petec

I hate to say it but I told you so! The buyback is a multiyear event funded by FCF; the cash on hand is for the OMERS stakes in Brit, Eurolife, and Allied. On the positive side the free cash flow from Toys R Us alone is enough to buy back 35bps of the stock each year ;)

I thought the RE might be a key part of it. Am I right that it is debt free? Because if so they've basically paid 3x ebitda, or 1x after RE, which suggests kinda limited downside even if Amazon does come along and eat their lunch. Not so sure about Bill Gregson though - isn't he kinda busy? ;)

Why does the investing side detract from the management attention that can be given to the insurance side? The guy who can improve results at Allied is Andy Barnard; he *is* focussed on that, not on Toys R Us.

Agreed - although for completeness I have to point out there will be losses too on e.g. Blackberry. EDIT - I take that back - BB not down much ytd, it's just fallen from a spike in 1q - but equally, Eurobank not up much, it just recovered from a dip in 1q.

$238mil is the gain on FFH's direct holding in IIFL. Through Fairfax India's holding they will recognize an additional $212mil. D'oh. Thanks!

Sorry if I’m being dim but why does recognising a $238m gain drive a $450m gain?

Yeah not a call I'd make with any confidence. I have more in Fairfax than Berkshire on the basis of deeper knowledge (mine), smaller size (Fairfax's), younger managers (Fairfax's), and the possibly dubious assumption that there are fatter gains to be made in value stocks than in operating businesses (which are an ever-bigger part of Berkshire) but I expect Berkshire to compound nicely too.

After about a year, can you report back on whether you've made a hole in your jeans? I smell false economy ;)

It makes a lot of sense and I’ve been thinking the same. Unfortunately under MIFID2 in Europe ALL securities that pay out based on a reference value are termed PRIIPS and PRIIPS can only be sold in Europe if they come with a Key Information Document or KID. KIDs are 3 pages long and are so summarised as to be actively unhelpful but apparently we are all too stupid to read the real underlying documents. Anyway, obviously only European issuers give a fuck about it this and issue KIDs so the unintended consequence is Europeans can’t buy floating rate bonds or prefs issued outside Europe. (Or maybe it was an intended consequence - capital controls anyone?) And this comes in just when rates are rising and the only way to protect yourself is ... a floating rate bond! And while I’m on a roll, the final straw is that I’m British, and we are leaving the EU, so why are we implementing the damned regulation anyway? In case you can’t tell I’m bloody annoyed. I’ve emailed Dundee and asked them to consider preparing a KID. They are pretty basic and can refer to a prospectus so it would take about five minutes. Got no reply. (Atlantic Power replied and said no.) Grrr.

Yes! Although there is a charge on the early redemption I believe, so it's not free money.

I just don't think the CR tells you enough to make that call. Consider 3 companies: 1) a specialist company that can write at 70% - but the niche is small and growth is limited. 2) a company at the more commoditised end of the market that loses all its customers writing at 95%. 3) a company at the more commoditised end of the market, writing at 100% when competitors with less cost discipline are writing at 101%, growing huge by slowly taking share, and making a ton of money investing the float. Which is more valuable? And is the CR alone a good measure of whether the company has its costs under control? Also, all else being equal (which it never is), long tail float written at 100% is more valuable than short tail float written at 99%, in the same way that a long term loan that costs you nothing is more valuable than a short term loan that costs you less than nothing. My point is that the CR is as much about strategy as it is about costs. It’s not a sufficient tool (although it is a necessary tool) for judging management or efficiency or value creation. Its component parts - the loss ratio and the efficiency ratio - would be more helpful but even then you need to compare them to other insurance companies in the same segment, not to the other companies in Fairfax. For all the CR on its own tells you, these could be the two most efficient companies in their industries. Take Crum for instance. Crum has transformed itself over the last decade. The mix of business is transformed and it’s stemmed what were huge underwriting losses. It generates a lot of float and IIRC it is one of the biggest payers of dividends to Fairfax. Looking at the headline CR and inferring that it is inefficient or badly managed just doesn't jive with me. I can't prove that you're wrong, but I can say with some confidence that the CR alone is not sufficient proof that you're right. My understanding (correct me if I am wrong - I know less about Berkshire than Fairfax) is that Ajit has power over the Berkshire insurance subs but that they have not been consolidated into one company. Andy Barnard is the same - he already has huge practical influence over all the subsidiaries as I understand it and he has done a lot of work to remove duplicated costs and spread best practise, both of which can be done without consolidating companies legally. He also has all the info needed to make a decision like this. I am fairly sure that if he thought it was a good idea it would get done. Maybe it will be but I won't be surprised or disappointed if it isn't.

There's a case to be made that the US is structurally overvalued due to its use as the world's reserve/trading currency. To the extent that other currencies win a bigger share of global trade, demand for the dollar will be lower (you won't have to own dollars before you, for example, buy oil). More significantly, if you can trade in other currencies you can risk holding some of your reserves in other currencies - so the Chinese could sell treasuries. All of that pushes demand for the dollar down and, all else equal, the gold price (in dollars) would then rise.

