petec

No, it won’t happen soon. I’m just speculating that the buyback might have become part of the reason for the discount, and that might persist. These things are weirdly self-perpetuating.

How do you see one set of shareholders benefiting and the others not? Is there a way dividends can be directed for example?

Yes, but IIRC FIH wasn’t buying back stock then, or at least not in size. Buybacks, even at a discount, become a *problem* for minority shareholders if they increase the risk that a controller can take the company private at a big discount. I’ve learnt this the hard way in the past and that is what might be starting to happen here.

Why would a second Dutch auction at FIH be a catalyst when the first one wasn’t? If the discount is due to (il)liquidity and FFH’s rising stake, wouldn’t a Dutch auction make things worse?

I thought the two-column approach was float per share plus *non* insurance earnings?

So carrying value $195m vs value of $470m according to Chou (calculated as (1200-115)*.43). Could be a nice little write-up one day.

Remind me how much FFH own and what they carry it at?

What’s the loan secured against?

Not if you started with excess capital, which is effectively what Prem has said repeatedly in his letters with the “in the last hard market we wrote at 1.5x” claims.

If that is the case then we should not see any further capital injections, and might start to see dividends. That would be positive.

Surely your income on your cash and short bonds responds almost immediately to the Fed raising rates? If you also get paid to increase duration, that's a double win, but it is not the only one. Clearly the discount rate rises so the impact on p/bv from rates rising on short term investments should be negligible, but I am not entirely sure the market works that way. I think it may well capitalise "predictable" interest income at a higher rate than lumpy underwriting income or very lumpy investment income, so as interest income rises in absolute terms the p/bv may rise. I am speculating.

One thing that has been nagging at me for a couple of years - and I know it has been discussed here before, but I'm not sure we ever got to a good answer - is why does FFH keep putting capital into the subs when according to the annual report the subs have capacity (in aggregate) to grow underwriting 1.5x or more on the existing capital base? Anyone got a clear view on this? It's potentially very important, because if the subs can grow without needing capital - or even, while dividending it to the holdco - that really changes the cash flow dynamics at the holdco and the possibilities for buying back shares or stakes.

Yes, but in a couple of years' time we will be able to take the convert down even lower

Agreed but then the eps growth forecast was above what I’d assumed possible. Overall I’ll take it.

I half agree and half don't. Counterpoints below: When you have debt you have no flexibility around when to repay. When you have an option, you do. Flexibility has absolutely increased. Without knowing the full terms of the deal we simply cannot know what the consequences of failing to "repay" would be. So long as OMERS et al are happy with their returns they may have no issue with Fairfax not "repaying" the loans. In fact they may see repayment as a threat, given the difficulty of redeploying capital. Certainly Fairfax's "failure" to "repay" the Allied "loan" does not seem to have stopped them attracting similar capital so far. Repurchasing shares in FFH and repurchasing positions in Brit/Allied/Odyssey amount to virtually the same thing - both ways, we end up owning more of what we already own. The deals effectively create liquidity on tap if FFH want to do another big buyback. Again, this arguably increases flexibility, given that liquidity may well be a barrier to buybacks for the mothership. The problem is, without knowing the terms of the deal and specifically the buyback price-setting mechanism, we cannot assess the value they might get. As I have argued before, having an in-house lender of last resort increases flexibility, rather than hindering it. I know you are not arguing counter to this but it's a point worth making.

Does the interest accrue? I thought it took the form of preferential dividend rights, and was paid in cash. It doesn't affect Fairfax's value, so long as the deal allows them to buy it back at a fair price. What it might affect, as SJ says, is their flexibility to use cash for other things. Absolutely.

I am sure it is not, but Prem has explicitly named the head of OMERS as a partner in the past and I infer that there *is* a personal relationship at the top.

I think the keys to the OMERS-type deals is the buyback provisions. Does FFH have a right to buy back at a fixed price, or a fixed multiple of book, or what? Only when we know this can we assess whether the deals were good. What I would say is that: 1) These deals are not quasi-debt. I know very well what SJ is getting at here, and it is a good point to raise, but debt HAS to be repaid and therefore represents a risk that does not apply here. 2) We cannot assume that the fair value of Odyssey is 1.67x book because if the minorities have a dividend rights and can be bought out for pre-agreed terms, then they do not own common equity and the value may be different. 3) When FFH's insurance subsidiaries are marked up to something approximating fair value, we have to be more careful comparing FFH's p/bv with other insurers. If all assets were marked in this way the company should trade at 1x BV, while a company built entirely organically, which effectively carries all its assets at depreciated cost, might trade well above book value.

It's funny, because I was quite positive on all these when they happened, because I thought they looked like good value (i.e. classic Fairfax investments). I also made the same arguments you are making about Atlas & Stelco management and Stelco's balance sheet (but not Atlas' - that's only solid because they issued a shit-ton of equity after Fairfax invested). But what has really made these two into exceptional investments is the cycle. Period. OK. Maybe I am just misreading you. I've had the sense that you are arguing that there has been a radical change. when I think it just looks like a radical change because the cycle has turned. However if you're just arguing there have been incremental changes in emphasis within a deep value framework, then we agree, and I think I have argued the same before, especially around Atlas and Stelco.

I'm not sure it helped the NPL management - trends there were positive before the merger and have not really changed (except for the fact that a lot of NPLs have been divested, which had nothing to do with the merger). What I did leave out though was the addition of a significant fee stream, which is something a normal equity raise would not have provided. But it was an equity raise, and that was the main benefit, and my guess is that the reason they did it as a merger rather than a raise was to lessen optical dilution to Fairfax. Call me a cynic.

I am going to be blunt here, but please do not take it as rude. Practically all of this is a figment if your imagination. That doesn't necessarily make it wrong, but unless you have access to internal discussions at Fairfax that I don't, you're sitting at home inferring process changes from the outputs, and coming to one relatively unlikely conclusion when the evidence equally supports other more likely ones. None of that is a problem - it all adds to the rich diversity of views on this board - but I call it out because it increases the chances that you will be disappointed down the line. I have no doubt Prem is always learning from his successes and failures, like the rest of us. But the idea that he has delegated capital allocation is delusional, in my humble view.

I agree on both these points. I also wonder if we are approaching the point where the hard market slows, reserves have been rebuilt, and normal course buybacks funded from operating earnings can start in earnest. That would be great with the share price where it is now!

For two reasons: 1) Something did change: they stopped shorting. This is clear and has been communicated clearly. 2) Their investments are in a much better position because the cycle changed. But what we are seeing now is the fruits of going bargain hunting at the bottom of the cycle. It has taken years, but the fruits are here. We are not seeing the results of a sudden change in management philosophy. If we were, I think they would have told us so.

Every single one of these has to do with a change in the relationship between the price and the intrinsic value of the security concerned (or, in the case of the KW private credit investment, the alternatives). Not one of these has to do with a change in strategy or approach. Fairfax goes where they see value. They did not buy back shares in FIH when it traded at or above book. They started to when it traded at two-thirds of book. This was not because they decided to invest internally.

Out of interest, why do you think 1+1=2.5 in this case? I don't see that the economics of any of the underlying assets changed as a result of the merger - Grivalia's properties did not rise in value, Eurobanks NPE's did not suddenly spring back to life, and there were no significant cost reductions. What did happen is that Eurobank's capital increased, because it issued equity. But that would have happened if it had issued equity for cash, rather than for Grivalia shares. Am I wrong?