Fairfax 2022

[glider3834]

2 hours ago, nwoodman said:

I haven’t see it reported elsewhere that IRDAI had knocked back the conversion of the Digit Prefs.  

yes - from prospectus looks like Fairfax are continuing to engage with IRDAI on this issue - so we will see what happens.

 

I think IRDAI concern is if there is conversion, Go Digit Infoworks becomes a subsidiary of Fairfax Asia (FAL). Go Digit Infoworks is a promoter of Digit Insurance, and an 'Indian promoter' under IRDAI regs can be a company but not a subsidiary.

 

Having said that, it appears an 'Indian promoter' can be an LLP. This 'subsidiary' issue came up with Poonwalla controlled PFL, the promoter of Magma HDI General Insurance. The controlling shareholder of PFL got around this IRDAI reg by selling shares  in the underlying insurer, Magma, to an LLP controlled by Poonwallas.

https://www.business-standard.com/content/press-releases-ani/proposed-divestment-of-stake-in-magma-hdi-general-insurance-121110300487_1.html

 

I am not sure if Fairfax could use this or another corp reorg strategy?

 

By my estimates, these convertible preferreds could be converted to a 37% equity interest in Go Digit Infoworks or an indirect 32% economic interest in Digit - so if Fairfax doesn't convert them, this is an asset that can be sold to a third party who could & possibly this could be another option.

 

Also I was thinking whether it would be possible for Go Digit Infoworks to use the listed price (fair value) of Digit to sell its Digit shares to FAL. FAL could in turn sell its convertible preferred shares back to Go Digit - so  FAL would then have its Go Digit Infoworks Stake plus direct ownership in Digit Insurance. So maybe this could be a way for Fairfax to become controlling shareholder of Digit Insurance??

 

Go Digit Infoworks are selling up to 109 mil shares in this IPO ,  it will be interesting to see how these proceeds are used. 

 

 

 

 

 

 

 

 

 

Edited August 28, 2022 by glider3834

[nwoodman]

@glider3834  Thanks for the heads up.  I hadn’t read through the prospectus in detail until now, from p.51. regarding the compulsorily convertible preference shares (CCPS)

 

The promoter of FAL Corp is Fairfax Financial Holdings Limited, which is listed on the Toronto Stock Exchange, the majority designated partner of Oben (holding 99.99%) is Kamesh Goyal, and the promoters of GDISPL are Kamesh Goyal, Oben and FAL Corp, which respectively hold 14.96%, 39,79% and 45.25% of the equity share capital of GDISPL. For details of the shareholding pattern of GDISPL, see “Our Promoters and Promoter Group” on page 261.
Additionally, FAL Corp holds 7,800,000 CCPS issued by GDISPL (aggregating to 100% of the preference share capital of GDISPL). The aforesaid CCPS has a fixed conversion ratio for conversion into equity shares of GDISPL being (i) 2.324 CCPS for each equity share, for 6,300,000 CCPS (”Ratio 1”) ; and (ii) 3.55 CCPS for each equity share for the remaining 1,500,000 CCPS (“Ratio 2”). Upon conversion of the CCPS, the parties have agreed that the shareholding of FAL Corp in GDISPL will represent maximum of up to 82.07% of the share capital of GDISPL. Further, consequent to conversion of the CCPS, the indirect shareholding of FAL Corp in our Company (on a fully diluted basis ) will be a maximum of up to 68.65%. While we believe that upon the CCPS conversion, none of our Promoters shall cease to act as promoters of our Company, we cannot assure you that the regulators will not take an adverse view, in which case such an event may have an adverse effect on our Company or its shareholders.
On June 7, 2022, our Company applied to the IRDAI, seeking its approval for conversion of the 7,800,000 CCPS into equity shares of GDISPL. However, the IRDAI, by way of its letter dated July 26, 2022, communicated that this application cannot be considered by it, since the proposed conversion of the CCPS would result in GDISPL becoming a subsidiary of FAL Corp which is not allowed under the IRDAI (Registration of Indian Insurance Companies) Regulations, 2000, which defines an ‘Indian promoter’ to mean a company, as defined in the Companies Act, which is not a subsidiary, as defined in Section 2(87) of the Companies Act. For further details in relation to the above, see “Our Promoters and Promoter Group” on page 261.
While upon the CCPS conversion, none of our Promoters shall cease to act as promoters of our Company, and our Company and our Promoters intend to continue to engage with the IRDAI in relation to such conversion of CCPS, as per the provisions of applicable law, we cannot assure you that the IRDAI will approve such conversion in the future. Consequently, we cannot assure you that the CCPS will be converted by FAL Corp in a timely manner, or at all. Further, each of FAL Corp and, subject to FAL Corp’s consent and right of first refusal, Kamesh Goyal and Oben has the ability, should they choose to do so, to sell their respective shareholding in GDISPL to a third party, which, if sufficient in size, could result in a change of control of our Company. 

