petec

FWIW, I think this is the wrong way to look at things. What matters is total resources available, not where the resources are. The subsidiaries have significant dividend capacity - and that capacity would grow rapidly if they decided not to grow premiums written. In extremis, a subsidiary can even be sold. So, if the holdco needs cash it can have it. But that would be a choice, and the cost would be growing slower or investing less. It took me a while to grasp this, but I think the structure here is smart and an advantage. Same for Brookfield, where I ended up thinking the way the leverage was designed in BPY was an active advantage, not a weakness. I have come to actively seek businesses that are levered in such a way that the leverage cannot impair the franchise in a short term crisis.

I certainly think they've thought hard about what they do and are arguably a bit better at it. But I think there's a danger in assuming that that is what has led to better outcomes so far. The biggest driver of improved performance by far has been the uplift in all commodity prices (from interest rates to shipping rates to gas and steel and timber). For example Eurobank was always a good franchise run by a good (perhaps great) team. It's probably a better company than Keells and certainly a simpler one. But the only reason it was so cheap is that it was in the wrong country. I think we should expect Fairfax to continue to make investments like that that take years - even decades - to pay off. Similarly Digit is no different, conceptually, to ICICI Lombard. Seaspan and Stelco are different to Resolute. They're better run and better capitalised. That's great but a) there's been a lot of luck in the timing of the cycle turn and b) don't forget that the only reason Seaspan is better capitalised is that it issued an absolute boat-load (pun intended) of stock after Fairfax invested. Recipe is the same company it always was, just at a lower price. I'd be a lot more confident in your thesis if Fairfax was actively selling failed historic investments, like Blackberry and Farmer's Edge. But as far as I can recall, despite all the talk of "monetisation", they have only sold one: Resolute. None of this is a criticism. I have held Fairfax since 2008, in part because they swim where others see only sharks. Sometimes they get bitten. I am fine with that. Today, it's my biggest holding, because I think they are going the right way and very cheap. I am just cautioning against reading too much into what is primarily a cyclical turn. I believe they are more focussed on investing, having taken their eye off the ball. I think they are better at doing what they have always done. But I do not believe the leopard has changed its spots.

No, but higher prices, surely? No way. Just the shipping across the Pacific would be similar to cash drilling costs in Alberta, and that’s before drilling in Russia, piping to China, cooling, regasifying, and piping to Alberta. What am I missing?

Inflation is the only thing that drives currency diversion over the long term. So the only factors that might cause divergence are a) mismeasurement of inflation or b) cyclicality. What I mean by the latter is that currencies can diverge from their “correct” inflation adjusted value for long periods of time depending on economic performance, perceived risk, etc. These divergences can last decades. Looking at where a currency sits relative to its long term inflation adjusted average is a good way of assessing whether it is cheap.

So is it fair to say you don’t buy into the Peyto argument that NGTL expansion + intra-basin demand + LNG Canada = Alberta egress > production in the next few years? Presumably needs a change of government?

Thanks. What’s your view on whether more capacity (after LNG Canada phase 1) gets built on the West Coast to serve Asia?

Do you mean NGL or LNG? I thought the debate was about LNG but you mention both?

DM me if you want a chat - I ran a Latam fund for years!

I’m not entirely sure I agree with that, on the assumption that if Fairfax doesn’t sell and the buybacks continue eventually Fairfax owns 100%, but I’d have to do a worked example to be sure. Anyway, I think there’s only one minority in each of the FFH subsidiaries?

To an FFH shareholder there’s no difference assuming the price paid is the same.

Agreed - it’s interesting that they think buying Allied is better value than buying FFH, which isn’t at all obvious to me. I wish we knew more about the economics of the minority stakes.

I know what you’re saying but it’s a slightly false distinction. Buying in the minorities at Allied, Brit, and Odyssey effectively *is* a huge buyback, even if it doesn’t bring the share count down. You still own more of those business per share, and they’re the beating heart of Fairfax.

This is an interesting mental exercise but I’m not sure Fairfax think about their investing in these buckets. In fact, I think you could simplify the whole thing dramatically. In my view they aim for two things when they invest: value and great people. They do not aim for certainty of outcome, unlike some. They’re perfectly happy to lose money on some, and make spectacular gains on others. (And it shows!)

Might it not be as simple as the market is starting to position for a recession and a slower pace of rate increases, while insurance is somewhat procyclical and Fairfax in particular is a beneficiary of rising rates?

I’m not sure whether it’s likely but I think it’s prudent to stress test for it. Rates have been low for nearly 15 years now without causing a hard market. That’s at least partly because managers, shareholders, and external capital accepted a lower than double digit roe (precisely because rates were low). It’s also partly because falling rates drove a bond bull market and BV gains. I think it’s the change in direction in rates, as much as the level of rates, that’s driven the insurance market hardening. If rates continue to rise that driver remains in place, but I think it would be conservative to assume they level off at a historically low level (say 3 or 3.5%) and then look at what level of CR is needed for an industry ROE of say 8%. Note: these numbers are not predictions just suggestions for a bear case normalized CR modelling exercise.

Once* the SIB is done FFH will own C$145m/year worth of normalised FCF at Stelco, plus a one-third share of any excess cash flow it generates before (if) steel markets return to historic norms. They won’t control this cash flow (yet!), and it will be lumpy, but Prem shows every sign of valuing the lumpy dollar as highly as the smooth one. I could be dead wrong but I don’t see him selling into the SIB. *I should have said if, but if the SIB doesn’t complete Stelco has a massive cash pile so the value remains.

Viking, have you run a “normalised” scenario assuming long run underwriting at say 98% or 99%?

1. Question is, how much of that embedded optionality do you capitalise up front. To my mind it's fair to (and the market does) capitalise some of it in competitively advantaged businesses with cookie-cutter formulae for allocating capital, such as CSU. However its much harder to justify capitalising capital allocation up front when you have no idea what the capital will be spent on, and the market generally doesn't capitalise it up front in those situations, even when the jockey is Warren Buffett. Given that Atlas/Sokol's attempts to allocate away from shipping have been lacklustre so far, I wouldn't expect a buyer to capitalise that optionality up front. 2. If I'm right, then this is a fair offer and does not reinforce Fairfax's reputation in the way you suggest. Interesting to see how others see this - thanks all for the debate as usual.

As someone who runs a restaurant, I am always a little sceptical of these claims. Get it right - right offer, right brand, right location - and it is a very cash generative business. The problem is that few restaurants fit that mould, so 90% fail - but that's not a reason to write off the other 10%. I think the risk is reinvesting in it - which is why I agree I hope they use it as a cash cow, and if they do grow it I hope they do it on someone else's balance sheet by franchising it.

Agreed. And if they're paying 7.5x future FCF it will be a serious growth engine. Recipe have always talked about international growth. I have always idly wondered if any of Recipe's assets could be franchised to Fairfax India.

Sorry to hear it! But my word, we all have stories like that. It is the nature of the game!

Yes!

Because buybacks are funded out of equity and takeouts are funded out of float, is my guess.

Except for the TRS

Again, why is this t*tf*ckery? 1) Why is 7x 2-year-forward and possibly peak earnings a lowball offer for a levered lessor in a commoditised industry? Nobody has explained this to me yet. 2) If Fairfax didn't own a share of Atlas, but offered this price, and shareholders could accept or refuse by minority vote, would we be angry? I think our (on both sides) reactions are more reflective of the pre-existing positions of each commentator than of anything else.