Viking Posted February 25 Posted February 25 (edited) Earnings Estimate Summary for 2026 and 2027 - Part 1 I look forward to hearing from other board members. Please share your thoughts. That is how we all learn and improve. Introduction “What possible assurance do you have that (a stock you own) will go up in price? And if you are buying, how much should you pay? What you’re asking here is what makes a company valuable, and why it will be more valuable tomorrow than it is today. There are many theories, but to me, it always comes down to earnings and assets. Especially earnings.” Peter Lynch - One Up on Wall Street Estimating the earnings power of a company like Fairfax is far from simple. Fairfax is a non-traditional property and casualty (P/C) insurer with significant equity holdings. It under-earned for much of the past decade, and over the last five years has undergone a significant internal transformation. For investors, this means that historical financials—especially pre-2021—are often distorted and difficult to use as a guide. That is now changing. The turnaround in operating earnings that began in 2021 gathered momentum through 2022, 2023, and 2024. With 2025 results in hand, investors now have five years of cleaner, more representative data—information that better reflects Fairfax’s true fundamentals. The historical record is finally becoming a reliable input for forecasting. Yet the challenge remains: no standard, “off-the-shelf” model exists for a company as unique as Fairfax. That complexity, however, creates opportunity for investors prepared to put in the time and do the work. To address this gap, we constructed our own model to estimate Fairfax’s future earnings. The remainder of this article outlines that model and the logic behind it. The Structure of the Earnings Model Fairfax’s reported earnings are driven by five distinct income streams: Underwriting profit Interest and dividend income Share of profit of associates Non-insurance consolidated companies Net gains on investments (realized and unrealized) If we can forecast each of these reasonably well, the overall earnings estimate should be reasonably accurate. Later sections of this chapter walk through each income stream in turn. Capital Allocation – Reinvestment Perhaps the most difficult part of forecasting is anticipating how Fairfax will reinvest the substantial earnings it is now generating (roughly $4.5 billion per year). Fairfax has lots of good options and what they decide will determine which of the five income streams grow the fastest. Because we cannot predict management’s precise actions, our model must make educated assumptions about which streams benefit and by how much. As new information arrives each quarter, we update these assumptions. Looking at the last five years, the management team at Fairfax has done an outstanding job with capital allocation. My guess is they will continue to make good decisions (on balance) and this will benefit shareholders in the coming years – likely providing a tailwind to my forecasted numbers. Forecasting Methodology Our forecasting process follows a consistent sequence: Begin with historical results. We maintain detailed data going back to 2016. Adjust for both “old news” and “new news.” Internal developments - reported results, acquisitions and divestitures, management commentary, and updates from major equity holdings. External developments – P/C insurance pricing cycle, catastrophe activity, interest-rate trends, and broader economic conditions. Layer in additional assumptions where necessary. Each section of this chapter provides enough transparency for readers to understand the reasoning behind each forecast—making it easy to adjust assumptions if desired. Forecasting Time Horizon Our goal is to be as fact-based as possible. For this reason, our forecasts typically extend no more than two years: Current-year forecast (2026): An estimate of where actual earnings will land. Next-year forecast (2027): An estimate of “normalized” earnings—what Fairfax should earn in an average or typical year. Why not forecast 2028, 2029, and beyond? Because there are simply too many variables: catastrophe size and frequency, capital-allocation decisions, equity-market performance, the path of interest-rates, and the direction of the P/C pricing cycle. Attempting to build detailed multi-year forecasts introduces false precision. This does not mean I am bearish about Fairfax’s long-term prospects. Quite the opposite. I believe the company can deliver a mid-teens (or higher) ROE over the next five years, which implies strong earnings growth into 2028, 2029, 2030, and beyond. I do not need to know the exact mix of drivers today to have confidence in that outcome. For long-range estimates, the quality of the management team is arguably the most important factor—especially in P/C insurance. Fairfax’s execution over the past five years has been best-in-class among peers. That track record