 

The level of disclosure in the prospectus is refreshing.  The scrutiny by the IRDAI is frustrating but also creates quite the moat.

Edited August 28, 2022 by nwoodman

[Viking]

When i monitor Fairfax i focus on three broad buckets. Most importantly, what is the near term trend with each bucket:

1.) underwriting income

2.) interest and dividend income

3.) realized/unrealized investment gains

 

Powell on Friday suggested Fed funds would be moving higher in 2022/2023 likely to something in the 4% range. He also said rates would stay higher for longer (perhaps all of 2023). (Not all of Fairfax’s fixed income investments are in the US.)
 

Higher interest rates for longer in the US is very good news for Fairfax. This will also lead to higher rates in Canada (given the economic linkages). Europe is likely a different story (interest rates are likely to remain anemic).

 

Fairfax has a cash/bond portfolio of about $35/$36 billion. Most of it is domiciled in US/CAN. Let’s assume a 3.5% average interest rate is a reasonable estimate looking 1 year out = $1.2 billion in interest income/year = $300 million/quarter. For reference, Fairfax earned $176 million in interest income in Q2, 2022. 

 

Fairfax also has an average duration of 1.2 years on its bond portfolio (most insurers are about 4 years). This means about 21% of Fairfax’s total bond portfolio will reset to higher rates each quarter. Each quarter we should see interest income continue to move higher. 
 

Fairfax earned $176 million in interest income in Q2, 2022 = 2% average yield. Up from $154 million in Q1 = 1.8%. In Q3 interest income could be over $200 million = 2.3% yield.
 

Will Fairfax get to $300 million in interest income/quarter? It is possible.
 

Another catalyst will be if Fairfax is given the opportunity to shift into securities with higher yields (if we get a significant widening in spreads). And increasing duration could also result in higher yields (should we see yields move higher in 3-5 year duration). So an average yield of 3.5% on the bond portfolio might end up being a low estimate in another 12 or 18 months.

—————

This does not include dividends = $43 million in Q2. Or investment expenses = $17 million in Q2

—————

Interest & dividends = interest income + dividends - investment expenses.

Edited August 28, 2022 by Viking

[TwoCitiesCapital]

5-year treasuries have a yield of 3.25%. Would love to see this locked in on SOME portion of the portfolio. 

 

I simply fear a repeat of 2018 - short duration on the way up (missing interest income but avoiding capital losses) and short duration on the way down (missing interest income and capital gains). 

 

A 5-year type ladder wouldn't be taking excessive duration risk and could largely lock in a knowable cash flow of ~750 million - $1 billion per year on the fixed income (pending rate moves). They'd not be taking unnecessary duration risk, they'd solidify a knowable income stream, and retain the optionality of extending duration and/or benefitting from rising rates by having a lower duration portfolio. 

 

Really hoping to see the bond portfolio extended out from 1-2 year bonds here in the near future. 

 

[Cigarbutt]

To value FFH:

1-underwriting income

2-capital allocation and investment float management

3-downside risks

Obviously 2- and 3- are the fun parts. Here are some thoughts on 1-.

TL;DR version: In the next two to five years with an expected baseline accident year 95% combined ratio, expect +/- 2.5% adverse reserve development instead of a +/- 2.5% favorable development on reserves which (based on more or less 1:1 premium equity ratio) results in a 5% difference in ROE from the underwriting side.

-----

This has been peripherally discussed before here and the change that is (slowly) developing will impact results going forward in the medium term. The last leg of the underwriting cycle has been very very unusual in the sense that 1-the duration of the last soft part has been very long and 2-the hardening actually occurred before underwriting capacity was badly hurt. For Fairfax, this means that one should expect lower underwriting profitability although FFH's book of business has grown more lately during the hardening which should produce stronger accident year results going forward to mitigate previous years' development.

One can look at industry wide data but, above in thread, the excess and surplus market is mentioned and it is a nice example (as a relevant sample versus commercial lines in general including specialty lines; workers comp's dynamics playing out differently) of this cycle turning in a very unusual way. 

Reserve releases are coming down and, overall, adverse development is expected in the next few years until the later years have "run-off" from the long-tail portfolios. 2020 was a bizarre year with an unusual and focal favorable development for 2020 (2021 looks similar) masking previous prior years' adverse development but it is reasonable to assume that the 2020 year was an anomaly (tribal comments omitted from this post).

In a relevant way, the following is interesting:

Excess & Surplus Lines Market Hardens Further | Insurance Thought Leadership

In a relevant way, here are some numbers from FFH:

The prior years' development has been favorable but the trend is clear, fits with the E+S market, specialty market and commercial lines in general, even more clearly when "run-off" reserves' development is taken into account.

Not the end of the world*, but the above likely means lower expectations for underwriting return on capital for a few years.

*The above does not take into account catastrophes etc. Today, there's an article describing how the month of August has been the quietest month for hurricanes since 1960 and it was suggested that climate change may explain that.

