance is that it is very short tail and therefore the CR will be lower out of necessity.  Is this not accurate?

 

 

[indirect]

Guys,

 

My hypothesis : FFH writes long tail and intermediate insurance and 5 yrs ago they had to write more to survive and may be at that time they were not as discriminating. We are paying for it now?

 

AS a corollary will the same thing happen now when co like CNA, AIG etc need premium growth?

[arbitragr]

Just so long as the cost of float doesn't exceed the cost of debt for most years I think it's okay.

 

Suprising how bad it (underwriting loss) was this year however.

 

Nonetheless, FFH's investments were good enough to overshadow the "ordinary" underwriting results for this year at least.

[Grenville]

"The underlying business, our insurance and reinsurance business, excluding these exceptional items and in a very challenging insurance environment, had a combined ratio of 96.2%. Accident year combined ratios for all our companies are close to 100% in all cases with good reserving."

- From the conference call.

 

I hope everyone has listened to the conference call. They address the fx issue, and the underwriting pretty well. Based on my limited understanding of underwriting, they did a pretty good job overall. I agree if you were trying to go for perfect and nit pick, the underwriting is the worst performing area relatively. I think if you zoom out they are doing a fine job with their underwriting and they would tell you up front if they weren't.

 

One question I have, does the CR ratio take into account how they are reserving? If they are super conservative and over reserving will that mean that the CR ratio looks artificially high or does the effect of over reserving average itself out over time?

 

 

 

[Rabbitisrich]

It's not entirely fair to look at FFH as two components, the investment house and the insurance business. The uncertain cash flows of the insurance side bear upon the investment side, and perhaps vice versa.

 

For example, on the conference call Prem stated, "As we generally hedge our foreign exchange claim (ph) exposures, these translation losses are almost fully matched by translation gains on our investment portfolio." The hedges don't appear on the combined ratio, but as investors we should back out the translation loss for the amount of the currency swap gains (if possible).

[SharperDingaan]

 

We're really talking about the table on p23 of the 2008 Full Year Interim Report, & how poorly we understand it.

 

We see a 2008 UW loss of (457.7) & think that's terrible, when we should really be looking at the operating income which was +18.4

- They write longer tail business & they don't discount, so the UW loss is higher that a competitor would report. Example: Assume 100 of total expense at the end of the year distributed as 70 in Yr1, 20 in Yr2, & 10 in Yr + total premium income at the end of year 1 of 97.5. UW cost is -2.5 (97.5-100). But discount at 10% & the UW cost becomes +1.0 (88.6-87.6). Would we rather have the 'pretty' version, or the 'real' numbers ?

- UW is their business - & they write it obtain float at a reasonable cost. Investment is our business - but we couldn't predict the collapse of Lehmans last year, & the impact of the auto-industry restructuring this year. Yet somehow when the equivalent events happen in their business we're expecting them to have either priced perfectly or exited the market at the right time. IE: We can have extraordinary investment losses, but underwriters cant ?

- You hedge investment results via index options, CDS's, etc; you hedge UW with excess of loss coverage, reinsurance, & exiting markets. But we dont understand the UW hedges so we give them no benefit ?

- Long tail CR is the result of business decisions made over the last 1-3 years. 2008 Consolidated CR was 110.1, 2007 was 94.0, & like it or not investment & UW are intimately related. 2008 was a very good year, but why ? They exited UW markets that weren't profitable, they did not 'reach' for cash yield in interest & dividends, they increased runoff by commution, & they sold off pre-established market positions. We were expecting them to write tons of junk business to inflate premium & make the UW loss minimal ? & then load up on junk bonds to maximize interest income & inflate operating income ?

 

Reality is that you have to adjust numbers to get a better idea of what 'normal' looked like, & compare against peers writing the same type of business in similar quantities. I don't want to do it, is not an option.

- Most folks would look at the interest & dividend line, normalize the average yield over some period, & then recalculate based on the normalized yield. A UW/Investment adjustment

- You would do a similar thing with run-off, but would assess against market conditions at those times when run-off changed significantly. A UW adjustment & a UW quality assessment.

- A similar thing with lagged UW loss (net of Katrina's, Ike's, etc) vs premium. Another UW adjustment/quality assessment     

 

Make those adjustments & you'll find that the UW is actually very good. If it wasn't good wouldn't HW have let the team go ?  Given that he hasn't done so isn't that telling you something ? 

 

Folks its easy to criticize, but from what I see we'd be useless at UW.

Let them get on with running the business. If you have an objection, sell your shares.

 

SD 

 

 

[Guest misterstockwell]

::)

 

Ok, how about Chubb? Is that a peer? Is there some reason we should exclude its results?

 

08  88.7

07  82.9

06  84.2

05  92.3

 

This includes all catastrophe costs.

 

Long tail or short tail, if you underwrite correctly, you should be profitable, whether it was business written last year or 5 years ago. FFH is extraordinary at investing, and barely ordinary at underwriting. This is my last comment on the subject, and please don't respond with the "...then sell your shares!" I own part of this business, and I have the right to look critically at what happens there. Open minded investors should all do the same.

[woodstove]

Great discussion!  I know zip about underwriting.  Paying something for float (2.5 pct) seems decent given probable investment returns.  And there is utility in adjusting for fluctuations, to see whether normal operations (under unit managers control) are prudent, well run etc.

