petec

If you include dividends Fairfax is now in the black on Stelco. Something of a record - only took two years ;)

https://ca.finance.yahoo.com/news/atlas-mara-limited-announces-strategic-155000677.html The transaction will include upfront cash consideration payable at closing equal to approximately 0.8 times book value as of 30 June 2020, plus additional cash consideration payable 24 months after closing of the transaction, subject to certain conditions. https://www.benzinga.com/pressreleases/20/11/ac18543273/atlas-mara-limited-announces-strategic-transaction-with-kcb-group-plc As part of the Transaction, KCB will acquire Atlas Mara's 62.06% shareholding in Banque Populaire du Rwanda Plc ("BPR") for cash consideration representing approximately 1.09 times book value and, through the Company's subsidiary ABC Holdings Limited, all of Atlas Mara's indirect interests in African Banking Corporation Tanzania Limited ("BancABC Tanzania") for cash consideration representing approximately 0.42 times book value. The actual cash consideration payable by KCB will be determined based on the final book value of the two banks at completion of the transactions. The transactions are expected to close during the first half of 2021, assuming regulatory approvals are received by then. https://www.bnnbloomberg.ca/nigeria-s-biggest-bank-in-talks-to-buy-atlas-mara-assets-1.1530545 Do not be surprised if this happens close to BV as well which leaves them with UBN and following are the latest results for the same https://thenationonlineng.net/union-bank-grosses-n118-8b-in-q3/ Chief Financial Officer, Union Bank of Nigeria (UBN) Plc, Joe Mbulu noted that the bank’s asset quality has continued to improve with non-performing loans (NPLs) down to 3.6 per cent from 5.8 per cent as at December 2019, supported by ongoing efforts to diversify loan book to include viable businesses and households. “Our Capital Adequacy Ratio remains robust at 19.5 per cent, well above the regulatory threshold. With the $40 million financing secured from the International Finance Corporation for on-lending to trade finance customers, we are continuing to expand our funding engagements with DFIs to support our strategic business initiatives. Again not surprised with zero transparency from FFH and FAH mgmt on why ATMA was sold to FFH for 0.25*BV except saying helios did not want it . FAH is leaving close to 100 mil on the table by doing this deal with Helios which brings only 5 mil to bottom line for now on annual basis. If Helios was so keen on partnering with FFH , they could have done this deal with Helios and kept ATMA at the same time . They can sell it for 40 mil whenever they want or much more if they market it properly. And moreover FFH is only paying 20mil upfront. What a joke. I do not know why helios did not want it , may be they got the jitters after the deal with Equity group fell through. The only explanation is that FFH benefits at the expense of FAH shareholders and no one seems to care since there is no 'smart' money in FAH to seek an explanation from FFH. The Access/Mozambique and KCB transactions are certainly positive, although small. It would be a very positive sign if Access buys the rest, and I think you're right that that could go at around or above 1x bv since the major asset there is ABC Botswana which has a stub trading at 1.3x book on the local exchange. A control stake would presumably go for more. The only thing you neglect to mention wrt the valuation of ATMA is that UBN trades for 0.6x bv and is a melting ice cube in real terms (ROE<inflation) in a market where the downside risk is meaningful (e.g. Central Bank sets a floor for the loan/deposit ratio, which has the potential to be very scary). The benchmark asset in this space is Guaranty Trust Bank, which has nearly 3x UBN's ROE. I think UBN is probably a very valuable asset in the long term but I am not totally sure it is there yet. (Idle speculation: I wonder whether FFH's UBN stake might end up in CIB.) But even allowing for that, it is clear that ATMA trades (and the FFH deal was done) below SOTP value. I don't dispute that, and in fact I think I have made that case at various points upthread. I've certainly pointed out that if ATMA ever gets >50% of UBN (not sure why that is taking so long) the optics of its financial statements will change overnight. I just don't think the undervaluation is relevant in the way you do. To my mind there are three key points: 1. Helios don't want ATMA. They don't see the value. (That alone tells you something, given their record. In fact, they saw so much risk that they wanted FFH to guarantee FAH's loan to ATMA.) This leaves Fairfax with a choice: a) Walk away from the deal. b) Put the deal at risk by delaying it until ATMA's SOTP value is realised either via a sale or a breakup. There is absolutely no evidence that either of these can be done fast. Sale? There are no obvious strategic buyers for the whole. You might find a value buyer, but they'd want a low price and they'd need to do extensive due diligence. This is 1H20, and a global pandemic is getting started. This makes future value highly uncertain and makes getting on a plane to do due diligence impossible. Breakup? Given enough months you can probably find strategic buyers for each of the subsidiaries, but selling subsidiaries requires regulatory approval which takes many more months. Fairfax have been trying to break ATMA up for 2 years now and know very well how hard it is. Conclusion? Ask any major asset manager: to sell fast you have to sell cheap. To get a good price for ATMA is going to take 1-3 years. Helios might be long gone by then. c) Buy ATMA from FAH and get the deal done. 2. FFH cannot pay much more than market for ATMA. Even ignoring the fact that a pandemic is just getting started which could erode significant book value, the fact is that FFH board's key responsibility is to FFH shareholders. ATMA trades on a recognised stock exchange. There is a market price at which FFH can buy if they want to. Accepted, the stock is not liquid, so perhaps a block premium is justified. If so, it is nearer 20% than 100%. If you were complaining that the price should have been $0.45, or $0.50, I wouldn't be arguing with you. But you think FAH left $100m on the table. If FFH had paid that full of a price, I guarantee you that this board would be full of the Prem haters saying that he'd used the FFH balance sheet to save face by bailing out FAH to get a deal done, and couldn't be trusted. And they'd be right. 3. The Helios deal is transformational and may well create more future value for FAH minorities than has been "lost" on the ATMA trade. Last week FAH was a failed holding company, with no track record of creating value and a number of very challenged subsidiaries. Next week, when the deal closes, it will become a more diversified, more cash generative, better run company with the ability to leverage 3rd party capital to grow in an asset-light manner. Unfortunately there is no amount of disclosure that will help you figure out how much value the deal creates, because what you need to know is how much 3P capital the new company will raise and the IRR on its future deals. Only time will tell. Bottom line is that I just don't think FFH had many choices here, and I think they deserve some credit for getting a potentially superb deal over the line.