meaningfully raises the probability that Fairfax will continue to deliver in year three and beyond, even though we cannot yet know the exact details. The Dynamic Nature of Forecasting Forecasts become outdated as soon as they are published. This is because as time passes, we get new and betterinformation which causes earnings estimates to change – sometimes materially. As a result, our forecasts are updated frequently, usually every quarter. Is Forecasting a Good Use of Time? For me, forecasting is extremely valuable—but not because of the precise diluted EPS figure the model spits out at the end. The real value lies in the research and reasoning required to construct the forecast. When results differ from expectations, the gap becomes a prompt for investigation. Why was I wrong? Does the thesis need revision? Forecasting, in short, is a tool for deepening understanding. Better understanding improves decision-making, which improves investment outcomes. Accounting Results Versus Economic Results “…managers and investors alike must understand that accounting numbers are the beginning, not the end, of business valuation.” Warren Buffett – Berkshire Hathaway 1982AR Our model is an accounting model: it calculates earnings per share and book value. But shareholders ultimately care about economic value—what is happening to Fairfax’s intrinsic business value and earning power over time. This distinction is crucial because accounting results do not capture all the value creation that is happening at Fairfax - EPS and BV have been under-reporting the increase in business value, and in recent years the gap has been growing. One example is excess of fair value over carrying value of Fairfax’s non-insurance associates and consolidated holdings. As of December 31, 2025, this totaled $3.1 billion, or roughly $140 per diluted share (pre-tax). This is real value that has been created that has not yet appeared in reported results like EPS and BV. This is just one example of hidden value that is residing within Fairfax. Over the past five years, Fairfax’s economic results have consistently exceeded its accounting results. Investors need to recognize this gap when assessing management’s performance and valuing the company. Understanding Annual Volatility “Because our year-to-year results are inherently volatile, we believe a five-year rolling average to be appropriate for judging the historical record.” Warren Buffett – Berkshire Hathaway 1984AR Annual results are important. However, because of their business model, annual results for companies like Fairfax and Berkshire Hathaway will be volatile. This is the case for two very different reasons: P/C insurance results are volatile, due to catastrophes. Investment results are volatile, due to fluctuations in financial markets. Some years the fluctuations in reported results will be large. As a result, it makes sense for investors to use a rolling average of a couple of years when evaluating the performance of a company like Fairfax. Part 2 continues below Edited February 25 by Viking
Viking Posted February 25 Posted February 25 Earnings Estimate Summary for 2026 and 2027 - Part2 This post updates my 2026 forecast and introduces an initial 2027 estimate for Fairfax. The projections incorporate insights from the Q4 interim report and material developments over recent months. At the beginning of 2025, my EPS estimate for Fairfax was $152 per share. Actual diluted EPS came in at $214. For several years now, I have underestimated Fairfax’s earnings power. That pattern continued in 2025. The central question is whether the higher earnings base established since 2021 is durable. I believe it is. Executive Summary Looked at through the lens of earnings power, Fairfax appears positioned to continue delivering strong results. 2026 Outlook Diluted EPS: ~$187 Excess of Fair Value over Carrying Value (FV–CV): +$25/share (after tax) Economic EPS: ~$212 2027 Outlook Diluted EPS: ~$204 Excess of FV–CV: +$18/share (after tax) Economic EPS: ~$223 Street Estimates (as of February 25, 2026) Sell-side forecasts for diluted EPS forecasts: 2026: $184 2027: $183 It should be noted, these numbers exclude FV–CV value creation. Analysts understand Fairfax much better than they did three or four years ago. However, they still struggle with Fairfax’s equity holdings and capital allocation and its impact on future results – their forecast for 2027 appears overly pessimistic. Normalized Earnings Power From 2023 through 2025: Average diluted EPS: $183 Average FV–CV gains: $32/share Normalized economic EPS: ~$215 These figures likely represent a reasonable baseline for future modelling. Valuation Snapshot With Fairfax shares trading currently near $1,700: P/BV: 1.35 ($1,700 / $1260 BVPS) PE (economic EPS): ~7.9x ($1,700 / $215) For a business with Fairfax’s earnings durability, decentralized operating model, and demonstrated capital allocation skill, this valuation appears inexpensive. Forecast