Fairfax 2021

[Phoenix01]

FFH taking advantage of the BB price spike will be huge for Q1 results.  I have been thinking about this and was wondering if they might have used a term repo to lock in some financial flexibility.

 

This would have allowed FFH to sell BB shares at a given price (let's say $15) with an obligation of buying them back at a higher price at a time of their choosing (let's say$17 in 2 years).  In effect this becomes a loan with no monthly payments and it also reduces leverage.

 

If the price drops before the end of the term, it allows them to avoid the tax liability on the gain.  If the price increases they get to keep the potential upside.

 

Does this make sense?  How and when would FFH have to disclose such a transaction?

 

 

[SafetyinNumbers]

I assume if FFH could do enter into the long side of total return swap on its own stock then it could enter into an inverse total return swap on BB. It’s pretty easy to execute since, FFH can lend its own shares to the counterparty making the borrow cost and risk free.

 

Based on no SEDI filings necessary for the FFH total return swap, I don’t see why they would need to disclose for BB.

[chrispy]

I assume if FFH could do enter into the long side of total return swap on its own stock then it could enter into an inverse total return swap on BB. It’s pretty easy to execute since, FFH can lend its own shares to the counterparty making the borrow cost and risk free.

 

Based on no SEDI filings necessary for the FFH total return swap, I don’t see why they would need to disclose for BB.

 

Based on what we learned from the most recent filings and call, I think that thought is probable as well

[Xerxes]

Just based on how confident he sounded on Q4 call, the complete lack information he provided on BB, and having the mechanics in place to do something, plus a healthy dosage of value investor mentality (must squeeze when overvalued), i think we can be fairly certain that he found a way to squeeze the BB lemon.

 

What i do want to see is the same TRS treatment on Stelco and Resolute (on the long side), since it is unlikely that he would have added to the shares.

[FairFacts]

It would be unforgivable not to have locked in at least a meaningful portion of the gain. Unless there is a technical/contractual limitation I will be amazed if something hasn’t been done. Would certainly cause one to question why not?

[Viking]

Not the start to 2021 that insurance companies were hoping for. Early days in terms of understanding actual costs but if this estimate is in the ball park... ouch! Anyone know what Fairfax’s exposure is to Texas? The silver lining for insurers is this will likely prolong the hard market for pricing,

 

Biden declares major disaster in Texas as focus shifts to who is responsible for the winter weather crisis

- https://www.washingtonpost.com/nation/2021/02/20/winter-storms-texas-weather-updates/

 

The catastrophic winter storm was expected to become the “largest insurance claim event in [Texas] history,” said the Insurance Council of Texas, a trade group, which estimated the damage would far outpace the $19 billion in claims from Hurricane Harvey in 2017.

[Phoenix01]

Based on the historical percentage of US cat losses, an 18B event would cost FFH around $150M.

[Cigarbutt]

^There are many ways to guestimate this. The point also is to make sure exposure was adequate, appropriately priced and to make sure that you can write policies in the future that will reflect the risk.

The estimated net underwriting cost may lie between 300-350M.

Here are the assumptions:

-Uri will be comparable to Harvey (2017) (assume Uri will have total insurable costs slightly lower; significant residual uncertainty there)

-In 2017, Harvey cost FFH 252.4M (122.9 at Allied World; the ‘catastrophe’ underwriting results in 2017 had raised questions about the acquisition) (assume there were no major reserves adjustments after, apart from generally favorable trend over the years)

-In 2017, this loss = 2.6 combined ratio points (11.8 points for AW!)

-Comparing 2020 to 2017 yr-end numbers, FFH’s SE has increased by 0.3% but NPW has increased by 48.9% (with AW and ORH really growing in 2020), so it may be reasonable to expect something like a 2.2 to 2.5% CR cost on the total premium base

 

[petec]

https://www.thehindubusinessline.com/companies/fairfax-preps-for-bigger-play-in-infra-sector/article33896181.ece

[Viking]

Stelco on Thursday announced it will be re-instating its $0.10/share dividend (suspended in 2020 due to pandemic.

