StubbleJumper

Yes, Daphne, we should try to be responsible. How do you feel about FFH's decision to suddenly invest $50m of shareholders' money with Lissom and specifically under Ben's management? As a shareholder, does this give you comfort? Or would you be more comforted by having it invested by Hamblin-Watsa like all of the rest of the portfolio? SJ

To be clear, Ben Watsa does not appear to actually work at FFH or in any of the subs. Rather, Prem was on the board of directors of a large Canadian charity for a number of years. One other fellow who was also on that board of directors happens to run an investment firm in Toronto. Coincidentally, a few years later, it appears as if Ben Watsa was hired by that fellow's investment firm. Now, that's not that unusual, in and of itself; most of the jobs we get are due to relationships that we've built over the years, so if Ben impressed the boss of that investment firm when he and Prem encountered the fellow socially, that would not at all be unusual. What is unusual is that the Q3 report discloses that FFH has now shifted $50m to this fellow's investment firm, and the specific person managing that $50m is Ben Watsa. So, why was that done? You've already got plenty of internal investment management capacity at Hamblin-Watsa. You've already got Brian Bradstreet in-house. Why do you suddenly need to out-source the management of $50m? And, if you do need to outsource the management of $50m, why specifically does it go to Ben Watsa's firm? After all, there's 100s of investment firms in Toronto that have a far longer track record than Ben Watsa, to say nothing of FFH's long standing relationship with Mason Hawkins and Francis Chou (it's pretty obvious to me that Mason Hawkins has forgotten more about investments than most of us will ever know). So, it doesn't look good. The related-party transaction was disclosed in the filings, but the disclosure was paltry. How much are shareholders paying Ben Watsa to manage our money? Is it 100 bps, 200 bps, some other amount? Is this a real transaction done for the demonstrable benefit of shareholders, or is this a case where Prem is having shareholders pay 200 bps on $50m to ensure that his son will have a well-paid job? Who the hell knows? The one thing that I do know is that it erodes Prem's reputation when he does wacky things like this. SJ

Ahhh, so Albertans should absorb the financial risk of a long-term rail car lease with the goal of absolving the private sector of said risk. I have a better idea. How about governments act in citizens' interests in building pipelines? How about governments spend one dollar on favourable propaganda for every dollar that foreigners spend on anti-pipeline fear mongering? How about governments outlaw the use of foreign money for anti-pipeline crusades and when foreigners spend money to influence Canadian public policy, why don't we make them inadmissable to Canada for life? Why don't we slap a prohibitive tax on foreign contributions to Canadian environmental lobby groups? All of that strikes me as much better aligned with Canadian interests than having government absorb financial risk that should fall on industry. Cheers SJ

How would provincially owned railcars help? It is your viewpoint that GATX is not leasing out enough cars? Look, the reality is that Alberta doesn't have a deep-water port and needs to use either a pipeline or rail to move product to salt water. A pipeline is far more cost efficient, but certain governments are shamelessly pandering to foreign-funded environmental lobby movements. See how the US is going ape-shit over the Russians interfering in their most recent election? Well, that's the kind of bullshit that goes on in Canada, with foreigners spending their money to buy Canadian public policy decisions. It's shameful. The leftists howl about Canadian sovereignty and then they get in bed with moneyed foreign interests. SJ

Value of the culture? Zero. What's the value of a culture that screws minority owners of ORH? Seriously, Prem screwed us on taking it private. What's the value of a culture that usurps minority owners' voting rights? It took two votes to make it happen, presumably with some serious arm twisting of institutional holders. What's the value of a culture where the chairman uses his multiple voting shares to appoint his son, who is still wet behind the ears, to the board of directors? What a joke. What's the value of a culture where the chairman channels $50m of shareholders' money to a firm that employs his son so that his son has a portfolio to manage? Seriously? I'd be embarrassed if I needed daddy to do that to ensure my success. The culture is worth ZERO. The collective brains are the only thing that has worth, but beware that those collective brains are not always focused on the interests of minority holders. SJ

Are you suggesting that underwriting (ex. CAT) will be a lot weaker next year? or they'll make more investment losses? I'm guessing that it's general cynicism and ennui with FFH. U/W profit from 2014 to 2017 was $705m, $552m, $576m and -$642m. I don't feel badly at all about pencilling in 2018 Net Written Premiums of $12B at a CR of 95, yielding $600m of U/W profit when they've bounced around that level with lower premiums and lower CR. If anything, I could be chastised for not pencilling in a CR of 90 or 92 to reflect that insurance prices have trended up in response to last year's ridiculous industry level cat losses. With most of the port in cash, t-bills and bonds, there shouldn't be major investment income surprises. I guess if interest rates rise meaningfully, the $9B of bonds could get written down to $8.7 or $8.5B, but that's not such a problem. SJ