Framework Fairfax’s earnings have three fundamental drivers: Underwriting profit – How good is the P/C insurance business? Total return on the investment portfolio – How good is the team at Hamblin Watsa? Capital allocation – How good is senior management? High-Level Assumptions Over the past 3 years, Fairfax has delivered an average: Combine ratio (CR): 93.0% Total return on investments: 10.0% 2026 Assumptions CR: 93.5% Total return on investments: 8.5% 2027 Assumptions CR: 93.5% Total return on investments: 8.3% Total return includes annual changes in excess of fair value over carrying value for non-insurance associate and consolidated holdings. Six-Year Snapshot of Operating Earnings: The Transformation Since 2021 The chart below clearly communicates the dramatic transformation that has happened with earnings, beginning in 2021. Importantly, the increase has been driven primarily by operating income: 2021: $1.9B, or $71 per share 2025: $5.5B, or $237 per share (+234% vs 2021) 2026E: $5.8B, or $259 per share (+9% YoY) 2027E: $6.0B, or $275 per share (+6% YoY) The key point: earnings growth has been driven primarily by operating income — not one-time gains. Even as the hard market moderates, the earnings base appears durable. Detailed Assumptions for Each Line-Item in Forecast for 2026 (Preliminary; will be refined upon release of the 2025 Annual Report.) 1. Underwriting Profit Estimate: $1.73B Net premiums written growth of ~2% (moderating hard market) Combined ratio: 93.5% Average catastrophe year Continued reserve releases (following strong 2025 trend) 2. Interest and Dividend Income Estimate: $2.55B Q4 2025 run rate: $646M Tailwinds: Growth in fixed-income portfolio: $50B → $52B Expansion of mortgage loan portfolio via Kennedy Wilson Headwind: Moderation in short-term rates Yield Assumption: ~4.9% (vs ~5.0% in 2025) 3. Share of Profit of Associates Estimate: $1.05B Eurobank, Poseidon steady Improved performance from EXCO Resources (US natural gas producer) Sanmar Chemicals shifting from headwind to a tailwind 4. Non-Insurance Consolidated Operations Estimate: $450M Key contributors: Recipe, Sleep Country, Peak Achievement, AGT Food & Indredients This earnings stream is breaking out and looks positioned for meaningful growth. When building the initial forecast, separate numbers are input for 5. IFRS 17 and 6. Life Insurance/Run-Off. For quarterly actuals, these two items are combined. Fairfax does not provide these separately, but we can back into them (as a total) based on other disclosed data. 5. IFRS 17: Discounting and Risk Adjustment Two variables drive this bucket: Net written premium growth: ~2% Trend in interest rates: assumes stable interest rates (December 31, 2025 levels) This category remains difficult to model; confidence in the estimate is low. 6. Life Insurance and Run-Off Expected adverse reserve development for run-off: ~$250 million, roughly the average of the past 2 years Sale of Eurolife’s life insurance business in Greece is expected to close in Q2 ~$950M proceeds; use of proceeds TBD Models will be updated post-closing. 7. Interest Expense Estimate: $875M Based on Q4 2025 run rate ($211M) plus expected issuance. 8. Corporate Overhead and Other Estimate: $490M Modest increase vs 2025 ($480M) 9. Net Gains on Investments Estimate: $1.5B Drivers: Mark-to-market gains (FFH-TRS, Orla Mining, others). Realized gains from asset sales A major revaluation event is not built into the base case. As hidden value continues to grow, these become more likely – we just don’t know the timing. 10. Gain on Sale / Deconsolidation of Insurance Subsidiaries This this bucket captures large asset sales, usually insurance. Over the past five years, large one-time gains from asset sales/revaluations have averaged about $400 million per year. Estimate: $350M Reflects expected gain from Eurolife transaction (expected Q2 close) 11. Income Taxes Estimate: ~22% Up from 18% in 2025 Guided range (Q4 call): 22% to 25% Investment gains, taxed at lower rates, are expected to be an important driver of earnings 12. Non-Controlling Interests Estimate: $440M (~9.5% of net earnings, similar to 2025) Potential Allied World minority buyout not yet modeled. 13. Effective Shares Outstanding Estimate: 20.2M Assumes share reduction in 2026: ~650,000, or 3.1% 2025YE: 20.9M (down ~812,000, or 3.7% YoY) Additional Notes: Underwriting profit includes insurance and reinsurance (excludes life/run-off) Interest and dividend income and share of profit of associates includes insurance, reinsurance, and life/run-off
mananainvesting Posted February 26 Posted February 26 We usually talk about liquidation value (Book value) / earnings power value. I wonder what would be the Asset reproduction value of $FFH.TO? Very very likely nowhere near where the current Mcap/EV is. One can't build a culture and network that Fairfax has even in a few years, likely takes decades, what's the PV of that?!! ----- random thought!
