Fairfax 2020

[KFS]

A couple of notes from the conference call:
- Annual report and Q4 info will be available March 6
- net written premiums grew at accelerating rate each quarter; sees trend continuing in 2020
- supporting growth of insurance subs top priority; this is because hard markets occur infrequently and only last a short period of time (a couple of years)
- share repurchases will be done with cash not needed to grow insurance subs
- in the non-insurance segment, TCook/Quess demerger resulted in $191 million impairment
- US runoff had $216 million loss due to asbestos strengthening; example of social inflation
- $600 million will be coming to FFH in Q1 when the Riverstone UK deal closes; after sale closes Riverstone UK will have opportunity (with OMERS) to grow business. Fairfax may take Riverstone public at some point.
- minority insurance partners buyout: 10% of Brit soon at $100 million (seems low?); Eurolife will be small amount; Allied agreement opens up mid year and Fairfax has 3-4 years to buy out minority partner
- Go Digit: the company recently raised 10% at $800 million valuation. The $300 million gain recognized by Fairfax was Go Digit convertible shares (owned via Quess) not the common stock owned by Fairfax (hope i got this right)
- BIAL sale was to third party investor; purchaser was not named

Some additional notes:
- Repurchased 479k shares in 2019.  1.2 million since 2017.  Repurchases are expected to be significant but over a long term (i.e. 10 years) as stated in previous annual report, not necessarily in short term or every quarter.  (Singleton repurchased 85% of shares outstanding over a 10-15 year stretch.) 
- Coronavirus:  minimal or insignificant impact on insurance. 
- Stock market overall valuations are high, but... for example, from 1999 to 2002 the overall market dropped 50% while Fairfax value-focused investments increased 100% over the same period. 
- Fairfax india clearly undervalued.  Great potential for BIAL, 3rd largest city in India, community of software engineers; re-election of Modi. 

[petec]

Just started reading the transcript. “Realised gains in Seaspan and Brookfield” - sounds like they’ve sold some SSW and owner BAM? Unless there’s been a transcription error...

[StubbleJumper]

Just started reading the transcript. “Realised gains in Seaspan and Brookfield” - sounds like they’ve sold some SSW and owner BAM? Unless there’s been a transcription error...

Perhaps gains on SSW have something to do with the APR merger.  That's the sort of thing that I was alluding to in my comments from yesterday about quality of earnings.  There have been a few transactions over the past couple of years which have triggered accounting gains without any real change to the economic reality.  Keg and Recipe are an example.  Eurolife and Grivalia is another.  SSW and APR might be a third example, if that transaction has already hit the books? 

It's almost worthwhile inventing a pro-forma income statement to strip out some of this.


SJ

[petec]

There wasn’t a gain on APR so that’s not it and I don’t think the transaction will contribute earnings, high quality or low.

I suspect what they’re referring to is gains on the exercise of SSW warrants.

[StubbleJumper]

There wasn’t a gain on APR so that’s not it and I don’t think the transaction will contribute earnings, high quality or low.

I suspect what they’re referring to is gains on the exercise of SSW warrants.

That could very well be the case.  They would definitely have a gain from exercising the SSW warrants, but it doesn't much change the economic reality (outside of FFH having to add some cash to the existing investment).

Keg/Recipe wasn't the correct transaction for me to reference.  In 2018 the "accounting" transactions were Grivalia being consolidated, and Thomas Cook/Quess triggering a bunch of paper gains.  In 2019, it was Quess triggering a bunch of paper losses and Eurolife/Grivalia triggering a bunch of paper gains. 

For the past few years, most of us have cooked up an adjusted BV to get an idea of the significance of some of the excess in value over book.  But, really it might be time to create an adjusted net income to strip out the impact of some of the one-time transactions to better portray ongoing economic performance.


SJ

[petec]

While we’re at it I don’t recall Grivalia/Eurobank generating a paper gain either, and it’s clearly been value-enhancing because it was effectively a capital raise for the bank.

[StubbleJumper]

While we’re at it I don’t recall Grivalia/Eurobank generating a paper gain either, and it’s clearly been value-enhancing because it was effectively a capital raise for the bank.

