Fairfax Financial's 2025Q4 13F: Prem Watsa's Deep-Value Tape

Fairfax Financial Holdings Ltd/Can, the Toronto-headquartered insurer-investor run by Prem Watsa for four decades, filed a 2025Q4 13F showing $2.08 billion of US-listed equity positions across 29 holdings. The filing is small by mega-cap-manager standards, but read closely, it is one of the cleaner reads of the year on deep-value contrarian discipline. The top of the book is anchored in gold (Orla Mining at 20.33%), oil (Occidental at 11.94%), and a stack of unloved consumer and steel names that almost every long-only growth manager has cut to underweight or zero. Watsa's portfolio reads like a list of consensus-bearish ideas — which is exactly the way it has always read, and exactly why it deserves a careful breakdown.

The Top of the Book Is the Whole Story

Eight positions account for $1.68 billion of the $2.08 billion total — 80.6% concentration in the top eight. None of those eight is a household momentum name; six of the eight have traded sideways or down on a one-year basis. The full ranked list:

The signature value-investor pattern is visible in every line. Each of these securities has at least one of three things in common: (a) a multi-year drawdown that scared off momentum money, (b) a structural cyclical exposure (steel, oil, real estate, gold) that benchmark managers underweight, or (c) a long-duration turnaround thesis that needs years of patience to resolve. UAA at 10% of the book is the single boldest expression — Under Armour has been compressed by Nike and Lululemon for four years; Watsa's bet is that the brand survives consolidation and re-rates with a multi-year operational reset.

ORLA: The Gold Position That Reframes the Whole Book

The single position deserving the most attention is Orla Mining, a mid-cap gold producer at 20.33% of the disclosed portfolio. Fairfax has been macro-hedged for years against deflation risk and US dollar debasement; a concentrated single-name gold position is one of the cleaner expressions of that macro view, far cleaner than a basket of gold ETFs or producers. Orla operates the Camino Rojo mine in Mexico and the Musselwhite mine in Ontario — both lower-cost properties with multi-year reserve lives. The position sizing implies a real macro view, not a barbell tilt.

For comparison, the next-largest US-listed manager 13F to hold ORLA at a similar percentage of book is roughly 10× smaller in AUM. This is one of the relatively rare situations in our coverage universe where a multi-billion-dollar 13F carries a single-name exposure that no peer of comparable size is matching. Investors tracking precious-metals factor exposure inside diversified value books can compare this against the broader gold-holder universe via the combined holdings tool.

BlackBerry: The Legacy Position Tells You About the Process

BlackBerry at $132.2M and 6.35% of the book is the position most worth contextualizing for new readers of Fairfax. Watsa first established this stake in 2013 when BlackBerry was attempting to sell itself; Fairfax was the lead bidder, the deal collapsed, and the firm has retained a structural position ever since. Through twelve years of operational pivots — handset wind-down, QNX automotive software pivot, AI security re-positioning — Fairfax has not exited. This is not a discretionary trading position; it is a near-permanent strategic relationship that reads on the 13F like a stock pick. The pattern is repeated, in less stark form, across multiple holdings: CVS, KW, and KHC are all multi-quarter stable positions that have weathered drawdowns without trim. The portfolio turnover rate inside Fairfax's US book has historically been below 20% per quarter — well below the active-manager median.

Concentration Curve and AUM Stability

The eight-quarter AUM history shows a portfolio that has been deliberately small and deliberately stable. AUM has ranged from $1.87B to $2.08B over the period — a 10% band — while the holding count has stayed in the 27-31 range. That is the inverse of the Sequoia-style venture-distribution shape that produces big sequential resets; Fairfax's US 13F is engineered to be steady, with positions sized to weather multi-year drawdowns rather than to capture short-cycle mean reversion. The 2024Q3 to 2025Q4 progression — $1.95B → $2.08B with 30 → 29 holdings — confirms the patience profile.

Note the data quirk visible in the 2025Q4 holdings list: every position is flagged "NEW" by the comparison engine. This is not a real portfolio reset. It reflects a Form-vehicle reclassification — Fairfax has historically split its US 13F filings across multiple registered subsidiaries, and the canonical filer-CIK we display here picked up a clean set of holdings in 2025Q4. The economic positions did not change; the reporting vehicle did. WhaleScore on the consolidated portfolio sits at 74.75, the high end of the value-disciplined cluster.

What This Means for Long-Only Investors

Three reads are actionable for institutional investors who do not directly mirror Fairfax:

• Cross-reference the Watsa names against your factor model — ORLA + OXY + CLF + UAA + KHC together capture a precise factor exposure (commodity + cyclical + value + brand turnaround). If your existing US large-cap value sleeve carries similar dollar weight in these names, the implied factor concentration is double-counted. If your sleeve carries zero, this is the active manager's read of where deep value lives.
• Read the ORLA holder list as the macro proxy — A 20% Fairfax stake in Orla Mining is a sizeable absorption of the available float. If you want exposure to gold via an active-money lens rather than via an ETF, the ORLA holder page shows who else is sized similarly.
• Watch the next 13F for any structural trim of BB or UAA — These are the two positions most likely to signal a thesis change, because Fairfax has held both through multiple drawdowns. A trim greater than 25% in 2026Q1 would be the cleanest possible signal that Watsa's patience has run out on one of these resolution stories. Track both via the Fairfax filer page.

What to Watch

• Fairfax's 2026Q1 13F (due May 15, 2026) — first full quarter for trim direction on the eight top positions. Specifically watch ORLA share count: any reduction below the 2025Q4 share count flags a macro view change.
• Fairfax's Q1 2026 quarterly letter (typically released late April) — Watsa's commentary on macro positioning is usually the cleanest interpretive layer over the 13F mechanics.
• Berkshire's Q2 2026 13F — Fairfax and Berkshire co-hold OXY and KHC. Cross-comparison of position changes between the two value-disciplined managers is the closest thing to a second opinion on these long-duration names. The Q2 2026 Berkshire filing is due August 14, 2026.

The 2025Q4 Fairfax 13F-HR is on EDGAR under CIK 0000915191; the underlying AUM-by-quarter trajectory, position turnover summary, and full holder cross-reference live on the Fairfax filer page. For the SEC primary, search the EDGAR 13F filing history.