[longlake95]

I recently met with a friend who has worked in insurance her whole career, she is 53-ish…she is a senior underwriter in commercial lines, here in Canada.

 

I asked her what inning of the ball game are we in the “hard market”, without missing a beat she replied “ the 7th inning”. 
 

yes, it’s anecdotal, but it’s real-time, from a source I trust.

 

 

[petec]

On 8/31/2022 at 1:25 PM, Cigarbutt said:

To value FFH:

1-underwriting income

2-capital allocation and investment float management

3-downside risks

Obviously 2- and 3- are the fun parts. Here are some thoughts on 1-.

TL;DR version: In the next two to five years with an expected baseline accident year 95% combined ratio, expect +/- 2.5% adverse reserve development instead of a +/- 2.5% favorable development on reserves which (based on more or less 1:1 premium equity ratio) results in a 5% difference in ROE from the underwriting side.

-----

This has been peripherally discussed before here and the change that is (slowly) developing will impact results going forward in the medium term. The last leg of the underwriting cycle has been very very unusual in the sense that 1-the duration of the last soft part has been very long and 2-the hardening actually occurred before underwriting capacity was badly hurt. For Fairfax, this means that one should expect lower underwriting profitability although FFH's book of business has grown more lately during the hardening which should produce stronger accident year results going forward to mitigate previous years' development.

One can look at industry wide data but, above in thread, the excess and surplus market is mentioned and it is a nice example (as a relevant sample versus commercial lines in general including specialty lines; workers comp's dynamics playing out differently) of this cycle turning in a very unusual way. 

Reserve releases are coming down and, overall, adverse development is expected in the next few years until the later years have "run-off" from the long-tail portfolios. 2020 was a bizarre year with an unusual and focal favorable development for 2020 (2021 looks similar) masking previous prior years' adverse development but it is reasonable to assume that the 2020 year was an anomaly (tribal comments omitted from this post).

In a relevant way, the following is interesting:

Excess & Surplus Lines Market Hardens Further | Insurance Thought Leadership

In a relevant way, here are some numbers from FFH:

The prior years' development has been favorable but the trend is clear, fits with the E+S market, specialty market and commercial lines in general, even more clearly when "run-off" reserves' development is taken into account.

Not the end of the world*, but the above likely means lower expectations for underwriting return on capital for a few years.

*The above does not take into account catastrophes etc. Today, there's an article describing how the month of August has been the quietest month for hurricanes since 1960 and it was suggested that climate change may explain that.

 

I suspect I know less about this than you do, but I would have assumed

1) That the weakening in favourable development over the last few years is due to the lagged effect of the soft market - i.e., during the soft years it got harder to underwrite well so favourable development slowly reduced, and

2) since we are now in a hard market, which all the commentary I have read is largely driven by price not exposure, ceteris paribus one would expect increasingly favourable development in the next few years.

 

Why is this wrong, if we ignore inflation (which I don't think is what'd driving your prediction, but correct me if I am wrong)?

 

 

[Viking]

The 'holy grail' of investing is finding a company that is:

1) growing total earnings meaningfully (above trend/expectations) each year

2.) decreasing share count meaningfully each year

3.) getting a higher multiple from Mr Market (as investors, over time, move from hate to love).

When one of these happens, investors usually do well. When two of these happen, investors usually do very well. When all three happen at about the same time, investors hit the ball out of the park (#3 can happen with a bit of a lag). 

----------

So what has been happening with share count at Fairfax? From 2008 to 2017 Fairfax increased ‘common stock effectively outstanding’ from 17.5 to 27.8 million = +60%. This was done to fund the expansion of its insurance business: Zenith (2010) - and then internationally - Brit, Eurolife, ICICI Lombard (2015), Indonesia, Eastern Europe, Latin America, South Africa (2016), Allied World (2017). These were the years Fairfax was taking massive losses with its ‘equity hedge’ positions so expansion had to be funded largely with share issuance.

 

The surprising thing is the price Fairfax was able to issue shares at. From 2015-2017 Fairfax issues a whopping 7.2 million shares at an average price of... US$462. Not that far away from where shares are trading today at US$485. And all shares issued from 2008 to 2017 were issued at an average price of US$425. Interesting.

 

What has happened since 2017? The share count has fallen for 4 straight years (2018-2021). Share count has fallen by 3.8 million = 14%. That is a meaningful amount. What price were shares bought back at? 2 million were bought back in 2021 at $500 so this is likely a good average number to use.

 

So Fairfax issues shares at $462 to fund a couple of big acquisitions and then buys back a big chunk of the shares 5 years later at $500. Meanwhile i think it is safe to assume the insurance businesses purchased back in 2015-17 are now worth (in aggregate) at least 2X what was initially paid.

 

What will we see from Fairfax moving forward: more issuance? Or more buybacks? The answer is easy: more buybacks… and perhaps one or more large buybacks, like what happened in 2021.

