petec

I'd forgotten about this. Quite astonishing really. They bought a good business at a cheap price plus free levered upside optionality in Greek government bonds when they were trading cheap but the underlying problems were being resolved, and they did it partly with cheap funding from OMERS. An extraordinary deal.

A very minor and pedantic point for which I apologise in advance. Cyprus is an island. Cypress is a tree!

Yes. For me, they've gone from "we are long term investors and are happy to wait for our 15% return" to "we are long term investors but we still work hard to ensure that our 15% return happens in a reasonable timeframe, and that we can realise it into book value so that we can see it".

I think this is one of the biggest changes from the bad old days - they've realised that IV has to be reflected in BV or the market won't pay for it. However, for all the reasons we have discussed, they also have a lot more value to realise these days than they did. It's much easier to monetise investments when they've gone up!

Or dividends. This might be well be the main motive for doing the deal. Nice bit of reg arb.

Possibly, but which way? I think protecting downside because you're not being paid to take the risk of a possible macro outcome is a micro/pricing bet not a macro one. But I am also starting to think the argument I (re)started is rather pointless

This is also my take. Mostly.

This is correct in terms of short term reporting. As discussed above, I don't think it is correct when actually looking at whether claims are sufficiently covered, taking into account what usually drives interest rates - i.e., inflation.

Absolutely.

Yes of course. Sorry - I should have been clearer. I am not suggesting this example is/was reflective of FFH's position. I just find extreme examples useful for thinking through principles. The point is you can seriously impair your balance sheet if you take duration risk without being paid for it. This is your strongest argument, imho.

Agreed. But the logical output of that argument is that FFH (and other insurecos) always take a macro view when they invest in nominal fixed income securities. Not that they're taking a macro view when duration spreads don't pay them to go long.

I think this is highly dependent on interest rates (which impact float income and combined ratios). At 1% rates, I doubt that FFH will make a 15% ROE. At 5% rates, I would be surprised if they do not.

I would agree if the bonds were inflation-linked so that both inflation and duration were matched. But that isn't practical - the reality is that the bond portfolio will always be mostly nominal while the insurance liabilities they back are inflation-linked. If I have a 10-year bond backing a 10-year insurance liability, and inflation unexpectedly goes to 10%, by the end of 10 years I am going to have a worthless asset and a huge liability - even though I matched my duration. At a time of historically low rates and low inflation, therefore, I don't think it is really a macro bet to say: I don't know what's going to happen or when, but there is a risk that inflation turns out higher than expected and my balance sheet gets impaired, and I am not being paid to take that risk. I think that decision was about price, not macro predictions. If long bond yields had been 15%, all else equal, they'd have gone as long as possible!

Valid point about comp, and that may be why McMorrow doesn't want to do an MBO and give someone else the keys! However KW is selling stabilised assets to improve the balance sheet. Perhaps not at the pace you'd like, but it's been a difficult market. Obviously the market tends to respond immediately to rate increases while inflationary rent increases feed through slowly. I think we are past the nadir of that cycle and it will be much easier to sell assets near full value in the next few years than it was in the last few. Also, KW embarked on a lot of new construction just before covid - appalling timing in retrospect but those assets have been completing and stabilising over the last couple of years. For both those reasons I expect the pace of sales to quicken. Full disclosure, I built a small stake in KW recently.

Viking, I think this proves my point as much as it does yours. Of the 10 investments you review Fairfax has got rid of two (APR and Mosaic). One is a star performer today (Eurobank). Two more are performing reasonably well (Recipe, AGT). Two more are significant but rapidly becoming smaller as the portfolio grows (EXCO, CIB). Two may yet come good (Helios and Astarta, which I too know little about). Have recent investments produced much higher IRRs than the ones made in 2014-17? Yes. Is that because Fairfax have completely revamped their approach? I doubt it. I am sure they have refined it, but it is not transformed. I think there is a degree of randomness here. They've had great patches before, and are in one now. They've had soft patches before, and I suspect they'll have them in the future too.

+1

I think you understand it very well! It's pure capital cycle economics. When capital is abundant and losses are low, rates come down. When capital is less abundant and losses high, rates go up. This is why I think rates drive *both* float income and combined ratios, which has had an outsize impact on FFH's earnings over the last few years.

+1

I am not sure what you're apologising for but you don't need to!

The only comment here I disagree with is the one about hard insurance markets. Yes of course there are niches which behave differently, but insurance does definitely go through hard and soft cycles. I have no idea when the next soft market is coming, but it is coming. Most of the rest I agree with, which is why Fairfax remains a big holding for me.

I still don't see that as a macro bet. Saying: "we think interest rates will rise so we will keep duration low" is a macro bet. Saying: "the interest we would earn if rates stay low does not compensate us for the duration loss we'd incur if rates rise" is actually macro-agnostic value investing. That's how I have always interpreted Perm's "reach for yield" wording. Not that it matters.

I am really intrigued that FFH hasn't done an MBO with KW. It would be an interesting investment skillset to have in-house and I think there is value at this price.

Yes, when stabilised. And KW has paid some extra for the development pipeline and the team.

$347m deal with KW funding ~$90m and partners the rest. This is equity, and KW's share of the complete or near-complete assets drops from 37% to 5-10% as part of the deal. In other words what the partners are buying is 27-32% of the equity in these assets for ~$257m. The partners get a better deal on these assets because KW is also getting the development pipeline and team. If there is a fixed income opportunity it is likely later as the completed assets get refinanced, and as the development pipeline needs to be financed.

Likewise, Viking, and for the record I think your work is superb. However I don't think your post here is quite accurate. I don't think anyone thought the P&C business was bad - in fact, there was clear evidence that Fairfax were very good at improving insurance assets over time. The issue was that insurance was in a loooooong soft market. In addition, the fixed income float was earning nothing and the market didn't value Fairfax's refusal to reach for yield. In addition, the portfolio was viewed as low-quality crap: Eurobank wasn't earning anything at the NIM level let alone after bad debts, Resolute and Blackberry were clearly mistakes, management kept pouring money into stuff the market perceived to be low quality, like Stelco and Seaspan and Dexterra. And the hedges - dear God, the hedges... What changed? The vast majority of what changed was that the earnings power of the assets improved for cyclical reasons beyond management's control. insurance went into a hard market, allowing for higher CR's and much more float. interest rates rose, transforming float earnings and Eurobank's NIM. brief post-covid commodity price spikes allowed Resolute and Stelco to delever, buy back shares, and sell themselves. Did management change their approach? Somewhat. The hedges are the best example, and I think they realised you can't just sit in value names and wait - you have to actively work to realise that value. But are they really doing anything different? Not much. Have they changed their underwriting philosophy? Not a jot. Are they still value investors? Absolutely. Would they buy 2011 Eurobank or 2019 Stelco again? Absolutely. I think this matters becase I think it is important not to capitalise cyclical earnings too far into the future. That doesn't mean I think earnings will collapse back to 2020 levels, because I don't think rates will go back to 0.