The Fairfax Way with David Thomas $FFH.TO
Summary
Fairfax’s 19.2% cumulative annual return from 1985 through 2024 is the headline, but the deeper story is that different engines carried the company at different times. Investments offset years of weak insurance underwriting; insurance later improved while hedging losses were costly; now insurance, investments, and non-insurance operations are contributing together. David Thomas calls Fairfax “almost like a teenager as they turn 40” because its second wind is arriving after decades of hard lessons.
The foundational mistake was applying value-investing logic to insurers without respecting their hidden liabilities. Fairfax repeatedly bought businesses below book value, then discovered under-reserving, asbestos exposure, and other long-tail claims that could take years to surface and repair. Watsa said he would not today buy an insurer in that condition; the later acquisitions, especially Allied World in 2017, were bolt-ons rather than “broken cheap rundown assets.”
Prem Watsa’s macro record contains both remarkably prescient calls and a $4 billion lesson in asymmetric risk. Fairfax was cautious into 1987, sold down Japan in the late 1980s and later shorted its index, shorted technology in 2000, and built its mortgage-credit trade from 2003–04 by asking whether its own reinsurers—and ultimately Fannie and Freddie—could fail. Yet its 2010–16 equity hedges became “very costly protection,” teaching Fairfax that it had been “doing it for the right reason, but…doing it wrong.”
Fairfax’s investment process is less about favored sectors than about backing managers who fit a decentralized culture. Watsa’s formulation is “management, management, management”: Seaspan, Recipe Unlimited, Eurobank, Sleep Country, and even BlackBerry reflect bets on operators more than a single macro template. The portfolio changes materially across decades, while the emerging collection of controlled non-insurance businesses has become a genuine third earnings engine.
BlackBerry shows both the attraction and danger of Fairfax’s willingness to give troubled companies a “longer leash.” Watsa believed John Chen could stabilize a once-great company after conflict between its founder co-CEOs, but progress never became sustained momentum; a proposed take-private instead became roughly $1 billion of convertible financing. Thomas cannot explain the precise gating factor behind that change—an important honest non-answer—but sees the investment as a manager-led turnaround bet, not a technology thesis.
The short-seller battle was neither a clean vindication of Fairfax nor a simple fraud story. Under-reserved insurers, complicated cross-border capital movements, minimal media engagement, and Fairfax’s 74% Odyssey Re stake—followed by a convertible transaction to restore the 80% tax-consolidation threshold—gave skeptics concrete reasons to question the company, even though the underlying interpretation remained contested. The campaign nevertheless escalated into “another Enron” and “fraud of the century” allegations, with alleged questionable investigative tactics and plans for Bob Dylan to play Fairfax’s supposed funeral party. Fairfax’s 2006 lawsuit sought $6 billion, produced some settlements, and was weakened by jurisdictional issues.
Today’s central question is whether Fairfax has escaped its “seven lean years, seven fat years” rhythm without losing Watsa’s exceptional judgment. Buybacks have retired more than 20% of shares during the latest run, but recently slowed as valuation rose; the book discusses Prem’s son becoming chairman, alongside named successors across corporate leadership, insurance, and investing. Thomas’s strongest structural observation is that “there’s no part of their business that’s broken,” though Fairfax’s 15% long-term book-value-growth objective remains an aspiration “over time,” not a promise for every year.
Deep dive
1. Fairfax compounded at 19.2% by surviving its own learning curve
Thomas came to Fairfax as a Canadian business journalist and longtime reader of Watsa’s shareholder letters. Their relationship began with a wide-ranging interview about capitalism as “a force for good,” Watsa’s belief that companies should “stand for good,” and the obligation to treat people properly.
Watsa had little perceived need for media, so access developed slowly. Thomas ultimately told him, in effect, that he would start writing and prove he could accurately translate the company’s history; the resulting research and interview “dance” took three years.
The numerical case for caring is a 19.2% cumulative annual return from Fairfax’s 1985 founding through year-end 2024. Early returns briefly ran around 35%–40%, but Thomas finds the recent acceleration unusual: Fairfax is enjoying a “second wind” at 40 rather than merely decelerating with scale.
Walker’s qualification is important: Berkshire’s slowing returns partly reflect becoming “so, so effing big.” Fairfax is big but does not face Berkshire’s same scale constraint.
2. Cheap insurers taught Fairfax that book value can conceal decades of pain
Walker notes that the first five insurance companies Fairfax bought were under-reserved. Fairfax began with strong investors who assumed operating insurance “wasn’t that hard,” but value discipline proved dangerous when asbestos and other long-tail liabilities emerged years later.