The numbers are 100% wrong, there's no doubt. The question is if you eliminate the realised gains, will there be a material EPS? And I will go out on a limb and say yes, in the absence of some unusual series of cats, there will be a material EPS from underwriting, interest/divvies and non insurance. But, as a long time shareholder, I understand your cynicism. SJ To your estimate you need to add a new stream of revenu from FIH : in january they will pay FFH 114M in performance fee. We can hope for something from Fairfax Africa too in the futur Yes, I believe that in the non-GAAP statements, FFH normally lumps investment fees in with the corporate overhead. This time, it looks like corporate overhead, interest and investment fees all lumped together. So interest goes up with the new debt taken on during 2018, but that's partially offset by better investment fees. But, I'd say that particular line of my crappy pro-forma statement is one of the least accurate! SJ

The numbers are 100% wrong, there's no doubt. The question is if you eliminate the realised gains, will there be a material EPS? And I will go out on a limb and say yes, in the absence of some unusual series of cats, there will be a material EPS from underwriting, interest/divvies and non insurance. But, as a long time shareholder, I understand your cynicism. SJ

I heartily disagree. 1) Underwriting - if you run-rate Q4 you get $11B net written. Tack on price growth from a firming market, you are at $12B net written. Given past performance, I'd say they can write the $12B at a 95 CR, which gives $600m UW profit. 2) Interest and divvies - FFH has $20B of cash/T-bills/etc. Interest rates have risen. If they can average 2% during 2018, that's $400m. They have $9B in bonds, which might average 2.5%, so that's another $225m. And then $5B in common stocks might average 1%, so that's another $50m. Call it ~$600m for the sake of conservatism (knowing that it's likely more like $700m)? So, let's rack it up: UW profit: $600 Interest and divvies: $600 Run-off: -$250 Non-insurance profit: $200 Corporate O/H & interest expense: -$300 Net gains: 0 Pre-tax income: $850m So, in my book, there's $20+/share of after-tax earnings from operations. The realised gains are gravy. SJ

And, the stock price is essentially flat at this morning's bell. Interesting. SJ

I'm not sure that it's a simple question of being negative. The adverse development is definitely a concern. Does it end in Q4 or do we see more in the future? Was it caused by bad luck, or is it a symptom of bad underwriting (under pricing and poor adjustment practices)? Is it something that fixes itself, or can FFH find somebody to go in and fix it? Those are legitimate questions. Allied might be a decent deal, but the first few quarters are harkening back to the massive reserve adjustments of TIG and C&F. But, this time FFH doesn't have the SwissRe cover, right? The good news is that TIG and C&F were long-tail which is where the worst adverse development tends to appear. But, I'd say the jury is still out (this is true of all acquisitions -- it takes a bit of time to determine whether you bought a lemon or a turd). SJ

It's doubtful that I will buy more. I'm at 22% FFH, which is already far too high. The shares look cheap, but Prem has done so many wacky things over the past 5 years that it's a tough thing to take the plunge to go to 30% or 35% (I've been there in the past, but the recent wackiness is a real hurdle for me). SJ

Well, the numbers are nice, but it does bring to mind all of the cautions about lumpiness of results. A few random observations: 1) Net written spiked nicely in Q4 and was up overall in 2017. Given FFH's capital levels and underwriting capacity, this is a good sign of things to come. If you are brave enough to pencil in a combined ratio of 95 and $500m in interest and divvies, it provides the basis of a decent and predictable operating income. 2) Nice to see reserve releases, especially for ORH. WTF is going with Allied's adverse development, and can FFH fix it? 3) The bond and equities ports are running pretty lean, which is an excellent position in the context of rising rates. Shitloads of cash, t-bills and short term bonds is a nice place to be. Can we be brave enough to pencil in 2% for the cash equivalents for 2018, or is that dreaming in technicolour? 4) Holdco cash is high, which is no surpise after some of the asset sales. But what's the deal with expanding the revolver to $2b? So the holdco now has ~$4.0B-4.5B of immediate available cash, when taking into account the expanded LOC. Why? Presumably there's a standby fee for the revolver, so doubling it is a bit strange when FFH is already swimming in cash. I understand that you want to acquire your credit when you don't actually need it, but this one seems strange. 5) Unrealised gains on equities were considerable. How much of this has been given back in the past week or 10 days? Some of the large legacy equity positions have finally moved the right direction, but has the market turned before FFH had the opportunity to exit? Overall, I like the current position with very little of the port in equities and 5+ year bonds. If the cash actually starts to yield something in 2018 and 2019 (as I questioned above, dare we dream 2%) and net written expands with the continued prospect of writing a CR of 95, there's a solid prospect of sizeable operating income. And today it traded pretty much bang-on with the adjusted book. Hard to believe. SJ