SafetyinNumbers Posted February 26 Posted February 26 2 hours ago, mananainvesting said: We usually talk about liquidation value (Book value) / earnings power value. I wonder what would be the Asset reproduction value of $FFH.TO? Very very likely nowhere near where the current Mcap/EV is. One can't build a culture and network that Fairfax has even in a few years, likely takes decades, what's the PV of that?!! ----- random thought! Arguably a fair way to measure the intrinsic value of Fairfax is the value of the investment portfolio. Underwriting more than covers head office costs and financing costs. People would happily pay full NAV for a bond ETF and equity ETF. By buying FFH we get the returns on the full portfolio and get paid to own it. Buying at up to a 40% discount to that number probably offers very good margin of safety.
mananainvesting Posted February 26 Posted February 26 51 minutes ago, SafetyinNumbers said: Arguably a fair way to measure the intrinsic value of Fairfax is the value of the investment portfolio. Underwriting more than covers head office costs and financing costs. People would happily pay full NAV for a bond ETF and equity ETF. By buying FFH we get the returns on the full portfolio and get paid to own it. Buying at up to a 40% discount to that number probably offers very good margin of safety. don't you think a premium to NAV is warranted for the culture of the company (which protects the terminal value) and the network effect accumulated over multi-decades (OMERS, Lloyds, India, Greece,Sri Lanka...) that will help compound returns well into the future?
Hoodlum Posted February 26 Posted February 26 15 hours ago, Viking said: Earnings Estimate Summary for 2026 and 2027 - Part2 This post updates my 2026 forecast and introduces an initial 2027 estimate for Fairfax. The projections incorporate insights from the Q4 interim report and material developments over recent months. At the beginning of 2025, my EPS estimate for Fairfax was $152 per share. Actual diluted EPS came in at $214. For several years now, I have underestimated Fairfax’s earnings power. That pattern continued in 2025. The central question is whether the higher earnings base established since 2021 is durable. I believe it is. Executive Summary Looked at through the lens of earnings power, Fairfax appears positioned to continue delivering strong results. 2026 Outlook Diluted EPS: ~$187 Excess of Fair Value over Carrying Value (FV–CV): +$25/share (after tax) Economic EPS: ~$212 2027 Outlook Diluted EPS: ~$204 Excess of FV–CV: +$18/share (after tax) Economic EPS: ~$223 Street Estimates (as of February 25, 2026) Sell-side forecasts for diluted EPS forecasts: 2026: $184 2027: $183 It should be noted, these numbers exclude FV–CV value creation. Analysts understand Fairfax much better than they did three or four years ago. However, they still struggle with Fairfax’s equity holdings and capital allocation and its impact on future results – their forecast for 2027 appears overly pessimistic. Normalized Earnings Power From 2023 through 2025: Average diluted EPS: $183 Average FV–CV gains: $32/share Normalized economic EPS: ~$215 These figures likely represent a reasonable baseline for future modelling. Valuation Snapshot With Fairfax shares trading currently near $1,700: P/BV: 1.35 ($1,700 / $1260 BVPS) PE (economic EPS): ~7.9x ($1,700 / $215) For a business with Fairfax’s earnings durability, decentralized operating model, and demonstrated capital allocation skill, this valuation appears inexpensive. Forecast Framework Fairfax’s earnings have three fundamental drivers: Underwriting profit – How good is the P/C insurance business? Total return on the investment portfolio – How good is the team at Hamblin Watsa? Capital allocation – How good is senior management? High-Level Assumptions Over the past 3 years, Fairfax has delivered an average: Combine ratio (CR): 93.0% Total return on investments: 10.0% 2026 Assumptions CR: 93.5% Total return on investments: 8.5% 2027 Assumptions CR: 93.5% Total return on investments: 8.3% Total return includes annual changes in excess of fair value over carrying value for non-insurance associate and consolidated holdings. Six-Year Snapshot of Operating Earnings: The Transformation Since 2021 The chart below clearly communicates the dramatic transformation that has happened with earnings, beginning in 2021. Importantly, the increase has been driven primarily by operating income: 2021: $1.9B, or $71 per share 2025: $5.5B, or $237 per share (+234% vs 2021) 2026E: $5.8B, or $259 per share (+9% YoY) 2027E: $6.0B, or $275 per share (+6% YoY) The key point: earnings growth has been driven primarily by operating income — not one-time gains. Even as the hard market moderates, the earnings base appears durable. Detailed Assumptions for Each Line-Item in Forecast for 2026 (Preliminary; will be refined upon release of the 2025 Annual Report.) 