 

Fairfax owns 12.2 million Stelco shares so this will result in dividend income of CAN $1.22/ quarter or $4.88 million per year. Stelco also will likely be paying special dividends at some point in 2021 due to the spike in steel prices; during the last spike in prices in 2018 Stelco paid out $2.70/share in special dividends.

 

Bottom line, the ‘interest and dividend’ bucket has been dropping like a stone for Fairfax the past 4 quarters. Q1 will also be challenged. However, as 2021 progresses and the economy picks up steam we should see this trend bottom and then actually start to grow again. Stelco is not a big deal on its own (large special dividend would be); however, when combined with other Fairfax equities the increases for 2021 should add up to a meaningful number.

 

My guess is the ‘interest and dividend income ‘ bucket will be a headwind for Fairfax for 1H and a tailwind in 2H.

[Santayana]

https://www.thehindubusinessline.com/companies/fairfax-preps-for-bigger-play-in-infra-sector/article33896181.ece

 

The mention of container handling terminals made me wonder what sort of synergies there might be with Seaspan.  Then I realized how Stelco could play into that as well. 

 

[petec]

https://www.thehindubusinessline.com/companies/fairfax-preps-for-bigger-play-in-infra-sector/article33896181.ece

 

The mention of container handling terminals made me wonder what sort of synergies there might be with Seaspan.  Then I realized how Stelco could play into that as well.

 

I thought the railways comment was interesting. Dear lord does India need investment in its railways.

 

Seaspan and Stelco are independent so I wouldn’t draw those links necessarily, although something could happen. 

[petec]

Stelco on Thursday announced it will be re-instating its $0.10/share dividend (suspended in 2020 due to pandemic.

 

Fairfax owns 12.2 million Stelco shares so this will result in dividend income of CAN $1.22/ quarter or $4.88 million per year. Stelco also will likely be paying special dividends at some point in 2021 due to the spike in steel prices; during the last spike in prices in 2018 Stelco paid out $2.70/share in special dividends.

 

Bottom line, the ‘interest and dividend’ bucket has been dropping like a stone for Fairfax the past 4 quarters. Q1 will also be challenged. However, as 2021 progresses and the economy picks up steam we should see this trend bottom and then actually start to grow again. Stelco is not a big deal on its own (large special dividend would be); however, when combined with other Fairfax equities the increases for 2021 should add up to a meaningful number.

 

My guess is the ‘interest and dividend income ‘ bucket will be a headwind for Fairfax for 1H and a tailwind in 2H.

 

Biggest driver of this line will be rising float/equity (I think they will get to 4x) and any rise in interest rates.

[Xerxes]

wow

4x the current interest/dividend income over a long i imagine and over much larger float?

 

As an newbie on the topic, the way i see it, the think with a long-term bear market in bonds, is that as interest rate slowly go up, and if any fixed-income investor decide to lock in those higher rate, while they naturally get higher income, they also automatically expose themselves to incur long-term unrealized capital losses as decade goes by and interest rate continues to grind higher.

 

It is much easier to be in a long-term bull market in bonds.

[no_free_lunch]

Thank you Phoenix , cigarbutt for your loss estimates.

 

I happened to see that there are double digit price increases on home insurance in Florida. Could insurance industry act rationally and simply raise rates to recapture the loss?

[Phoenix01]

Global insured Cat losses average $52B per year over the past 26 years.  This will have a significant impact, but if we have a below average hurricane season it may not be that bad.

 

The premiums are going in the right direction and will make absorbing these Cat losses easier.  Net premiums of $4B in Q1 with a CR of 95% gives $200M of profit.  That should cover the Texas Cat with a little profit left over.

[petec]

wow

4x the current interest/dividend income over a long i imagine and over much larger float?