The investment portfolio compensated for prolonged repairs at Crum & Forster, TIG, and other troubled American operations. Fairfax needed additional reinsurance protection and flexible movement of capital among subsidiaries simply to ensure those businesses could meet claims and cash requirements.
Thomas rejects the easy counterfactual that Fairfax should have remained a hedge fund: the insurance-and-investment model was a deliberate long-term choice. The better question is whether it should have built that model differently, and Watsa said he would not today buy an asset requiring such a lengthy rehabilitation.
Allied World, Fairfax’s large 2017 acquisition, was the biggest of the later bolt-ons and represents the revised playbook. The turnaround required neither abandoning decentralization nor replacing leadership wholesale—it required recognizing that “broken” insurers are qualitatively different from cheap industrial assets.
3. “Management, management, management” binds an otherwise eclectic portfolio
Thomas’s clearest description of Watsa’s investment filter is “management, management, management.” Fairfax has sometimes hired people before defining their roles because cultural fit, capital-allocation judgment, and a record of profitably running businesses mattered more than filling an organizational box.
That judgment is inseparable from decentralization: Fairfax must trust subsidiary leaders because Toronto does not centrally operate each insurer, restaurant, ship, bank, or manufacturer. Long CEO tenures are therefore not incidental; they are infrastructure for the model.
The operating collection now includes Seaspan shipping through Atlas/Poseidon, Recipe Unlimited’s restaurant brands, Sleep Country, banks in Egypt, and developing digital insurers. Thomas’s point is not that mattresses or restaurants express a grand macro view, but that capable managers created investable situations.
This differs from the handful of enduring public equities commonly associated with Berkshire. Fairfax invests for the long term, but its public portfolio changes materially across decades and includes securities many outside shareholders cannot readily replicate.
4. Eurobank rewarded patience; BlackBerry exposed its limits
Eurobank is Thomas’s sharpest value-investing specimen: Fairfax entered during Greece’s post-crisis and deflationary distress, watched the position fall almost to nothing, and recapitalized repeatedly. After years of pain, it became Fairfax’s largest holding and a major win.
Walker sees an inflation-protection flavor in holdings such as Orla Mining and Occidental Petroleum, but Thomas resists forcing the equity portfolio into one macro template. Fairfax’s macro expression has historically been clearest in bonds, rates, hedges, and aggregate exposure rather than every common-stock selection.
BlackBerry was a different kind of rescue. Watsa believed John Chen could restore momentum after conflict between the founder co-CEOs at RIM, but despite progress, the turnaround “never really got the momentum going”; Fairfax’s proposed acquisition ultimately became approximately $1 billion of convertible financing.
Walker presses on why Fairfax publicly offered to buy BlackBerry and then changed course. Thomas’s honest non-answer: “I can’t even begin to speak to that”—there were many moving parts, but he cannot supply a definitive behind-the-scenes gating factor.
5. Fairfax’s best macro trade began as counterparty risk management
Fairfax was cautious into the 1987 crash, sold down Japan in the late 1980s, later shorted the Japanese index, and profited from shorting technology around 2000. Those gains arrived when its weakened insurance subsidiaries badly needed investment support.
The global-financial-crisis trade began around 2003–04 with warnings about asset-backed securities and the mortgage market. Fairfax was down a lot on its positions and repeatedly added to them years before the break because its team remained convinced the underlying counterparty chain was fragile.
The analytical leap impressing Walker was not merely forecasting bad mortgages. Fairfax asked whether reinsurers providing its own protection could fail, whether their funding and guarantees were dependable, and eventually whether Fannie and Freddie could survive the same stress.
That structure was crucial: Fairfax knew the loss on the protection while retaining multiples of upside if the system broke. Jim Chanos later argued that the windfall was large enough to overcome the insurance-reserving weakness central to his bearish Fairfax thesis.
6. A $4 billion hedging loss rewrote the rules of defense
After the crisis, Fairfax feared a Depression- or Japan-style deflationary spiral: inflation was at 0% and dipping into deflation, sovereign stress was hitting the PIIGS, debt loads looked unsustainable, and several countries appeared close to breaking. What it underestimated was unprecedented, coordinated central-bank intervention through quantitative easing.
From 2010 through 2016, equity hedges and shorts cost roughly $4 billion and effectively wiped out earnings. Unlike the earlier credit-default swaps, the exposure could keep losing as markets rose; maintaining it eventually forced Fairfax to sell high-quality equities bought cheaply during the crisis.
