Six names: the subject, the parent that owns 43% of it and collects its fees, two private assets used to demonstrate the marking argument, one listed Indian peer and one Canadian comparator for the buyback behaviour. Prem Watsa, Benjamin Watsa, Henry Singleton, Narendra Modi and the commentator "Brown" are people, not securities; OMERS (11.5% of Anchorage) and Hamblin Watsa (the wholly-owned manager) are entities within the structure rather than investable names. Research legend: QT Qualtrim · SA Seeking Alpha · STK Stock Analysis. Written post with no timestamps — the At link opens the article.
| Ticker | Name | Research | View | What he said | At |
|---|---|---|---|---|---|
| FIH-U.TO | Fairfax India Holdings ($FIH.U) | QT · SA · STK · FA | Positive | "A very attractive investment opportunity" — and an explicit pass, on fees rather than on merit. A $4bn holding company, 84% in infrastructure and financial services, BVPS compounding $9.35 (2015) → $22.94 (2025), a 10.3% CAGR that is said to understate the truth; debt/equity below 20%. Valuation: shares at ~$18 against a $22.94 book, ~0.9x P/B, falling to ~0.7x if BIAL marks at $4bn — management's own phrase, "a discount on a discount." Capital allocation in the Singleton mould: shares issued at the 2018 peak, then 23.2m shares bought back for $303.9m at an average $13.09 — 14.8% of shares issued since inception. Chairman Benjamin Watsa, Prem's son and designated successor. The verdict: "If Fairfax India didn't have to pay fees to Fairfax, we would probably buy them for the Tiny Titans Portfolio" — i.e. owned indirectly, not directly. | read ↗ |
| FFH.TO | Fairfax Financial Holdings | QT · SA · STK · FA | Positive | The holding, and the intended beneficiary of every argument in the piece. The India exposure is quantified for the first time: an investment portfolio of roughly $75bn with $4.3bn in India — about 5.7% — "we believe this percentage will increase significantly going forward." Fairfax owns 43% of Fairfax India and, more importantly, collects its fees through Hamblin Watsa: a 1.5% management fee (0.5% on cash) plus 20% of any BVPS gain above a 15% three-year hurdle (~4.8% a year) — about 1.8% annually of the portfolio, $572m over eleven years. The same conservative-marking argument is then applied to the parent via Poseidon, and used to defend its own ~1.3x price-to-book: "on first sight, that looks expensive. But dig a bit deeper… and you get the idea that book value underestimates intrinsic value." | read ↗ |
| private | Bangalore International Airport Limited (BIAL) / Anchorage | — | Positive | "The crown jewel" — 55% of Fairfax India's investment portfolio, and the whole valuation case. Traffic +8% to 43.8m passengers, management targeting 80m by 2029; revenue has compounded at 21% a year over six years while the carrying value went from ~$2.6bn (2019) to ~$2.9bn (2025) — 1.8% a year. Carried at 10x EBITDA and 13.8x free cash flow against a peer-transaction average of 21x (Brussels 29x, Sydney 25x, Copenhagen 26x). Structure: Fairfax India owns 30.4% of BIAL directly and 88.5% of Anchorage, which owns another 43.6%; OMERS bought 11.5% of Anchorage for $129m in 2021, implying a $2.6bn equity value for all of BIAL. The catalyst is the Anchorage IPO — "the proportion of Fairfax India's investments subject to public mark-to-market valuations would increase from 31% to 85%." | read ↗ |
| GMRAIRPORT.NS | GMR Airports | STK | Neutral | "The closest public peer for BIAL. They own and operate several Indian Airports." Used purely as the mark: an EV/Sales multiple of 10.1x — higher than BIAL's EV/EBITDA in Fairfax India's books. "Just read that again!" No view is offered on GMR itself, and no argument that its own multiple is justified; it is the yardstick, not a recommendation. | read ↗ |
| private | Poseidon (Fairfax holding) | — | Neutral | The proof that the marking gap is real, and realised at the parent. A private company in which Fairfax held a ~45% stake, carried at $15.50 per share. "In late May, Fairfax sold a ~23% stake for $28.30 per share, more than 85% above its carrying value. The sale generated $1.91 billion in proceeds and a pretax gain of $837 million." Cited as evidence rather than as a holding view — it is now largely sold. | read ↗ |
| TVK.TO | TerraVest Industries | STK · FA | Neutral | A single-sentence comparator for the share-count behaviour: after noting that Fairfax India issued stock at the 2018 peak and repurchased it after the fall — "it all sounds obvious, but we are aware of very few companies that actually do this" — the post adds, "TerraVest is another company that comes to mind that is following a similar strategy." No figures, no valuation, no rating; a name to file rather than a call. | read ↗ |
Three cautions the piece does not raise about its own argument. (1) The 19%-versus-8% comparison is not like-for-like. Realised returns are measured on the subset management chose to sell — which is where the value was proven — while unrealised private returns are measured across everything still held, including whatever has not worked. Conservative marking is one explanation; adverse selection in what gets monetised is another, and the post considers only the first. (2) The fee that disqualifies Fairfax India is paid by Fairfax India's shareholders to a company Compounding Quality owns — which is a reason to prefer the parent, and also a reason to read the enthusiasm for the subsidiary with that in mind. It is disclosed plainly, to the author's credit. (3) The peer table is Fairfax India's own exhibit, from its shareholder letter: a selection of recent transactions chosen by the party arguing its asset is undervalued. The GMR EV/Sales-versus-EV/EBITDA line, likewise, compares two different multiples and is presented as if it were one comparison.