Well, BRK had $109B of cash equivalents on its books the last time I checked. I'm guessing that about $70b of that could easily be used for acquisitions or repurchases without placing the insurance companies at risk during a large cat. It's a shit-tonne of cash, and there's more rolling in every day. SJ

Only that that's about the dumbest reason for buying Berkshire that I've ever heard. Ouch. That's a pretty direct response for a new board member.

Yes, it's always necessary to add in those amounts to the BV calc, and I guess now we ought to subtract US$10 for the divvy. I did an adjusted BV calc a couple of months ago, but I don't have it handy. My memory was that it was less than US$500 at the time, and now it would be $10 less because of the dividend. But, in any case, we are not far from BV today. SJ

Come off it! This selloff hasn't created the kinds of values that would make them go all in. Unless it's in the 2y treasury for a yield pickup over cash with little duration risk. They'll keep leveraging their position as a preferred provider of capital but the fact that the S&P is back where it was 6 weeks ago isn't going to tempt them. Agreed. We seem to have collectively forgotten what constitutes a sea-change in the markets. That's what FFH is waiting for. As you said, the bump in short term interest rates won't hurt them any, but I don't see many bargains yet in equities. The one exception to that might soon be FFH's own shares. Another few days of this fun, and we might be bouncing around BV, which IMO, would be a decent place to initiate a large repurchase. SJ If you add in the unmarked gains I think it will get you to book value or less. By unmarked gains, you mean investments like BB that have gone up, but haven't been marked to market? I haven't paid any attention over the past few days, but I wonder whether all of those unmarked gains still exist. Pretty much everybody's gotten a haircut over the past week, but I haven't specifically looked at FFH's major holdings. SJ

Come off it! This selloff hasn't created the kinds of values that would make them go all in. Unless it's in the 2y treasury for a yield pickup over cash with little duration risk. They'll keep leveraging their position as a preferred provider of capital but the fact that the S&P is back where it was 6 weeks ago isn't going to tempt them. Agreed. We seem to have collectively forgotten what constitutes a sea-change in the markets. That's what FFH is waiting for. As you said, the bump in short term interest rates won't hurt them any, but I don't see many bargains yet in equities. The one exception to that might soon be FFH's own shares. Another few days of this fun, and we might be bouncing around BV, which IMO, would be a decent place to initiate a large repurchase. SJ Yes but even there they can only invest the holdco cash, not the float. Some time over the next few months they'd need to dividend up a pile of money to the parent. Last I checked, there was plenty of cash and plenty of statutory capital in the subs, so they could dividend a healthy amount to the holdco without really affecting underwriting capacity. But, my guess is that they sit tight and wait for the next shoe(s) to drop. SJ

Come off it! This selloff hasn't created the kinds of values that would make them go all in. Unless it's in the 2y treasury for a yield pickup over cash with little duration risk. They'll keep leveraging their position as a preferred provider of capital but the fact that the S&P is back where it was 6 weeks ago isn't going to tempt them. Agreed. We seem to have collectively forgotten what constitutes a sea-change in the markets. That's what FFH is waiting for. As you said, the bump in short term interest rates won't hurt them any, but I don't see many bargains yet in equities. The one exception to that might soon be FFH's own shares. Another few days of this fun, and we might be bouncing around BV, which IMO, would be a decent place to initiate a large repurchase. SJ

Yes, but if I've understood correctly, the Cdn ops had rev of ~$1B. Then I've gone the next perilous step of inventing an EBITDA margin and slapping some hypothetical multiples on it, and finally comparing it to the ~$750m of Carillon's Canadian debt. The only purpose of that exercise was to guess at the amount of cash going out the door from FFH. So, can we assume that FFH guarantees the $750m of debt and throws maybe $250m of cash at the parent? SJ

Hmmm. Well, If it has ~$1b in revenues, would that be ~$200m in EBITDA? If gross margins are ~20% then FFH can't really pay much more than the $750m debt that Carillon's Canadian operations owe. I mean, you might be able to do 5X or 6X ebitda (ie, $1B or $1.2B), but 10X would be outrageous? SJ