See above. They paid a higher price for the bloc that enabled them to remake the board and name the CEO. So that has inflated YE17 average cost. I'm not sure what the MV is off the top of my head - it's not listed.

It's in the Fairfax India 2018 letter. The second bloc was at a higher price because it enabled them to remake the board, name the CEO, and "generally to allow it to be managed according to Fairfax India's standards of corporate governance and guiding principles". The first bloc was at 8.7x FCF excluding the land bank so even the second bloc, at almost twice the price, was inexpensive.

Agreed, but I was struck by the tone in the FAH letter - there seemed to be a lot of "we" in the discussions of what Atlas are doing, rather than "they". That could just be down to their way of writing or it could imply that there is a lot of involvement.

Very interesting. I would love to know whether Fairfax have had much impact on how this is managed. The FAH letter implies they are deeply involved.

Yes I guess so. The market seems to pay a premium for what it knows. So, it might pay 1.1x bvps because it knows there is a steady buyback driving bvps growth, but it won't pay 1.1x bvps for cash on the balance sheet that could be used for a buyback at a very attractive price. That's my sense anyway.

By that logic, why vote for Christine N. McLean? She is Prem's daughter. Because I didn’t know that this is the case, otherwise I would have withheld the vote for her as well. McLean's relationship was laid out clearly in the shareholder letter. I don't mean this aggressively, but I'm interested: why would you vote at all if you haven't researched the people?

Agreed. I'm fairly sanguine. I agree they make sense. But given that they make sense, and given that FFH are clearly thinking about them, if they are not doing them yet then they clearly have what they think is a better use for the cash. That works for me. Also, incidentally, if you're looking for a rerating then a sustained buyback with FCF is better than cheap one-offs. The market, in its madness, does love consistency.

That Eurolife profit includes the Greek bond gains I think - not likely a sustainable rate nor one you’d put 9-12x on, but wonderful nonetheless! I understand the valuation argument. It just seems to me there’s a danger of getting excited about a big buyback because we want it to happen rather than because the company shows any signs of doing it-which they don’t.

That would be a gigantic error. They have a lot of talented people beyond Andy Barnard and the reason they stay is because they can run their own businesses under Fairfax’s decentralised system. Centralise, and not only do you lose them all but you send a signal that no-one who wants to keep building their business should ever sell to them again. Plus, the insurance business are not a homogenous mass that can be run together. They’re a massively diverse group of niches that need focussed management. Fairfax and Andy Barnard’s influence is already felt across the group - for example in the changes they plan to make regarding cat tolerance at AWH. He couldn’t do a lot more, but the downside would be enormous.

Dazel: Eurolife is an exceptional asset (as you seem to have realised in your second post) and they bought it for nothing. It would be crazy to sell it - I imagine there would be a big tax hit and the economy is on the verge of reflation. Also, the won't get a better price on Brit or AWH due to last year's cats. As I explained at length in my earlier posts, the price they pay to OMERS is based on OMERS' cost plus a hurdle, not on the current book value of the subsidiary. They don't have to exercise the option, but if they do that's the price they will pay. If anything the one they will get a great deal on is Eurolife, if the deal structure is the same. And they don't have the option to buy AWH until 2020 I don't think. SJ: Eurolife does not trade on 3x. Dazel is saying it trades on 3x at FFH's cost, but I am not sure where he got that from because Eurolife is not listed (last I checked it was 40/40/20 FFH/OMERS/Eurobank). What I do know is that Eurolife was loaded with Greek government bonds when they bought it which promptly went up a lot so it has been an absolute home run. I don't really understand the obsession with buybacks here. Sure the price is nice but the stock isn't liquid enough to do anything meaningful, and they have other good uses for the cash buying in minorities. They've clearly communicated that they will use the cash to buy in the minorities and that they will buy back with free cash flow over time, which is what the liquidity of the stock will bear.

I get to 600m using $1.88bn * 27.5% * 1.08^3 - $46m $1.88 * 27.5% is OMERS' cost for remaining shares 1.08^3 is 8% annual hurdle for 3 years since deal $46m is the dividend paid in 2017 (there may have been more in 2016/8). I'm nowhere near 100% sure on the methodology but we are in the same ballpark. Agreed re dividend.

Ha - I am an FFH holder not an FIH one! So I agree. But it's a good point. Problem is consolidating it all it is a bigger job than it seems - for example you could argue their 49% in Quantum Advisors should go in to FIH too, but that's unlisted so there is no easy way to price it. Probably just too complex so I suspect they keep it as is, although I wonder if they will take Thomas Cook private once the Quess spin is done.

I don't know the exact terms but it's something like this: Brit was bought in 2015 and OMERS got 29.9% of it. Brit doesn't have to pay a dividend but if it does OMERS gets preference up to an annual hurdle. From 3 years onwards FFH have the option to buy the OMERS stub for cost plus an annual hurdle rate. I think the hurdle rate is 7-8% and I assume any dividend paid counts towards it. (Clearly depending on how book value performs relative to the hurdle rate, the total P/BV paid for Brit will change.) I believe AWH works on a similar system so I assume Eurolife does too. Anyway the point is that the Brit option goes live in 2018, Eurolife probably in 2019, and AWH probably in 2020, and I expect all 3 to be exercised.