Q1 2021

[Daphne]

- Fairfax Financial Holdings Limited (FFH.TO) after trade Thursday announced net earnings of US$806.0 million ($28.91 net earnings per diluted share after payment of preferred share dividends) in the first quarter of 2021 compared to a net loss of $1,259.3 million ($47.38 net loss per diluted share after payment of preferred share dividends) in the first quarter of 2020. Book value per basic share at March 31, 2021 was $497.23 compared to $478.33 at December 31, 2020 (an increase of 6.1% adjusted for the $10 per common share dividend paid in the first quarter of 2021).

[StubbleJumper]

A few observations:

1) Nice headline EPS number this quarter, and if you're the type to calculate TTM earnings, those look pretty good too!  And this time, the quality of earnings is not really in question.

2) As we discussed a few weeks ago, the CR was a bit higher than we would have like to have seen.  There were more cat points from the US winter storms than I had anticipated, but more concerning is that the favourable development seems to be evaporating (Cigarbutt invested a bit of time last year to discuss the trend in favourable development, and I am afraid that he seems to have been proven correct).  As long as it doesn't flip over into adverse development....

3) Nice increases in net written, with the exceptions of Crum and Brit.  There was a good (and disturbing?) explanation of why Brit's net written shrunk in Q1, which is that it entered into a very large reinsurance arrangement that seems to cover everything but the kitchen sink.  But, that strikes me as a bit of a funny thing to do, and it's unclear what kind of messes this was intended to manage.  Anyway, for Brit, it is what it is.  Does anyone have a good understanding about why Crum's premium growth was so tepid?  Crum was capital constrained for part of 2020, but I had been hoping that was a thing of the past.  Taking into account all of the anecdotes about double-digit price increases, it looks to me as if Crum might be non-renewing some policies.  If anyone has any insight, I'd love to see it.

4) FFH hasn't done much with its fixed income duration and doesn't seem to be reaching for yield, so that strikes me as good.  Looking at the fixed income disclosure on page 12 of the Q1, FFH is up to almost $2 billion in "Other government" bonds (not Cdn Federal or provincial, not US Federal, state or muni).  It doesn't much matter in the grand scheme of things, but does anyone have any ideas what that might be?  Agency debt of some form?  Hopefully not Peruvian sovereign debt or some other dubious instrument!

5) The total return swaps are working out beautifully.  FFH didn't have liquidity to mount a meaningful buyback, but they still managed to benefit from a long position in one of the cheapest stocks out there -- FFH itself!

6) BV is $497.  Adjusted BV is what, about $515 or thereabout?  FFH is still pretty cheap.

 

 

SJ

[petec]

1.9x TBV. 

[Viking]

Results were pretty much in line with what was pre-announced; my guess was earnings would hopefully come in around $30/share and that is pretty much what we got. BV was up nicely to US $497 despite US $10 dividend payment. Great quarter. Bottom line, Fairfax needs to continue to execute and deliver results (under promise and over deliver).

- hard market is benefitting insurance companies

- rebound in equity markets is benefitting investment portfolio (understatement)

Interesting to see the increase in the number of FFH shares held via TRSwap (including April). I wonder what the exit strategy is for these?

Some of their equity holdings continue to suffer from Covid; this will slowly become more of a tailwind as vaccinations increase and economies open back up. 

The interest and dividend bucket continues to fall; this might be a big deal for analysts (who tend to look only at operating income and ignore investment gains).

Refinanced a bunch of debt in Q1 (lower interest rate and much longer 10 year term) at a cost in Q1 of $46 million. Great long term move despite hit to Q1 EPS.

In Q2 FFH will be receiving $375 million from OMERS for 14% of Brit. Also $730 million from CVC for Riverstone. = $1,105. From this $500 million will repay the credit facility. What will the remaining $600 million be used for? I wonder if we see more share repurchases (they really seem to like that insurance company called Fairfax). 

