petec

Makes sense!

I don’t disagree, but I’d look at premiums per share.

+1. The “look through” buyback is immense. Just need ATCO to join in!

Bear in mind they don’t mark it to market, so it probably won’t come through in bvps, but in adjusted terms I think the lower end of that bound is possible. And I think dividends could be significant on a 2-3y view.

The real value is if the recover drives negative net NPE formation and provision costs aren’t as high as 1%.

The press release did specify that it was a new class of security, so I think Safety is right that it’s a new class with special features. What those are, I couldn’t say, but history suggests a preferred dividend and a preset repurchase price will be in there.

Imagine a Berkshire with a Fairfax multiple and you’ll have a much better investment

I’m not sure the rate is high when you consider the buyers seem to be taking equity-like risk.

It does seem odd.

Well technically at this valuation they could sell all of Odyssey and buy back 70% of Fairfax, so I’m not sure this transaction is evidence that they’re tapping out their capacity - in fact the opposite. More pertinently, we are currently in a hard market, growing premia at 24%, and absorbing capital to do so. It’s reasonable to assume that at some point the market will soften and the insurance subsidiaries will generate capital. So it’s reasonable to assume that buyback capacity will increase in the future. I don’t see any logic behind saying that the TRS has to be applied to *this* buyback, or it’s a speculation.

Yes, I remember that exchange. For me the key points are 1) the “preferred dividend” is paid by Brit, not Fairfax. 2) Fairfax doesn’t have to buy back the shares. 3) The price Fairfax pays doesn’t seem to change. This makes me think Omers got a preferred return capped at a fairly high level (10%), but took equity risk to do so.

I don’t really understand why the TRS is not a speculation if it is used to lock in the price for this buyback, but is a speculation if it is used to lock in the price for another buyback in the future.

Why is this the buyback rather than a buyback? Why would it make sense to close out the TRS before the stock reaches (whatever Prem thinks is) fair value? Assuming fair value is above the $425-500 buyback range - and we know Prem thinks that it is - I think it makes sense to do the current buyback and hold the TRS to lock in a buyback at $373* later, when premium growth has stopped and the insurecos can dividend cash to the holdco. *IIRC this is the initiation price of the TRS.

I haven't looked at Brit's reporting but presumably there is a book value and a share count, so can't we calculate the BVPS? That might illuminate whether OMERS had a sweet deal or not? My sense is that OMERS got 10% preferential dividend, paid from Brit, but that's where their upside was capped - whereas their downside had no floor because FFH didn't have to buy the shares back. It looks like a fairly priced deal to me, but we don't really have the details to know.

What happened to Brit’s BVPS over this time? My vague sense is that it was flattish but I could be wrong. But that’s a key input.

I see FFH now owns 39.3% of RFP. If I was them I think I’d encourage RFP to buy back shares until I had control, and then put 100% of free cash to dividends.

Why use forward BV, not current?

No they don’t. There’s plenty of evidence that buybacks are pro-, not counter-, cyclical. Some management teams are good at them but most just do them when they have excess cash, not when the share price offers an opportunity, so they’re not a reliable signal of undervaluation. And if the shares aren’t undervalued, they don’t increase IV per share.

I don’t disagree, but Fairfax is not at this point. Maybe Berkshire is.

Sure, but why does it matter to you as a shareholder whether you own the same number of higher value shares, or a higher number of shares with the same value? It’s totally irrelevant (except for tax). EDIT: sorry, just realized this point has been made.

Exactly. And for the reasons Sanjeev gave, I think the dividend speaks powerfully to the culture they’ve built and to how they look after their people. I know the culture isn’t perfect but I think their thinking around the dividend is logical, rational, clear, and I think the long term impact on the culture and people is more important than the long term impact of the tax lost compared to a buyback (which is really the only reason a rational outside shareholder would rather a buyback than a dividend).

Imagine Fairfax owns 51% of a company. They will consolidate this company, since they control it. 100% of its assets and liabilities therefore go on the FFH balance sheet. However, they only own 51% of the equity. 49% is owned by other investors who do not have control. These noncontrolling interests need to be subtracted from FFH's equity, or FFH's equity would be overstated.

Thanks. I didn't follow the Putnam deal that closely but this makes sense.

Being really dull...can we shift the discussion about quarterly results to the Farifax 2021 thread? I find it much easier when each topic is concentrated in one place and I can't be the only one.

This is exactly right but if they didn't break it out, how would you know how to calculate a normalised CAT number? You wouldn't know how much of the long term average was normal and how much was CAT, and I think that helps you judge the risks. Said another way, if they didn't disclose it you can bet your bottom dollar that this board would be full of posts about how they should disclose more to help us understand the underwriting.