Walker’s pushback—worth keeping: Fairfax could have retained Berkshire, J&J, and similar holdings while shorting the index, creating something closer to net-neutral exposure despite basis risk. That would at least have allowed the long portfolio to offset part of a continuing market rise.
Watsa did not whitewash the result. Thomas recalls his shareholder-letter posture as, “Here’s how dumb we are…this is a good lesson. You should never do this”; Fairfax concluded it needed options or other protections with bounded downside.
7. Political turns matter to Watsa when they resemble management changes
Walker identifies three political inflection points: Trump’s 2016 election helped Fairfax conclude that its deflation fears had passed and end the equity hedges; Watsa leaned into India after Narendra Modi’s election; and Fairfax expanded in Greece during a favorable political turn. Thomas’s direct explanation centers on evaluating leaders like managers, especially in India.
Thomas links the Modi decision to Fairfax’s manager-first philosophy: Watsa assessed whether Modi could make progress on corruption and deregulation and unlock value in an economy with room to grow. The broader political interpretation remains Walker’s framing rather than a fully specified explanation from Thomas.
Walker reads Fairfax’s 2024 letter as distinctly cautious: the U.S. stock market represented roughly 70% of the world index versus 26% of the world economy, valuations appeared beyond the dot-com bubble, and concentration in the Magnificent 7 made a reversal in the wealth effect dangerous.
Walker also credits Fairfax with a major inflation-trade win in 2022, though he considers that episode less interesting than the earlier macro calls.
Thomas sees no evidence of a dramatic new portfolio shift. Watsa has always flagged expensive markets and distortions, sometimes correctly and sometimes prematurely: “That’s just Prem being Prem. Things are expensive…tread carefully.”
8. The short campaign exploited real complexity before becoming something darker
The battle intensified after Fairfax’s New York listing increased its American visibility. Shorts encountered genuinely vulnerable insurers, limited reserves, complicated global holding structures, and capital moving among affiliates—including Indian assets held through Mauritius—which could resemble concealment even if designed for flexibility.
Fairfax’s limited media engagement left an information vacuum. Hedge funds and associated investigators pushed descriptions including “crime of the century,” “fraud of the century,” and “another Enron”; alleged tactics became increasingly questionable, and antagonists reportedly planned a Hamptons funeral party with Bob Dylan after Fairfax’s expected collapse.
Walker’s most concrete challenge concerns Odyssey Re: Fairfax floated enough shares to reduce ownership to 74%, then needed a convertible transaction to recover the 80% required for tax consolidation. If management mishandled that visible threshold, he asks, why should investors trust its far harder insurance-reserve estimates?
Fairfax filed a $6 billion lawsuit in 2006 involving racketeering claims in New Jersey. Walker notes that suing short sellers is now often a red flag; Thomas answers that Fairfax needed a forum to put the alleged campaign on paper, though jurisdictional rulings weakened the case, some matters settled separately, and the larger dispute never cleanly concluded.
9. The next Fairfax will rely on three engines and a distributed succession
Fairfax has retired more than 20% of its shares during the latest repurchase cycle after Watsa outlined an aggressive decade-long program in 2017. Recent slowing might reflect valuation or preparation for opportunities, but Thomas has no private knowledge of the company’s intended capital deployment.
Their valuation exchange catches a practical error: Walker initially cites roughly 2× book value, while Thomas estimates nearer 1.3×; they conclude Walker likely compared a Canadian-dollar share price with U.S.-dollar book value. Both agree Watsa believes intrinsic value exceeds book, without specifying the multiple.
Kennedy Wilson fits the manager-led pattern. Its relationship with Fairfax spans the Bank of Ireland rescue and subsequent Greek investments; a take-private could also bring an acquired regional-bank bond portfolio and its operating team inside Fairfax, though Thomas offers no certainty that the transaction will close.
Succession now has identifiable depth: the book discusses Prem’s son becoming chairman, Peter Clark provides CEO-level continuity, Brian Young has worked for decades with insurance architect Andy Barnard, and Roger Lace, bond specialist Brian Bradstreet, and a younger team provide investment continuity.
Walker frames Fairfax’s history as “seven lean years, seven fat years,” but Thomas sees a structural break: profitable insurers, stronger bond income, and non-insurance businesses now work simultaneously. “There’s no part of their business that’s broken”—something Fairfax could not previously say.
The long-term objective has already softened from 20% to 15%, then from an annual ROE formulation to book-value growth “over time.” Watsa keeps the hurdle high to focus the organization; Thomas doubts it will be cut soon, but does not claim Fairfax can achieve it every year.