 

Why so confident Fairfax is done issuing new shares? Because that is what Prem has been telling us for years. In the past new shares were issued to fund Fairfax’s international expansion. Today Fairfax is happy with its global insurance footprint. There will be no more large, transformative acquisitions - just small bolt on acquisitions like Singapore Re in 2021.

 

Prem also had this to say in his 2018 letter to shareholders (written in early 2019): “I mentioned to you last year that we are focused on buying back our shares over the next ten years as and when we get the opportunity to do so at attractive prices. Henry Singleton from Teledyne was our hero as he reduced shares outstanding from approximately 88 million to 12 million over about 15 years. We began that process by buying back 1.1 million shares since we began in the fourth quarter of 2017 up until early 2019 – about half for cancellation and half for various long term incentive plans we have across our company.” The pandemic hit in  early 2020 and this effectively stopped any material buybacks at Fairfax for the next 12 months. But 2021 saw a big buyback and i think we will see another of size over the next year or two. Prem’s reference to Henry Singleton was done for a reason.

 

Bottom line, owners of Fairfax can expect the share count to continue to fall in the coming years - and perhaps by a lot!

----------

Common Stock Effectively Outstanding

	Dec 31	change
2021	23,865,600	-2,310,906
2020	26,176,506	-654,563
2019	26,831,069	-406,878
2018	27,237,947	-513,126
2017	27,751,073	4,657,507
2016	23,093,566	879,707
2015	22,213,859	1,037,691
2014	21,176,168	-23,834
2013	21,200,002	954,591
2012	20,245,411	-130,385
2011	20,375,796	-79,451
2010	20,455,247	466,377
2009	19,988,870	2,502,045
2008	17,486,825

[Gregmal]

What’s nice with Fairfax or Berkshire here, and additionally a number of smaller companies like maybe a Stelco and even smaller like potentially a PCYO, is that they’re prepared for this and have been. If you haven’t been a shareholder longer term ish, the underperformance stemming from preparing for this current environment isn’t your problem. So today you get to step in and get the finished product which looks excellent just the same as those who waited it all out, just without the frustration. With more cash and liquidity than they know what to do with, and a commitment to buybacks….there’s just not many ways where you end up in a spot where you don’t win big. As long as one views these investments as that, and doesn’t hang on the daily quotes, you have to kind up be a happy camper whether the market is up, down, or sideways knowing the value that’s being created.

[maxthetrade]

20 hours ago, Viking said:

Why so confident Fairfax is done issuing new shares? Because that is what Prem has been telling us for years. In the past new shares were issued to fund Fairfax’s international expansion. Today Fairfax is happy with its global insurance footprint. There will be no more large, transformative acquisitions - just small bolt on acquisitions like Singapore Re in 2021.

 

I sure hope you are right! With Fairfax I'm in the show me camp right now.

[SafetyinNumbers]

7 hours ago, maxthetrade said:

 

I sure hope you are right! With Fairfax I'm in the show me camp right now.

 

As Viking pointed out, FFH issues stock when it's at a big premium to book to buy other fairly or overvalued insurance companies. With the stock at a big discount to book and a fraction of float, it has to be a short list of what Prem could find interesting enough to buy with stock.

[maxthetrade]

8 hours ago, SafetyinNumbers said:

With the stock at a big discount to book and a fraction of float, it has to be a short list of what Prem could find interesting enough to buy with stock.

 

I'm more worried about him finding another crap co he can't resist buying instead of doing buybacks.

[Viking]

@maxthetrade i am waiting for:

1.) for the pet insurance sale to close (hopefully Sept Oct)

2.) end of hurricane season

I think the table is set for another large stock buyback in Q4. Prem often telegraphs what Fairfax’s plans are during the Q&A portion of quarterly calls. Below is what he said July 29 on the Q2 call.

—————

Fairfax thought their stock was crazy cheap 12 months ago and happily paid US$500 for 2 million shares. Fairfax is trading today at US$485 and it is a more valuable company 12 months later:

1.) 20% growth in net written premiums

2.) interest and dividend income run rate close to 2X what it was 12 months ago

3.) sale of pet insurance business for $1.4 billion

4.) sale of Resolute for $600 million

5.) large stable of equity holdings continue to build intrinsic value

—————

Tom MacKinnon: Yes, thanks. Good morning, Prem. Just a question with respect to the proceeds that are probably going to be coming in associated with the Pet Insurance deal. It looks to be about $1.2 billion in cash now. You're buybacks, I mean, he did the SIB. But since then they've been still relatively modest. What are your thoughts as to what to do with this $1.2 billion in cash from that Pet Insurance deal?

 

Prem Watsa: So, Tom, we are always flexible, of course, and we look at all the possibilities. We, in terms of acquisitions in the property casualty business, we are not focused on it. Because our business now is running at about $28 billion, Tom, that US dollars, of course, in 2015, it was $8 billion and in 2018 it was $15 billion and now it’s running at for 2022 , $28 billion. And that's not including the GIG, and the Middle Eastern company that we've got, which is another $3.9 billion digit close to $4 billion. So we've got a significant amount of operations decentralized all over the world and running at an underwriting profit, and very good reserving, I may add, so no acquisitions, I mean, small acquisitions here, and then Asia or Latin America, but nothing significant. And so we look at obviously, buying back our stock, that'll be the number one thing that we'd look at, but not at the expense of our financial position. I've said that many times for you, not at the expense of our financial position, but we would look at buying back our stock.