FFH hasn't attached a dollar-value to this acquisition: http://www.fairfax.ca/news/press-releases/press-release-details/2018/Fairfax-to-Acquire-Certain-Canadian-Assets-of-Carillion/default.aspx Does anybody have any idea of the magnitude? SJ

rb, I am a little curious how you know this. Up until two months ago, there were historically essentially zero capacity related IT failures amongst the major Canadian brokers. We seemed to have experienced a few days of wildly excessive demand which resulted in some of their platforms failing while others appear to be business as usual. Do you have some sort of industry IT knowledge that one broker built a good or proper system and one simply cobbled together some spare hardware and slapped together a bit of code? That situation is quite clearly a possibility, but without actually having worked in one of the brokerages, it's not clear to me how one would obtain that knowledge. I ask this because I am genuinely curious about why some brokers failed to deliver the goods while it was business as usual for others. In December, it seems clear that some of the brokers fell short on capacity -- did they have 80% or 90% of the capacity needed to manage the spike in pot stock trading, or did they have 60% or 70% of the required capacity for that demand spike? IB seems to have had at least 100% of the needed capacity, but was there still some spare capacity? Could IB have handled 50% more volume, would the platform take double the volume in December? Can we instead attribute IB's success to the fact that they might not have a legion of 20 year-olds who vigorously trade pot stocks because most of their clients are more serious, experienced investors? I am quite curious about all of this because the next generalized market event could quite possibly involve considerably higher volume than what we saw in December. I don't generally panic and sell when the market goes down, but it would be nice to at least be able sell.... SJ I did work for a couple of the banks (not for the online divisions though) and I know that they were woefully under invested in their online brokerages. They're kind of a joke. The feeling is that whatever it is it's good enough for the retail investors. If sometimes they don't work that's just life. I also think that you're thinking about this capacity issue the wrong way. I'm not a tech guy (so don;t take me as an authority) but I know a bit about it. You can generally process huge amounts of traffic/data at data center level with relatively little equipment if it's well configured. But IB and the banks brokers are fundamentally different in how they run. With IB you use TWS which is stand alone on your computer and communicates with IBs servers using data streams over certain communications protocols. This is a (ralatively) simple and robust solution. Going about it this way IB may very well have the capacity to do 10 or 20 times their normal volume because that extra capacity would not be very expensive. The banks on the other hand use a web application to deliver the service. These apps are probably not very good to start - none of the bank tech is. It's also a much more complicated solution than what IB uses. Put it under stress and it craps out. You can't know for sure what happened unless you were there but if I were to bet, I would bet that it was user application that crapped out and not their trading engine in the back because it ran out of capacity. Oh and regarding a market crisis, they're guaranteed to not work. They weren't working on volatile days in 2015 when we had that pullback and those were not that bad historically speaking. Thanks for the insight. It's very helpful. It's entirely possible that the banks' web-server was a weak link, but the back end might have been fine. That makes perfect sense that a stand-alone app would work better than a web-based platform in that circumstance. I've never been too impressed by the brokers' websites, but from a user perspective it's all a black box, so it's helpful to get insight from somebody who has at least gone for a beer with people who work on that stuff. SJ

Now that's an interesting possibility that I hadn't considered. IB might have a single mega data centre located in, say, New Jersey, which services all of their clients around the world. I had always assumed that they would use de-centralized data centres, but I suppose that there's no real reason from a regulatory perspective to use a Canadian data centre for Canadian clients and a US data centre for US clients, etc. It's always good to challenge our assumptions! SJ

rb, I am a little curious how you know this. Up until two months ago, there were historically essentially zero capacity related IT failures amongst the major Canadian brokers. We seemed to have experienced a few days of wildly excessive demand which resulted in some of their platforms failing while others appear to be business as usual. Do you have some sort of industry IT knowledge that one broker built a good or proper system and one simply cobbled together some spare hardware and slapped together a bit of code? That situation is quite clearly a possibility, but without actually having worked in one of the brokerages, it's not clear to me how one would obtain that knowledge. I ask this because I am genuinely curious about why some brokers failed to deliver the goods while it was business as usual for others. In December, it seems clear that some of the brokers fell short on capacity -- did they have 80% or 90% of the capacity needed to manage the spike in pot stock trading, or did they have 60% or 70% of the required capacity for that demand spike? IB seems to have had at least 100% of the needed capacity, but was there still some spare capacity? Could IB have handled 50% more volume, would the platform take double the volume in December? Can we instead attribute IB's success to the fact that they might not have a legion of 20 year-olds who vigorously trade pot stocks because most of their clients are more serious, experienced investors? I am quite curious about all of this because the next generalized market event could quite possibly involve considerably higher volume than what we saw in December. I don't generally panic and sell when the market goes down, but it would be nice to at least be able sell.... SJ