And it put a smile on my face to read that they held NO short equity total return swaps  

Edited April 30, 2021 by Viking

[Viking]

Below are some notes from Q1 conference call. The key take away for me is deleveraging (debt reduction)  is the near term priority. $1 billion in debt reduction in Q2 and ‘further over time’. Smart to do this when times are good. Fairfax needs to re-build management credibility and this is another step in the right direction. This also perhaps explains why they increased the number of FFH shares owned via TRSwap (and why they did not just buy back more shares outright). 

The insurance side of Fairfax is performing well (thank you hard market). bond portfolio is positioned for higher rates. The stock portfolio is positioned for a rebound in economic activity (and higher stock prices for cyclicals). Stock is trading at 0.92 X BV which does not look expensive. If economic growth is strong in 2021 Fairfax’s stock price should continue to increase nicely. 
——————————

- virtuous circle: hard market resulting in strong growth and improving underwriting results. At same time value investing is coming back and delivering strong returns. Should continue.

- mentioned BAC as equity holding? 

- deleveraging appears to be a priority. Will be reducing debt from $7 billion to $6 billion in Q2 and further over time. 

- total return swaps (ex Fairfax position). ‘Lots bought have already been sold’ (i think this is what Prem). Short term in nature.

- significant holdings in cash and ST investments: will benefit if rates rise further. Bonds today offer no margin of safety.

- insurance companies today are very well capitalized. Benefitting both from underwriting earnings and strong investment gains.

- expense ratio improving as growth in premiums is being driven primarily by rate increases. 

- Riverstone UK and Brit transactions are linked. Riverstone UK transaction is expected to close in Q2. Hold up is CVC and their discussion with regulators. Once this deal closes, OMERS will be using part of their proceeds to complete their purchase of $375 million of Brit.

- why increase size of TRS on FFH and not simply buy more stock on open market? Think stock at current prices offers very good value for long term investors. 

- priorities: 1.) strong financial position 2.) take advantage of hard market 3.) take advantage of stock price

Edited April 30, 2021 by Viking

[Xerxes]

While Prem is not selling his Resolute into the rally, Brookfield is .... $1.25 billion worth 

Brookfield sells $1.25-billion of West Fraser Timber shares as lumber prices soar - The Globe and Mail

"The sales – totalling 14.8 million shares – have cut Brookfield’s ownership of West Fraser from nearly 20 per cent to 7.3 per cent, according to Brookfield’s filings with securities regulators. In the past two weeks, Brookfield has sold nearly 1.24 million shares for $128-million; in a single day, April 1, it sold more than six million shares for $517-million."

[StubbleJumper]

45 minutes ago, Xerxes said:

While Prem is not selling his Resolute into the rally, Brookfield is .... $1.25 billion worth 

Brookfield sells $1.25-billion of West Fraser Timber shares as lumber prices soar - The Globe and Mail

"The sales – totalling 14.8 million shares – have cut Brookfield’s ownership of West Fraser from nearly 20 per cent to 7.3 per cent, according to Brookfield’s filings with securities regulators. In the past two weeks, Brookfield has sold nearly 1.24 million shares for $128-million; in a single day, April 1, it sold more than six million shares for $517-million."

 

Give it a bit of time.  Dumping Resolute at the top of the lumber price cycle is probably the thing to do.  But, I wouldn't blame Prem if he was currently trying to find a sucker to takeover the whole company, which might provide an additional 15% or 20% control premium for existing shareholders.  If Prem can offer that potential sucker a lock-up on FFH's shares and possibly Chou's shares or some other large holder's shares, it would provide a nice beachhead for a takeover. 

If there's no action taken on the Resolute shares by the end of Q3 or Q4, we should really question Prem's bona fides as a value investor.

 

SJ

[Xerxes]

Agreed on timeline.

The only concern is that the strategic sucker who will take the other side of the trade must know enough about the cyclicality of the business, to be wary to pay premium on top of the market. I just hope Fairfax are being creative to hedge somehow the upside gained thus far. 