 

Edited September 4, 2022 by Viking

[maxthetrade]

@VikingI pretty much agree with all what you said. I remember that exchange on the call and it was indeed quite encouraging. Problem is that Prem doesn't always do what he said he will do. Remember those high quality stocks (JNJ and WFC if I recall correctly) he said he wanted to hold for the long term and were sold soon after? And let's say that I wasn't exactly thrilled by the Recipe aquisition. Hence my show me attitude.

But all in all I think a big buyback later this year is more likely than not if Q3 is benign on the insurance front.

I think that the TRS also makes a repurchase more likely, they're definitely incentivized.

 

BTW, just looked at your equity holdings spreadsheet in the other thread, thanks for all your work on this name!

Edited September 4, 2022 by maxthetrade

[SafetyinNumbers]

15 minutes ago, maxthetrade said:

@VikingI pretty much agree with all what you said. I remember that exchange on the call and it was indeed quite encouraging. Problem is that Prem doesn't always do what he said he will do. Remember those high quality stocks (JNJ and WFC if I recall correctly) he said he wanted to hold for the long term and were sold soon after? And let's say that I wasn't exactly thrilled by the Recipe aquisition. Hence my show me attitude.

But all in all I think a big buyback later this year is more likely than not if Q3 is benign on the insurance front.

I think that the TRS also makes a repurchase more likely, they're definitely incentivized.

 

BTW, just looked at your equity holdings spreadsheet in the other thread, thanks for all your work on this name!

 

Agree that Viking does great work! 

 

There is a real distrust in Prem and I guess that works in our favour as he buys back stock. Has there been a time when Fairfax has issued stock at a discount to book value to buy something?

[Cigarbutt]

On 9/2/2022 at 11:36 AM, petec said:

 

I suspect I know less about this than you do, but I would have assumed

1) That the weakening in favourable development over the last few years is due to the lagged effect of the soft market - i.e., during the soft years it got harder to underwrite well so favourable development slowly reduced, and

2) since we are now in a hard market, which all the commentary I have read is largely driven by price not exposure, ceteris paribus one would expect increasingly favourable development in the next few years.

Why is this wrong, if we ignore inflation (which I don't think is what'd driving your prediction, but correct me if I am wrong)?

 

 

You are directionally correct and the idea is to guess if the 'excess' reserves being built now (something which we will find out in the future) are enough to compensate for the developing realization (something that we will find out in the nearer future) that previous years were characterized by deficient reserves.

-----

Longer answer

In the reference above, there is this graph which shows what is known now about the past but which, for the last 2 to 5 years, will significantly change within the next 2 to 5 years:

If history is any guide, the years 2016 and after will likely deteriorate and this deterioration will be recognized in future years. 

For fun (historical fun), let's look at what happened to the medical liability market before, as quite a representative example of most commercial lines (although typically more volatile ie more of this guessing about the past and the future). So what was known around 2001:

What happened (recognition of the past) after a couple of years:

The graphs are not quite comparing oranges to oranges but enough to draw some insights. Around 2001, it was reported that reserves were determined to be sufficient up to and including the year 2000. around 2005, it was realized that from 1997 on, policies had been underpriced and reached maximum adverse development in 2000, when things started to look better.

Anyways, usually there is relative pain as a result of this transition (typically negative underwriting income for a while) but this is variable and this time may be different (the market has turned relatively early versus the typical triggers ie large losses, threatened surplus, companies leaving market etc) and FFH has shown to be slightly better than peers (underwriting discipline, higher growth during hardening etc) which is why there is an expectation of only a moderately negative effect on underwriting profitability during the next 2 to 5 years.

Also, out of transparency and in support of your (at least my understanding) assertion that the future will compensate for the past, when the market turned in the early 2000s, against all odds, negative trends abated significantly (number of claims, payout per claim etc) and the medical liability market was quite profitable for a while (partly as a result of reserve redundancies and subsequent releases that more than compensated market players hurting from the residue of the previous underwriting softness). 

-----

Irrelevant addition:

Cycles are fascinating (underwriting etc and this could be applied to investing in general (not to be discussed during a perma-bull environment versus the bezzle phenomenon (as described by JK Galbraith))) and it also applies to fraud occurrence and discovery. Frauds typically peak at tops of economic activity, especially at points of critical turning points (before the crowd notices) as some 'managers' get more pressure than usual along the usual incentives to reach targets. Of course, frauds 'incurred' are not reported contemporaneously. Frauds are typically uncovered, against some resistance, later on and this 'development' typically leads to investigations, committees and such that then look into the matter when a new bull market is largely under way. 