 

 

No, what I mean is that as they expand underwriting in the hard market float will grow. I think float/equity is about 3.3x at the moment. It can go to about 4x. That means more bonds and interest per unit of equity. Higher leverage basically, but via float, not debt.

 

I disagree re losses in a bond bear market. That’s true if they choose to extend duration before the end of the bear. But if they stay in the short end of the bond market, they can keep rolling bonds onto higher rates throughout the bear, without incurring losses.

[Cigarbutt]

...

...the way i see it, the think with a long-term bear market in bonds, is that as interest rate slowly go up, and if any fixed-income investor decide to lock in those higher rate, while they naturally get higher income, they also automatically expose themselves to incur long-term unrealized capital losses as decade goes by and interest rate continues to grind higher.

It is much easier to be in a long-term bull market in bonds.

i agree with petec if interest rates rise slowly. Look below:

As rates rise, reinvested coupons and reinvestments at maturity will (with some lag) make the bond yields to gradually rise. Inflation-indexed bonds could also be bought. They can also hold the longer duration bonds to maturity and the unrealized bond losses remain so.

Looking at the whole industry, there is some kind of consensus that inflation is coming:

There may be a problem though if winter is coming but that's another story and one which has come to maturity for Fairfax.

 

They have changed their stance since 2016 and have expressed that they are expecting inflation along general economic growth and seem to espouse the reflation narrative. In the 2019 report, Mr. Watsa mentioned: "Our fixed income portfolio, which effectively comprises 70% of our investment portfolio, has a very short duration of approximately 1.5 years and on average is rated AA-. Very high quality and very short term to maturity – so rising interest rates would not impact our portfolio!" They also have derivative exposure to protect against rising rates for the residual longer term FI portfolio. In 2020, they switched some short term government paper for mid term corporates but duration remains low. If (IMO that's a big if) inflation comes, they are ready to benefit. Wild inflation would hurt but the hurt would be widespread.

-----

A related aspect that needs to be considered for an insurer if inflation is for real is that the contracts for coverage are made with a specific inflation target. Especially for longer tail lines, if inflation is higher than anticipated, the eventual cost of claims is paid with more dollars. That was an issue in the 70s and early 80s and since insurers tend to react with a lag and since regulators may not accept significant rate increases, underwriting may suffer for a while.

-----

FWIW, i've kept a diary of what would have happened if they had kept their long term fixed income exposure after 2016.

[Xerxes]

So, the general idea is that we had a four-decade bull market (1980 through say 2020 for simplicity).

And for argument sake let's say we will have a slow grinding bear market over the next 40 years (2020 through 2060).

 

Fairfax long term view was that the bull market ended (or was soon to end) when they switched out in 2016, and have been keeping on the short end only. After a few false start, (i.e. The initial stages of the pandemic contributed to the pause on the narrative), lets say it is now slowly heading that way.

 

I understand if they were to stay on the short end, they will float like a boat and ride that wave, while never having material exposure to capital loss. But at some point, let's say 20 years from now for illustration's sake, they will gradually lock in because they perceive that the rate has peaked. That "lock-in" itself could be 20 years too early, if the actual peak rate arrives in 2060. That is what I meant.

 

These long term interest rate movements are very long term and very gradual, where you could be in the middle inning of what is actually a larger first inning (i.e an inning within inning). Think these are very much outside the scope of most observers. Even now, only in hindsight, economists are looking back to explain why we had a 40 years long drop in interest rate.

 

But maybe none of this matters, because by then we wont be around anyways i think and/or already enjoying retirement on Planet Mars, thanks to SpaceX-Tesla consortium.

[Viking]

And it is not a given that the current bull market in bonds is even over :-) Yes, we likely will see inflationary pressures in 2021 as the economy recovers. But in 2022 as the economy normalizes we may see disinflationary forces that were in place pre-covid re-established. No idea which way we go the next 10 years (mild inflation or disinflation / mild deflation).