So, by the time this transaction is finalized, OMERS will have held their stake in Riverstone for one year and will once again have gotten their ~9% annualized return? The same annualized return that they got from the Brit transaction? Is that a coincidence or was it engineered? SJ Neither. A) we don’t know the conditions governing the contingent payout. B) we don’t know if OMERS sold on the same terms, better, or worse.

Scratch that. See below. They're the same subs. Looks to me like Fairfax (60%) have made a nice gain on this sale ($750m vs a carrying value of $605m, with a possible $237m to come). This would imply OMERS (40%) got $500m plus $158m contingent, which means they may well have recorded a loss - unless of course they got better terms than FFH, which I suspect is likely. From the 3q interim: On March 31, 2020 the company contributed its wholly owned European run-off group ("European Run-off") to RiverStone (Barbados) Ltd. (“RiverStone Barbados”), a newly created joint venture entity, for cash proceeds of $599.5 and a 60.0% equity interest in RiverStone Barbados with a fair value of $605.0. OMERS, the pension plan for municipal employees in the province of Ontario, jointly manages RiverStone Barbados and had contemporaneously subscribed for a 40.0% equity interest for cash consideration of $599.5, based on the fair value of European Run-off at December 31, 2019 pursuant to the subscription agreement entered into on December 20, 2019. I am quite happy to see it go, frankly. I don't think the market gives runoff the credit it deserves and capitalising the cash flows up front to fund hard market growth in subs the market does understand is OK with me. i'm eager to learn but don't come to the same numbers. At end 2019, book value of the sub was 750.5m. Ultimately, with measurement adjustments after but based on yr-end, Omers paid 599.5m for a 40% interest, suggesting a 1.5b value for the unit without a control premium. This implies a 895m value for FFH which, i think, was written down to FV of 605m in correlation to negative results in Q1, including investment losses. i understand FFH will receive proceeds of 750+235.7=985.7m for their 60% deconsolidated equity interest so (assuming Omers gets the same terms) Omers should be getting 500+157.1=657.1m. One attractive aspect of FFH over the years (my humble perception) is (was?) that market recognition of value in the short term was not really relevant given a long term orientation. Even if felt to be obsolete, i have difficulty understanding why one should long for immediate market recognition when one of the principal lever of value creation would be share repurchases. Aren’t our numbers exactly the same? I guess my point is simply that a management team has, at any given moment, to assess the options in front of it. It’s not obviously daft (to my eyes) to sell European runoff at >BV and reinvest the proceeds in the other subs at BV at the start of a hard market that might see them valued at >BV, or in FFH shares at <BV.

Given that what's rallying is the stuff they've been long, and not the stuff they've been vocal is overvalued, I am hopeful.

I am not surprised. FFH often spend years (decades) building up great set of businesses, but when it comes to monitzation it gets offset by a large short position gone wrong in a single quarter. I am guessing market will give full recognition on the gains in Q4 when results come out in January and netted against the realized shorts. With FFH shorts, we never know what we get: the good, the bad or the ugly. Every dollar on the Blackberry share price above $6 is worth $100m to Fairfax, before tax. Fairfax’s market cap is $10bn. How much do you think the Fairfax share price *should* have moved, out of interest?