-----

The underwriting cycle is dead. Long live the underwriting cycle. 

Charles VI of France was first thought to be 'beloved' but then it was realized that he was 'mad' (i guess some kind of adverse development) and people, at his death, sort of knew that cycles were cycles. 

 

[Viking]

1 hour ago, Cigarbutt said:

You are directionally correct and the idea is to guess if the 'excess' reserves being built now (something which we will find out in the future) are enough to compensate for the developing realization (something that we will find out in the nearer future) that previous years were characterized by deficient reserves.

-----

Longer answer

In the reference above, there is this graph which shows what is known now about the past but which, for the last 2 to 5 years, will significantly change within the next 2 to 5 years:

If history is any guide, the years 2016 and after will likely deteriorate and this deterioration will be recognized in future years. 

For fun (historical fun), let's look at what happened to the medical liability market before, as quite a representative example of most commercial lines (although typically more volatile ie more of this guessing about the past and the future). So what was known around 2001:

What happened (recognition of the past) after a couple of years:

The graphs are not quite comparing oranges to oranges but enough to draw some insights. Around 2001, it was reported that reserves were determined to be sufficient up to and including the year 2000. around 2005, it was realized that from 1997 on, policies had been underpriced and reached maximum adverse development in 2000, when things started to look better.

Anyways, usually there is relative pain as a result of this transition (typically negative underwriting income for a while) but this is variable and this time may be different (the market has turned relatively early versus the typical triggers ie large losses, threatened surplus, companies leaving market etc) and FFH has shown to be slightly better than peers (underwriting discipline, higher growth during hardening etc) which is why there is an expectation of only a moderately negative effect on underwriting profitability during the next 2 to 5 years.

Also, out of transparency and in support of your (at least my understanding) assertion that the future will compensate for the past, when the market turned in the early 2000s, against all odds, negative trends abated significantly (number of claims, payout per claim etc) and the medical liability market was quite profitable for a while (partly as a result of reserve redundancies and subsequent releases that more than compensated market players hurting from the residue of the previous underwriting softness). 

-----

Irrelevant addition:

Cycles are fascinating (underwriting etc and this could be applied to investing in general (not to be discussed during a perma-bull environment versus the bezzle phenomenon (as described by JK Galbraith))) and it also applies to fraud occurrence and discovery. Frauds typically peak at tops of economic activity, especially at points of critical turning points (before the crowd notices) as some 'managers' get more pressure than usual along the usual incentives to reach targets. Of course, frauds 'incurred' are not reported contemporaneously. Frauds are typically uncovered, against some resistance, later on and this 'development' typically leads to investigations, committees and such that then look into the matter when a new bull market is largely under way. 

-----

The underwriting cycle is dead. Long live the underwriting cycle. 

Charles VI of France was first thought to be 'beloved' but then it was realized that he was 'mad' (i guess some kind of adverse development) and people, at his death, sort of knew that cycles were cycles. 

 

@Cigarbutt thanks for posting in detail on this important topic. Could your concerns (developing losses on longer tail lines) be a key driver of the current hard market (especially when combined with falling interest income due to low bond yields)? My understanding is the hard market began in 2H 2019 - if true, that means we have just finished year three. And everything i am reading suggests the hard market will continue to run well into 2023. My guess is 4 years of hard market price increases should be able to provide a fair bit of cover for mistakes made in past years. But i will readily admit i do not understand this aspect of P&C insurance very well. 
————-

This topic has been discussed on past WRBerkley conference calls. The question is usually something like “given hard market has been running a couple of years already, why are we not seeing larger numbers for prior year positive development?” WRB answer: “we are being conservative”. My interpretation: “it is coming, just not yet” 

[Crip1]

18 hours ago, Viking said:

@Cigarbutt thanks for posting in detail on this important topic. Could your concerns (developing losses on longer tail lines) be a key driver of the current hard market (especially when combined with falling interest income due to low bond yields)? My understanding is the hard market began in 2H 2019 - if true, that means we have just finished year three. And everything i am reading suggests the hard market will continue to run well into 2023. My guess is 4 years of hard market price increases should be able to provide a fair bit of cover for mistakes made in past years. But i will readily admit i do not understand this aspect of P&C insurance very well. 
————-

This topic has been discussed on past WRBerkley conference calls. The question is usually something like “given hard market has been running a couple of years already, why are we not seeing larger numbers for prior year positive development?” WRB answer: “we are being conservative”. My interpretation: “it is coming, just not yet” 

I am not professing to be an expert on this matter, but the inflationary environment has to be a big unknown. This would drive, presumably, more conservative reserving.

 

-Crip

[glider3834]

6 hours ago, Crip1 said:

I am not professing to be an expert on this matter, but the inflationary environment has to be a big unknown. This would drive, presumably, more conservative reserving.

 

-Crip

agreed  - I think we are looking for an answer that can only be known over time but I would look at Andy Barnard's track record at Odyssey & Fairfax's track record on reserving over the last 10 years as well under Andy's watch. 