 

Lacy Hunt would likely say the core issue continues to be total debt. And more debt = more disinflation or even mild deflation over time.

[petec]

Xerxes, you’re right that they might lock in too soon. But there are a lot of “ifs” in your thesis and personally I’ll wait until it’s happening before I worry about it!

 

The one thing I think you might have wrong is the speed of an inflationary bond bear market. Yes, the bull has lasted 30 years - but the inflationary bear that preceded it was much quicker. Once people think inflation is coming, they start getting rid of cash, the money velocity rises, and inflation can come thick and fast.

 

But as others have said, we may be a long way from the turn yet.

[petec]

Announcement re efforts to renegotiate holding company debt at Atlas Mara:

 

https://otp.tools.investis.com/clients/uk/atlas_mara1/rns/regulatory-story.aspx?cid=744&newsid=1453476

[TwoCitiesCapital]

I understand if they were to stay on the short end, they will float like a boat and ride that wave, while never having material exposure to capital loss. But at some point, let's say 20 years from now for illustration's sake, they will gradually lock in because they perceive that the rate has peaked. That "lock-in" itself could be 20 years too early, if the actual peak rate arrives in 2060. That is what I meant.

 

I would think they could begin locking in significantly before year 20. It obviously depends on starting yields/durations and how quickly rates are rising, but in prior bond bear markets you could recoup the principal losses from rising rates with higher income/higher reinvestment income by year ~4.

 

So even if you expect rates to rise, you can buy a 10-year bond today and still be ahead of short-term bonds by years 5-6 even if you're right about rates moving higher.

 

I don't mind Fairfax's move to short term bonds. It's certainly saved them from some pain. But I would hope that they'd start moving incrementally back to 10 year and longer bonds if rates exceeds 1.5 -2.0%.  Just in recognition that these things don't move in a straight line, roll down yield becomes more attractive as the curve gets steeper (short term still anchored @ 0%), and any unrealized losses from rising rates will likely be mitigated by year 4-5 of holding the bond anyways.

 

And given my ultimate views that we are NOT in a sustainable inflation environment, I would think this exposes them to potential gains from a disinflationary/deflationary environment that they missed in 2018, and again in 2020, when 10-year yields dropped from 3 25% all the way down to 0.5% over that 2.5-3 year period.

 

 

And it is not a given that the current bull market in bonds is even over :-) Yes, we likely will see inflationary pressures in 2021 as the economy recovers. But in 2022 as the economy normalizes we may see disinflationary forces that were in place pre-covid re-established. No idea which way we go the next 10 years (mild inflation or disinflation / mild deflation).

 

Lacy Hunt would likely say the core issue continues to be total debt. And more debt = more disinflation or even mild deflation over time.

 

This is my view. I'm very heavy into commodity companies - but mostly because they're cheap and not because I'm expecting massive inflation.

 

I think we'll get a pop in 2021 due to the trillions pushed into circulation during 2020, but I do think at some point in 2022 we get back to disinflation as rates can never rise significantly with debt loads/equity markets where they're at.

 

 

[petec]

Bear in mind rates and inflation don’t have to move the same way. I can easily imagine a period of financial repression, with policy designed to drive slightly higher inflation while keeping rates low. Dangerous game to play, but ultimately the only realistic solution to high debt levels.

[Viking]

Bear in mind rates and inflation don’t have to move the same way. I can easily imagine a period of financial repression, with policy designed to drive slightly higher inflation while keeping rates low. Dangerous game to play, but ultimately the only realistic solution to high debt levels.

 

Defining just exactly what ‘inflation’ is is part of the challenge. One example is real estate in Vancouver. Single family home prices are expected to increase this year 20-30% (maybe as soon as this spring). Crazy. And prices were already at bubble levels. This looks like asset inflation to me. Ever rising prices :-)

 

So i agree rates and ‘inflation’ often don’t move in the same way.