A jargon-free summary of the thesis behind each argued name. (Renders on each name's consolidated page.)
Fairfax India is a listed pot of Indian assets, run from Canada by Prem Watsa's son Benjamin. It owns about $4bn of things, mostly infrastructure and financial businesses, and more than half of that is one asset: Bangalore airport.
The interesting part is an accounting quirk. When a holding company owns shares in a listed business, the value on its balance sheet moves with the share price every day. When it owns a private business, there is no share price, so management estimates the value themselves. Fairfax India's estimates have been deliberately cautious, and there is a clean way to prove it: the returns it has actually banked by selling things are 19% a year, and the returns it reports on the private things it still holds are 8% a year — the same team, the same period. When something is finally sold, the real value shows up all at once. Two thirds of the portfolio is valued this cautious way.
So there are two discounts stacked on each other, which is management's own phrase. The assets are carried below what they are worth, and the shares then trade below even that: book value is $22.94 a share and the shares are around $18. Management has been buying those shares back — 23.2m of them at an average $13.09, nearly 15% of everything ever issued — which is exactly what you want a manager to do when their own stock is cheap.
The event that would close the gap is an initial public offering of Anchorage, the subsidiary that holds most of the airport stake. Listing it puts a market price on the biggest asset and, on management's own figures, would take the share of the portfolio valued at real market prices from 31% to 85%. If the airport is then worth $4bn rather than the $2.9bn carried, the shares are trading at roughly 0.7x book instead of 0.9x. There is no date, and regulatory approval is still pending.
Compounding Quality nevertheless does not buy it, and says why: Fairfax India pays management and performance fees of roughly 1.8% a year — $572m over eleven years — to Fairfax, the parent, which they already own. Owning the parent gives the same India exposure and collects the fee rather than paying it.
This piece is really a second look at the Fairfax purchase made eleven days earlier, filling in the number that was missing then. Fairfax invests about $75bn, of which $4.3bn — about 5.7% — is in India, and the expectation is that this share grows a lot. It also owns 43% of Fairfax India and, more usefully, is paid to manage it: 1.5% of assets a year plus a fifth of any gain above a 15% three-year hurdle.
The same conservative-valuation argument then gets applied to Fairfax itself, with a case that has already been settled in cash. Fairfax held a stake in a private business called Poseidon on its books at $15.50 a share. In late May it sold part of that stake at $28.30 — 85% more — banking $1.91bn and a gain of $837m. That is what it means to say the accounts lag reality.
It matters because Fairfax's shares look expensive on the usual insurance yardstick, about 1.3 times book value. If the book itself understates what the assets are worth — as Poseidon just demonstrated — then 1.3 times an understated number is not the multiple it appears to be. That is the whole valuation defence, and it is a good deal more than the 16 August purchase note offered.
Bangalore is India's technology capital — most big American tech firms have engineering centres there, and about one in four iPhones is now made in the city. Its airport handled 43.8m passengers last year, 8% more than the year before, and management expects roughly 80m by 2029. Airports are natural monopolies: a city gets one, it cannot be duplicated, and it charges everyone who passes through.
Here is the mismatch. The airport's revenue has grown 21% a year for six years. The value Fairfax India puts on it in its accounts has grown 1.8% a year — from about $2.6bn in 2019 to about $2.9bn now. Nothing about the business justifies that; it is simply a cautious internal estimate that has not been marked up.
Comparison makes it starker. Airports that have actually changed hands recently sold at an average of 21 times their annual operating profit — Sydney at 25, Brussels at 29, Copenhagen at 26. Bangalore sits in the books at 10 times. The nearest listed Indian operator, GMR Airports, trades at 10.1 times revenue, which is more than Bangalore is carried at against profit.
The way this gets recognised is a stock market listing of Anchorage, the vehicle holding most of the stake. Once listed, the market sets the price and the estimate stops being an estimate. Management says that would move the portion of Fairfax India valued at real prices from 31% to 85%. The caveat is timing: regulatory approvals are still outstanding, and "over the next few years" is as precise as it gets.
GMR runs several Indian airports and is the closest listed comparison to Bangalore. It is here as a measuring stick, not as a suggestion: it trades at about 10.1 times its annual sales, while Bangalore is carried on Fairfax India's books at about 10 times its annual operating profit — and profit is a far smaller number than sales.
One caution the piece does not offer. Comparing a multiple of sales to a multiple of profit is not a like-for-like comparison, and it flatters the argument. The underlying point — that a listed Indian airport is valued far more generously than an unlisted one — is fair, but the specific juxtaposition is doing more rhetorical work than analytical work.
Poseidon is a private company in which Fairfax owned roughly 45%, valued in its accounts at $15.50 a share. In late May it sold about a 23% stake at $28.30 a share — more than 85% above the carrying value — raising $1.91bn and booking an $837m gain.
It is included as evidence, not as an idea. The whole article rests on the claim that assets valued by management estimate are worth more than the estimate; Poseidon is the case where somebody actually paid cash and proved it. It is the strongest single fact in the piece, and it is the reason the same argument about Fairfax India's airport deserves a hearing.
TerraVest gets one sentence, as a second example of a company that issues its own shares when they are expensive and buys them back when they are cheap. The observation being made is that this is obvious in principle and rare in practice — most managements do the reverse, buying back stock at the top when cash is plentiful.
No figures, valuation or recommendation are attached. Treat it as a name to look at later on the strength of one behaviour the author values, not as a view.
Summary derived from the archived Compounding Quality post (text in transcript.txt) for personal study. Not investment advice. © Compounding Quality / Pieter Slegers for source material.