[Viking]

What Fairfax does with Resolute and Stelco will really depend on their assessment of whether we are beginning a commodity super cycle. If lumber and steel prices stay elevated for the year (as futures prices currently forecast) both stocks should trade much higher later in the year. And likely pay large special dividends.

Resolute has a carrying value of US $134 million. The current run up in the stock price is not impacting BV. A sale would drive a nice increase both EPS and BV in the quarter it is booked.

Stelco is mark to market so the impact on FFH EPS and BV is updated each quarter.

Edited April 30, 2021 by Viking

[Xerxes]

4 hours ago, Viking said:

- mentioned BAC as equity holding? 

I just finished listening to the re-run.

He mentioned BAC twice in fact - second time as a contributor to FFH gains in Q1 (if i heard correctly). I am guessing we will see it in the 13F for Q1 as a new addition. Most probably a Wade-like allocation. Glad to see to tilt to quality continue even though this might be a small size position vs. the larger ones. 

Depending when they bought in Q1, this could be as high as 35% gain on Bank of America.

[petec]

1 hour ago, Viking said:

What Fairfax does with Resolute and Stelco will really depend on their assessment of whether we are beginning a commodity super cycle. If lumber and steel prices stay elevated for the year (as futures prices currently forecast) both stocks should trade much higher later in the year. And likely pay large special dividends.

Resolute has a carrying value of US $134 million. The current run up in the stock price is not impacting BV. A sale would drive a nice increase both EPS and BV in the quarter it is booked.

Stelco is mark to market so the impact on FFH EPS and BV is updated each quarter.

Exactly. I suspect Stelco’s EV has actually gone down this year given how much cash they will generate. I don’t know the maths for Resolute. But I wouldn’t jump to the conclusion that just because the stocks are up, Fairfax should sell. 
 

I’m also somewhat sceptical Fairfax *can* sell Resolute in the market. They own too much. I wonder if they don’t end up controlling RFP through buybacks and just dividending cash to themselves. 

[StubbleJumper]

Q1 Conference call transcript has been posted at: https://seekingalpha.com/article/4423036-fairfax-financial-holdings-limited-frfhf-ceo-prem-watsa-on-q1-2021-results-earnings-call

[petec]

Couple of things I thought were interesting on the call.

First, $2bn of TRS on their own stock at an average price of $372 vs $1.4bn at $343 in the 4q. By my maths that means they added $600m at $460 per share. Is that how it works?

Second, re: being restricted on Blackberry, "we checked it ten times if we checked it once". I infer they really wanted to sell.

Third: it's taken years but they have swung fully from being worried about deflation to being worried about inflation. I am not sure whether to congratulate them for chancing their minds or criticise them for taking so long, but it's notable.

Edited May 1, 2021 by petec

[Viking]

On Friday RBC released its Q1 report on Fairfax RBC and raised its target price for Fairfax to US $550 (from $500).

• Fairfax has a very strong presence in specialty insurance and large account. These are the segments of the market that are experiencing the best rate increases and there is little sign that rates won’t continue higher throughout 2021 and likely into 2022. Equally, there is ample evidence that the rate increases achieved over the last 12-18 months are finding their way into accident year margins. For the quarter the accident year combined ratio was 90.9% we don’t the company has ever posted an accident year combined ratio that low... Bottom line, the P&C businesses have he best earnings visibility they have ever had and we don’t see any signs this will let up near-term.

Edited May 1, 2021 by Viking

[treasurehunt]

First, $2bn of TRS on their own stock at an average price of $372 vs $1.4bn at $343 in the 4q. By my maths that means they added $600m at $460 per share. Is that how it works?

I get $600m at about $440 per share. (2000*372 - 1400*343)/600 = 440.

[petec]

1 hour ago, treasurehunt said:

First, $2bn of TRS on their own stock at an average price of $372 vs $1.4bn at $343 in the 4q. By my maths that means they added $600m at $460 per share. Is that how it works?

I get $600m at about $440 per share. (2000*372 - 1400*343)/600 = 440.