Scratch that. See below. They're the same subs. Looks to me like Fairfax (60%) have made a nice gain on this sale ($750m vs a carrying value of $605m, with a possible $237m to come). This would imply OMERS (40%) got $500m plus $158m contingent, which means they may well have recorded a loss - unless of course they got better terms than FFH, which I suspect is likely. From the 3q interim: On March 31, 2020 the company contributed its wholly owned European run-off group ("European Run-off") to RiverStone (Barbados) Ltd. (“RiverStone Barbados”), a newly created joint venture entity, for cash proceeds of $599.5 and a 60.0% equity interest in RiverStone Barbados with a fair value of $605.0. OMERS, the pension plan for municipal employees in the province of Ontario, jointly manages RiverStone Barbados and had contemporaneously subscribed for a 40.0% equity interest for cash consideration of $599.5, based on the fair value of European Run-off at December 31, 2019 pursuant to the subscription agreement entered into on December 20, 2019. I am quite happy to see it go, frankly. I don't think the market gives runoff the credit it deserves and capitalising the cash flows up front to fund hard market growth in subs the market does understand is OK with me.

The unit OMERS invested in last year was Riverstone UK. CVC is buying Riverstone Europe. They have the same CEO but are they the same? Is UK a subsidiary of Europe?

Thanks, Pete. So the total position is still deeply under water, to say nothing of the time value of money. Oh well, at least it's moving in the right direction. SJ I would strongly suggest never including the time value of money in your assessment of Fairfax's investments, unless you have a stiff drink to hand. Atlas may be an exception.

I vaguely recall that the average cost is in the $17's. I think they disclosed this a few years ago. If you convert the bond at $6 you get something like $12.

This post has brightened my afternoon.

Especially when you rewrote your convert.

That’s quite sizeable!

I can’t help but think he should have focussed on running the company rather than writing this book.

Shhhhh! Have you learned nothing?! ;)

More directly on the book value points: 1) How do you derive the market value of Bangalore? 2) The only bit of your reasoning I disagree with is Digit. If 10% sells for X, it is reasonable to think that a control stake might sell for >X. The world is paying amazing multiples for tech growth these days.

Thanks. Out of interest what do you think Atlas has done to/with APR since purchase? As far as I can see they have - cut costs. - deployed assets on the short term Mexicali contract, but I think this was in place since they bought it. I asked the same question of Parsad on the Atlas thread and he replied with a long extract from the q2 call. But apart from the cost cutting, everything they discussed on that call was either in place before they bought APR or was rather vague conjecture about future plans. Assuming they have found genuine fat, cutting costs is very positive and frees up capital for reinvestment, so credit for that. But apart from that I don't think they have done anything with APR yet.

Factual question: was accounting BV much different from fair value BV in the Q3 reporting? I seem to remember being surprised at how similar they were but maybe I misread. Philosophical question: as rates fall the yield on the bond portfolio falls. This reduces Fairfax’s intrinsic value, all else equal. But in an efficient market (ha ha) the earnings yield at which Fairfax’s stock is priced should also fall. Mathematically, do these two effects offset each other perfectly or imperfectly?

Is this a prediction or have you seen some news?

Yes. My thought is these buys largely scream value, not defence.

My apologies for being bone idle, but would you mind putting a number on the free cash flow yields you see (and the underlying gold price assumption)? That way I can judge whether to do the work!

It wasn’t a big buy. But it would definitely have been good if they’d had the cash to add this year when it fell.

The $1bn number is 3y free cash flow. They invested $670 of that and paid $330 in dividends/buybacks. Going forward they have basically no more major capex after q4. So free cash flow will come to shareholders. The absolute key difference, it seems to me, between Stelco and the “shitty” things it was compared to (Resolute) is liabilities. Stelco has few (although be careful with the pension and the fact it is carried at a discounted value). That massively limits the downside risk. And the land provides a very nice upside option. It’s a much better judged investment than Resolute imho. EDIT: those are the figures I think they gave on the call. I want to check them.

And so you should! By way of explanation, I think it will ooze cash in upcycles, not lose much in downcycles, and allocate capital well. I think through-the-cycle cash flows will allow for a 10 (good) to 15% (great) return on the price paid, with very low risk because there isn't much debt; and I think the real estate is a nice option on the side. My view is that Prem paid a very low multiple of peak free cash flow and a reasonable multiple of average cash flow. The reason the price collapsed was that the cycle collapsed just afterwards. That does not make this a bad long term investment necessarily, but it does make it look badly timed. Then again, it wouldn't have happened any other way. Kestenbaum is not stupid and wouldn't have sold at the bottom of the cycle.

I’m going to stick my head above the parapet and say that Stelco is going to be a good-to-great investment for Fairfax over the next 3-5-10 years, even allowing for their too-high going-in price.