 

Also inflation continues to surprise to the upside you would expect interest rates to respond & Fairfax's short duration positioning means they can reinvest at higher yields. Whether we get to ROE target via underwriting or investment 'engine' doesn't really matter -  @Cigarbutt if you are subtracting your projected ROE due to lower underwriting result due to higher inflation, then I think we should also add to projected ROE via higher interest/investment income.

 

Fairfax has for last number of years been concerned about higher inflation, is it a reasonable expectation that they would also be building that into their reserving? Bearing in mind inflation is just one variable in the reserving process. 

 

Just a final one, paid losses ratio (actual payments on on claims/net earned premium) - this is a stat that WR Berkley has raised on their last two conference calls. This is worth monitoring. It has been trending down for last 5 years (ie higher percentage of their loss expenses are provisions versus actual payouts) as WRB & FFH both cautious on impact on their loss picks from economic/social inflation& due back log in court system due to covid. So we know they are sandbagging their loss reserves, whether these prove redundant or not will only be known over time.

Edited September 5, 2022 by glider3834

[Cigarbutt]

On 9/4/2022 at 5:59 PM, Viking said:

 concerns...

7 hours ago, Crip1 said:

...more conservative reserving.

1 hour ago, glider3834 said:

...should also add to projected ROE via higher interest/investment income.

 

 

-Inflation

Yes the inflation issue will tend to be incorporated into the reserving process and investment income should compensate at least partly. 

-Underwriting cycle

This last cycle turned before the usual triggers (see ROE picture below; hard markets usually triggered when ROEs fall much below 4%). Also, the industry reserve leverage hit a low of 0.75 at YE 2021 compared to 2.0 in the 80s. So the industry was well capitalized even in the context of 16 uninterrupted years of reserve releases. So, uncharacteristically, the cycle morphed into a hard market. FFH will tend to do better and maybe there is a new era of overall market discipline?

At any rate, if the idea was betting (with leverage) on Fairfax in the early 2000s in order to harvest the benefits of investment float (growing in a hard market) management, then prior years' adverse development was a concern.

At this point, there is no significant "concern" coming from the underwriting side, apart from a possible (likely IMO) need to moderately reduce expectations about underwriting profitability within the next 2 to 5 years (a period which may correspond to a relevant holding period if FFH has become a trading vehicle instead of a core holding).

-----

Of course, the alternative opinion (reserves will remain redundant, overall) could apply. This occurred in the medical liability lines after the hard market in the early 2000s.

BRK has been present and has become a large player in that part of the market. Equipped with a long term view, staying the course, the idea is to direct more capacity when the market hardens and consolidate your position (acquisitions) before the market enters a new hardening phase (happening now but the underwriting and reserve release profiles suggests quite another sudden and more prolonged increase in premiums). If possible, the ideal scenario is to enter the game without having to carry previous years' adverse development. This argument is similar to those who argue about the need to be fully invested versus available capacity (or financial flexibility to achieve capacity when needed).

It looks like FFH has played the relative game quite well lately but it may not completely escape industry-wide trends?

-----

Anyways, the key variables now are float management and downside risks. Keep at it.

[Daphne]

Reinsurance Rates Rising
 

Global geopolitical tensions, high inflation and climate change have heightened demand for risk protection and will lead to increased premiums, top reinsurers said at the industry’s annual gathering in Monte Carlo.

Reinsurers insure the insurers and have been pushing up premiums in recent years as they have faced higher losses.

“On top of impacts from COVID-19 and increasing losses from natural catastrophes, the reinsurance industry is now confronted with issues like inflation, risk of recession and geopolitical tensions,” said Moses Ojeisekhoba, Swiss Re’s chief executive officer reinsurance, in a statement on Monday.

STORY CONTINUES BELOW ADVERTISEMENT

“As we see cost drivers accelerating in this dynamic risk environment, insurance premiums must be carefully calibrated to keep pace,” he added.

Reinsurers meet their insurance clients in Monte Carlo to hammer out contracts ahead of the key Jan. 1 reinsurance renewal season.

Rates could rise in the “mid-single digit” per cent range, S&P analysts said last week, while a Moody’s customer survey showed expectations for double-digit rate rises in U.S. property reinsurance.

Rates could rise by 10 per cent or more in some markets, Munich Re and Hannover Re executives also told media briefings on Sunday and Monday, given the strength of inflation.

“It’s not an easy time,” Hannover Re CEO Jean-Jacques Henchoz said.

“We have a general environment which is very volatile because of the direct and indirect implications of the (Ukraine) war … inflation is the driving topic in many of our discussions here in Monte Carlo.”

Hannover Re highlighted the impact of the war in Ukraine, which Russia calls a “special military operation”, on the aviation and marine markets. 

 

However, the changing environment also provides some opportunities for insurers and reinsurers.

The Swiss Re Institute expects $33-billion to be generated in commercial premium volumes in the period from 2022 to 2026 as companies relocate their supply chains closer to their home countries.