Yes. Fat fingers on my part. But I’m still wondering if that’s actually how it works. Or could the overall average be impacted by rewriting older, expiring swaps at higher prices? I just don’t know how these things work. 

[Daphne]

Yes. So, Tom, in terms of Fairfax shares as you see in the press release, right, we have about 730 million, 2 million shares at approximately US$372. This is all of U.S dollars, 730 million is the total return swaps in Fairfax. And, of course, it's already doing well. [Indiscernible] on an opportunistic basis, we've looked at buying some common shares, Tom, but they're not long-term and they're the ones that we bought quite a bit, we've already sold. And so we continue to look at opportunities, but it's short-term stuff, meaning [multiple speakers] for a long -- longer period of time.

Would anyone on the board hazard a guess to what Prem is saying here and what it means?

[TwoCitiesCapital]

14 hours ago, petec said:

Yes. Fat fingers on my part. But I’m still wondering if that’s actually how it works. Or could the overall average be impacted by rewriting older, expiring swaps at higher prices? I just don’t know how these things work. 

Typically they swaps will "reset" on a monthly or quarterly basis. But they're not like options - there is no upfront premium so a dollar gain in the stock = a dollar gain in the swap (less financing costs).

So if you buy $1B of swaps that go up by 50%, the seller of the swap would pay you $500 million and you'd pay them the financing (LIBOR+spread) for the period. In reality, the payments are netted and flow one direction. 

Then, the swap "resets". The notional is now $1.5B and all P&L and financing will be paid on this $1.5B figure now.  The LIBOR+spread rate resets to the new observation point and P&L and collateral obligations are reset for the next period. 

Now, it's NOT clear to me if Fairfax is discussing these things at cost or not. The move from $1.4B to $2B - is that a difference in cost implying Fairfax added $600 million in additional swaps? Or is it reflective of the 30+% rally we've seen in USD terms meaning Fairfax only added $150-200 million on new swaps and the old swaps reset for $1.4B to $1.8B? This is what I would have to look more into to know for sure. 

[Xerxes]

4 hours ago, Daphne said:

Yes. So, Tom, in terms of Fairfax shares as you see in the press release, right, we have about 730 million, 2 million shares at approximately US$372. This is all of U.S dollars, 730 million is the total return swaps in Fairfax. And, of course, it's already doing well. [Indiscernible] on an opportunistic basis, we've looked at buying some common shares, Tom, but they're not long-term and they're the ones that we bought quite a bit, we've already sold. And so we continue to look at opportunities, but it's short-term stuff, meaning [multiple speakers] for a long -- longer period of time.

Would anyone on the board hazard a guess to what Prem is saying here and what it means?

He was asked about the TRS, but he diverted his answer and start talking about buying the market dip in 2020. He is referring to the trade that Wade and the team put in. That is why he said we already sold. See letter from March, where he says that they sold about $200 million of shares. It confused me as well, because i thought he was saying he un-winded the TRS trade, but he also said he added to it, which made no sense. Looking back i think he was talking about the Wade trade where he says they already sold some.

"After the March/April crash in the stock market, we could not resist buying Exxon shares at a dividend yield of 10.5%, Canadian banks at an average yield of 6.1% and some other companies like Royal Dutch Shell, Alphabet, FedEx and Helmerich & Payne at very attractive prices. We sold approximately half of them in 2020 for a profit of $212 million or an average gain of 40% on our investment."

Edited May 3, 2021 by Xerxes

[petec]

6 hours ago, Xerxes said:

He was asked about the TRS, but he diverted his answer and start talking about buying the market dip in 2020. He is referring to the trade that Wade and the team put in. That is why he said we already sold. See letter from March, where he says that they sold about $200 million of shares. It confused me as well, because i thought he was saying he un-winded the TRS trade, but he also said he added to it, which made no sense. Looking back i think he was talking about the Wade trade where he says they already sold some.

"After the March/April crash in the stock market, we could not resist buying Exxon shares at a dividend yield of 10.5%, Canadian banks at an average yield of 6.1% and some other companies like Royal Dutch Shell, Alphabet, FedEx and Helmerich & Payne at very attractive prices. We sold approximately half of them in 2020 for a profit of $212 million or an average gain of 40% on our investment."