"Net gains on long equity exposures of $1,631.1 million in 2019 was primarily comprised of unrealized appreciation of preferred shares of Go Digit Infoworks ($350.9 million), the sale of the company's remaining interest in ICICI Lombard ($240.0 million), a non-cash gain on the merger of Grivalia Properties into Eurobank ($171.3 million) and significant unrealized appreciation of common stocks."

[petec]

There wasn’t a gain on APR so that’s not it and I don’t think the transaction will contribute earnings, high quality or low.

I suspect what they’re referring to is gains on the exercise of SSW warrants.

That could very well be the case.  They would definitely have a gain from exercising the SSW warrants, but it doesn't much change the economic reality (outside of FFH having to add some cash to the existing investment).

Keg/Recipe wasn't the correct transaction for me to reference.  In 2018 the "accounting" transactions were Grivalia being consolidated, and Thomas Cook/Quess triggering a bunch of paper gains.  In 2019, it was Quess triggering a bunch of paper losses and Eurolife/Grivalia triggering a bunch of paper gains. 

For the past few years, most of us have cooked up an adjusted BV to get an idea of the significance of some of the excess in value over book.  But, really it might be time to create an adjusted net income to strip out the impact of some of the one-time transactions to better portray ongoing economic performance.


SJ

Ah I see what you’re getting at.

Personally I care more about BV than earnings so I’m happy with the disclosure they’ve often provided on what BV would be if they marked to market.

[StubbleJumper]

There wasn’t a gain on APR so that’s not it and I don’t think the transaction will contribute earnings, high quality or low.

I suspect what they’re referring to is gains on the exercise of SSW warrants.

That could very well be the case.  They would definitely have a gain from exercising the SSW warrants, but it doesn't much change the economic reality (outside of FFH having to add some cash to the existing investment).

Keg/Recipe wasn't the correct transaction for me to reference.  In 2018 the "accounting" transactions were Grivalia being consolidated, and Thomas Cook/Quess triggering a bunch of paper gains.  In 2019, it was Quess triggering a bunch of paper losses and Eurolife/Grivalia triggering a bunch of paper gains. 

For the past few years, most of us have cooked up an adjusted BV to get an idea of the significance of some of the excess in value over book.  But, really it might be time to create an adjusted net income to strip out the impact of some of the one-time transactions to better portray ongoing economic performance.


SJ

Ah I see what you’re getting at.

Personally I care more about BV than earnings so I’m happy with the disclosure they’ve often provided on what BV would be if they marked to market.

Agreed that the BV metric is the more important metric for valuing the insurance end of the operations, and it's been important to cook up an adjusted-BV estimate for the past few years to better reflect reality.  But, it's also important to try to measure operational performance against any number of metrics, including EPS, ROE, CR and investment return.  That's where some of these non-cash items muddy the water.  This year, the dollars are small, with Eurobank/Grivalia being $6 or $7 per share...but the Thomas Cook/Quess number from 2018 was absolutely enormous, and the Grivalia consolidation number from 2018 was a smaller number added to it.  A large head-line EPS number is nice to see and it definitely feels good, but...


SJ


SJ

[petec]

While we’re at it I don’t recall Grivalia/Eurobank generating a paper gain either, and it’s clearly been value-enhancing because it was effectively a capital raise for the bank.

"Net gains on long equity exposures of $1,631.1 million in 2019 was primarily comprised of unrealized appreciation of preferred shares of Go Digit Infoworks ($350.9 million), the sale of the company's remaining interest in ICICI Lombard ($240.0 million), a non-cash gain on the merger of Grivalia Properties into Eurobank ($171.3 million) and significant unrealized appreciation of common stocks."

Sorry - I doing a very poor job of expressing myself (possibly because I’m also trying to feed a 3 month old!).

What I meant to say was:
1) I couldn’t remember whether they booked a gain on the Eurobank/Grivalia deal, but
2) if they did I’m pretty sure it did reflect economic reality in the sense that it moved book value closer to the mark-to-market book value. Until the deal Grivalia was consolidated so the rise in the share price since acquisition wasn’t reflected in FFH BV.

I may be remembering wrong - don’t have my notes to hand.