 

What would be the impact of say a 1 pct increase in premiums, for same amount of risk?  Would cause some loss of business, ie float.  Is it more investment income loss than premium gain?  That seems to me the sort of tradeoff which might be considered at high level.  Maybe given an excess of cash at holding company, it would be a reasonable tradeoff - ie, look not at overall rate of investment return, but the marginal rate of return.

 

What would happen if Fairfax priced for a really hard market, say 20 pct increase in premiums for same amount of risk coverage?  Capital is scarce, AIG and KFS are impaired, cat bonds and alternatives to normal insurers are probably not considered prudent insurance purchases and might cost someone his job.  Why should Fairfax be generous with making its capital available to insurees given 15 pct returns in other sorts of investments?  It would be an interesting marketing test.  Probably is considered routinely by unit managers via normal operations.

 

Not second guessing.  I trust the guys doing the work.  Just interested in understanding how the various parts fit together and move.

[Partner24]

Dear Crip,

 

You bring an interesting point and, to me, your comments are fair. There is different aspects to check before making a judgement about the managers performance.

 

- In taking a look at how they have been able to improve the balance sheet over the last few years, I'm quite happy about the results. We're not a AAA company, but Fairfax has done significant steps in the right direct direction and I'm very satisfied.

 

- Investment returns: they are great. We were protected from a 50 to 100 years financial storm and we were not swimming naked. I don't know any public company that had these wise move made and this is simply terrific.

 

- Runoff. Great improvement! Less claims outstanding, less underwriting losses, etc. This really seem to be well contained now. A great relief!

 

- Underwriting. Yes, this is the sour point to me, but this is not a catastrophe. Ideally, it would be great to have no cost float over the mid and long terms (or ice on cake, an underwriting profit), but it is not terrible neither. I understand that I shouldn't include the currencies variations, but I will fully include underwriting losses from hurricanes and other catastrophes.

 

So, to me, Fairfax performance over the last few years can be quoted as A. The small percentage cost of float is the only reason why I don't see myself honestly give an A+...but frankly A+ is not far.

 

As an individual shareholder, am I satisfied with an A? Absolutely! There is so much D and E out there, especially in 2008. Do I see myself "firing" Fairfax by selling my shares because of insatisfaction? No.

 

Cheers!

 

 

 

 

 

[FlyingArrow]

Dear Crip,

 

You bring an interesting point and, to me, your comments are fair. There is different aspects to check before making a judgement about the managers performance.

 

- In taking a look at how they have been able to improve the balance sheet over the last few years, I'm quite happy about the results. We're not a AAA company, but Fairfax has done significant steps in the right direct direction and I'm very satisfied.

 

- Investment returns: they are great. We were protected from a 50 to 100 years financial storm and we were not swimming naked. I don't know any public company that had these wise move made and this is simply terrific.

 

- Runoff. Great improvement! Less claims outstanding, less underwriting losses, etc. This really seem to be well contained now. A great relief!

 

- Underwriting. Yes, this is the sour point to me, but this is not a catastrophe. Ideally, it would be great to have no cost float over the mid and long terms (or ice on cake, an underwriting profit), but it is not terrible neither. I understand that I shouldn't include the currencies variations, but I will fully include underwriting losses from hurricanes and other catastrophes.

 

So, to me, Fairfax performance over the last few years can be quoted as A. The small percentage cost of float is the only reason why I don't see myself honestly give an A+...but frankly A+ is not far.

 

As an individual shareholder, am I satisfied with an A? Absolutely! There is so much D and E out there, especially in 2008. Do I see myself "firing" Fairfax by selling my shares because of insatisfaction? No.

 

Cheers!

 

 

Very well said Partner. I have nothing to add to that except to reinforce your point on runoff. What a difference a couple of years makes.

[Viking]

I am hoping:

1.) with their recent ratings upgrade this will help them attract business at better rates

2.) other companies (to survive) are being less conservative with their reported underwriting numbers (which we will not know for years... i.e. look at all the stuff that came out of the closet in 2000-2005).

 

I would like to see FFH improve with their underwriting results. That would make this stock a core, long term hold in my mind.

[Guest misterstockwell]

To decide whether an underwriting profit is important or not, one need only go to the Berkshire chariman's letter and read---note how many paragraphs are written about underwriting profits and their importance. There is absolutely no reason for FFH to have to pay for float!

[dowfin1]

Also unlike FFH and MKL, WEB doesn't use CAT losses as the isolated exception.  Indeed it looks like BRK suffered little CAT losses when most others did. 

[oldye]

Why not focus on the part where insurers only earned about 8% on equity over the last 30 years and that most don't write insurance at a profit. Btw this is by design, it is the explicit goal of regulators to keep CR's around 100.

 

 

 You can huff and puff and wish all you want but until prices firm there isn't much they can do other than write less business.  Berkshire's insurance companies are not only AAA, one of them has an enormous marketing budget with a low cost structure.  Ajit Jain generates about 40% of the float with a group of 31 employees.  Lowest cost of capital+lowest cost producer = good underwriting.  It can't be replicated!

 

As an aside,

Of the 16 billion in Muni's Berkshire insurers, Fairfax owns about 3.5 billion, and I'm pretty sure Hamblin Watsa belong to the group of investors he referred to as "sophisticated."