I thought he was talking about TRS positions on stocks other than Fairfax, and he was saying that while the Fairfax ones are long term, the others weren’t and have been partly unwound. 

[petec]

8 hours ago, TwoCitiesCapital said:

Typically they swaps will "reset" on a monthly or quarterly basis. But they're not like options - there is no upfront premium so a dollar gain in the stock = a dollar gain in the swap (less financing costs).

So if you buy $1B of swaps that go up by 50%, the seller of the swap would pay you $500 million and you'd pay them the financing (LIBOR+spread) for the period. In reality, the payments are netted and flow one direction. 

Then, the swap "resets". The notional is now $1.5B and all P&L and financing will be paid on this $1.5B figure now.  The LIBOR+spread rate resets to the new observation point and P&L and collateral obligations are reset for the next period. 

Now, it's NOT clear to me if Fairfax is discussing these things at cost or not. The move from $1.4B to $2B - is that a difference in cost implying Fairfax added $600 million in additional swaps? Or is it reflective of the 30+% rally we've seen in USD terms meaning Fairfax only added $150-200 million on new swaps and the old swaps reset for $1.4B to $1.8B? This is what I would have to look more into to know for sure. 

Understood. So we can’t tell much from the data we have, in short. 

[Xerxes]

i believe you are right vis a vis TRS vs. commons.

The whole statement confused me.

[SharperDingaan]

At the operating level the swaps are producing a quarterly MTM settlement - with the market resetting every quater. Thereafter FFH receives $ if the share price goes up, pays $ if the share price goes down. At the stategic level, the swaps have locked in the purchase price on the notional quantity of FFH shares - for the life of the swap. All else equal when the swap matures, FFH buys in the underlying number of notional shares and cancels them.

It implies that over the life of the swap, FFH both expects their share price to rise, and that the eventual buy-in will be funded from cummulative free cash flow. Very elegant, and smart approach.

SD

 

 

Edited May 3, 2021 by SharperDingaan

[Xerxes]

49 minutes ago, SharperDingaan said:

At the operating level the swaps are producing a quarterly MTM settlement - with the market resetting every quater. Thereafter FFH receives $ if the share price goes up, pays $ if the share price goes down. At the stategic level, the swaps have locked in the purchase price on the notional quanity of FFH shares - for the life of the swap. All else equal when the swap matures, FFH buys in the underlying number of notional shares and cancels them.

It implies that over the life of the swap, FFH both expects their share price to rise, and that the eventual buy-in will be funded from cummulative free cash flow. Very elegant, and smart approach.

SD,

i dont think that last statement is correct, just based on my understanding of a TRS.

FFH can of course buy whatever quantity it wants of its own shares at the end, but i dont think it will be as part of TRS agreement or within that sandbox. Of course if FFH does a buyback at the end funded by TRS gains, than it would akin to doing a buyback in Q4 2020/Q1 2021 when share price was much lower. 

[SharperDingaan]

1 hour ago, Xerxes said:

SD,

i dont think that last statement is correct, just based on my understanding of a TRS.

FFH can of course buy whatever quantity it wants of its own shares at the end, but i dont think it will be as part of TRS agreement or within that sandbox. Of course if FFH does a buyback at the end funded by TRS gains, than it would akin to doing a buyback in Q4 2020/Q1 2021 when share price was much lower. 

Our bad. We meant a TRS agreement that expires as normal upon maturity. As well as an independent buyback of the notional shares, on the date the TRS agreement expires. The independent buyback most likely being the expiry date of a warehouse arrangement between FFH and some of the market brokers/banks. 

Net of the cumulative mark, the buyback is at the very low Q4 2020/Q1 2021 price, there is an extended period of warehouse buying support throughout the life of the TRS (assumed), and no sudden price spike because of the buyback. 

Very elegant, and very FFH.

 

SD