Swiss Re also plans to further grow its natural catastrophe portfolio, where the market is forecast to grow to $48-billion from $35-billion in the next four years.

[Viking]

Bond yields are spiking and most maturities are at or near decade? highs. Market is now forecasting Fed funds rate of 4.25% in March 2023.

 

Higher rates are very good news for Fairfax and their $35-$36 billion bond portfolio (1.2 year avg. duration). Fairfax said in Q2 release the current annual run rate for interest and dividend income was $950 million. This will be higher when they report Q3 results. If they are able to get an average yield of 4% on their bond portfolio = $350 million/quarter = $1.4 billion/year.
 

Dividends tracking at $35 million/quarter = $140 million/year

investment expenses tracking at $15 million/quarter = $60 million/year

 

Add the 3 items and we could see Fairfax earnings in the interest and dividend bucket $375 million/quarter = $1.5 billion/year = $63/share. 
—————

So what could Fairfax earn in operating income in 2023?
1.) underwriting (95CR) = $1 billion

2.) interest and dividend income = $1.5 billion

3.) share of profit of associates =$500 million (primarily Atlas + Eurobank)

Total = $3 billion/year = US$127/share (pre-tax)

 

Fairfax shares are trading at US$498 = 4 x estimated 2023 operating earnings (pre-tax). Any gains on the significant equity portfolio (much of which is  priced today at bear market prices) is just gravy.

—————

Hard market is boosting underwriting income. Rising bond yields is spiking interest and dividend income. Associate earnings are chugging along (Atlas/Eurobank). Stock is wicked cheap. Got it!

—————

My number above for underwriting does not include runoff, which will likely be a drag of $150 million/year = $6/share.

 

 

 

Edited September 13, 2022 by Viking

[Viking]

At current bond yields insurers will be taking another significant hit to earnings (mark to market losses) and book value when they report Q3 earnings. We could see three quarters in a row where many P&C insurers report very large hits to book value. Will regulators start to care? Does this result in hard market continuing into 2023?
 

Does Fairfax’s positioning of its bond portfolio (1.2 year average duration) start to benefit its underwriting (it can keep the petal to the metal on growth)? Will other insurers have to slow their growth? 
—————

US Treasury Yields

             Dec 31.       June 30.     Sept 13

1 month.  0.06.          1.28.              2.48

1 year.      0.39.          2.80.             3.88

2 year.      0.73.          2.92.             3.74

3 year.      0.97.          2.99.             3.75

5 year.      1.26.           3.01.              3.58

10 year.    1.52.           2.98              3.43

 

Most P&C insurers have an average duration of around 4 years on their bond portfolio. Fairfax, at 1.2 years, and WR Berkley, at 2.4 years, are the two outliers.

Edited September 13, 2022 by Viking

[glider3834]

'Climate, conflict and capital are coalescing to create a “tipping point” for the reinsurance market, with 2022 mid-year renewals seeing rate increases hit their highest levels since 2006, according to a new report by international insurance broker Howden.'

 

https://www.insurancebusinessmag.com/au/news/breaking-news/reinsurance-market-coming-close-to-a-tipping-point--report-419762.aspx

 

I did this chart below - for the last 10 years, Fairfax has tripled gross premiums written (GPW) but at same time prioritised insurance over reinsurance with reinsurance share of total gross premium falling from 34% to 24%  (& if you include non-consolidated insurers like GIG or Digit that write principally insurance, this gross premium split would skew further toward insurance)

 

 

Peter Clarke on Q2'22 call  - comment suggesting more rate pressure in reinsurance 

 

Peter Clarke

I think the only thing I'd add is we're still getting rate and we're still getting fairly good rates, like 7.5%. So that's going to drive growth alone. And then different lines of business are increasing still DNO, VNO for example, in the US has stabilized. But property CAT, a lot of raise, a lot of capacity there. And a lot of opportunity. So we're seeing a little bit more on the reinsurance side, less on the insurance side, but I think we'll see -- we'll still see strong growth in the next six months.

Edited September 14, 2022 by glider3834

[Viking]

Can we now reasonably say that the base case is that Fairfax should be able to earn about US$100/share moving forward? With the shares closing today below $500 that is a PE of 5.
 

Using the ‘back of the napkin’ method: With a CR of 95. And with an after tax return of 4% on $50 billion investment portfolio. $750 million + $2 billion = $2.75 billion / 23.7 million shares = $116/share. Bit of a buffer of about $350 million… enough to cover minority interests, loss from runoff, interest expense etc?
 

Looking out another year, if:

1.) the hard market continues (and we get another year of +15% top line growth)

2.) Fairfax is able to push duration of the bond portfolio out to +3 years at attractive yields

3.) Fairfax finds one or two assets to monetize

4.) Fairfax buys back $500 million or another $1 billion in stock

Well… $100/share will start to look low as a base case.

Edited September 17, 2022 by Viking