1. Underwriting Profit Estimate: $1.73B Net premiums written growth of ~2% (moderating hard market) Combined ratio: 93.5% Average catastrophe year Continued reserve releases (following strong 2025 trend) 2. Interest and Dividend Income Estimate: $2.55B Q4 2025 run rate: $646M Tailwinds: Growth in fixed-income portfolio: $50B → $52B Expansion of mortgage loan portfolio via Kennedy Wilson Headwind: Moderation in short-term rates Yield Assumption: ~4.9% (vs ~5.0% in 2025) 3. Share of Profit of Associates Estimate: $1.05B Eurobank, Poseidon steady Improved performance from EXCO Resources (US natural gas producer) Sanmar Chemicals shifting from headwind to a tailwind 4. Non-Insurance Consolidated Operations Estimate: $450M Key contributors: Recipe, Sleep Country, Peak Achievement, AGT Food & Indredients This earnings stream is breaking out and looks positioned for meaningful growth. When building the initial forecast, separate numbers are input for 5. IFRS 17 and 6. Life Insurance/Run-Off. For quarterly actuals, these two items are combined. Fairfax does not provide these separately, but we can back into them (as a total) based on other disclosed data. 5. IFRS 17: Discounting and Risk Adjustment Two variables drive this bucket: Net written premium growth: ~2% Trend in interest rates: assumes stable interest rates (December 31, 2025 levels) This category remains difficult to model; confidence in the estimate is low. 6. Life Insurance and Run-Off Expected adverse reserve development for run-off: ~$250 million, roughly the average of the past 2 years Sale of Eurolife’s life insurance business in Greece is expected to close in Q2 ~$950M proceeds; use of proceeds TBD Models will be updated post-closing. 7. Interest Expense Estimate: $875M Based on Q4 2025 run rate ($211M) plus expected issuance. 8. Corporate Overhead and Other Estimate: $490M Modest increase vs 2025 ($480M) 9. Net Gains on Investments Estimate: $1.5B Drivers: Mark-to-market gains (FFH-TRS, Orla Mining, others). Realized gains from asset sales A major revaluation event is not built into the base case. As hidden value continues to grow, these become more likely – we just don’t know the timing. 10. Gain on Sale / Deconsolidation of Insurance Subsidiaries This this bucket captures large asset sales, usually insurance. Over the past five years, large one-time gains from asset sales/revaluations have averaged about $400 million per year. Estimate: $350M Reflects expected gain from Eurolife transaction (expected Q2 close) 11. Income Taxes Estimate: ~22% Up from 18% in 2025 Guided range (Q4 call): 22% to 25% Investment gains, taxed at lower rates, are expected to be an important driver of earnings 12. Non-Controlling Interests Estimate: $440M (~9.5% of net earnings, similar to 2025) Potential Allied World minority buyout not yet modeled. 13. Effective Shares Outstanding Estimate: 20.2M Assumes share reduction in 2026: ~650,000, or 3.1% 2025YE: 20.9M (down ~812,000, or 3.7% YoY) Additional Notes: Underwriting profit includes insurance and reinsurance (excludes life/run-off) Interest and dividend income and share of profit of associates includes insurance, reinsurance, and life/run-off Thanks @Viking for posting these detailed estimates. I was just thinking about this over the past few days and I appreciate that you have taken a somewhat conservative approach, allowing for possible upside. The one caveat of course is a large Cat season (Hurricane). The Q1 Insurance CR comparable with last year will look good as there had been no significant Wildfire like we had last January.
mengan Posted February 26 Posted February 26 On 1/23/2026 at 9:44 PM, Viking said: A question for board members... How do you think about the significant (and growing) amount of "hidden value" that is residing in Fairfax's equity portfolio/balance sheet. Is there a "right way" to think about it? (academic and/or practical) For people on the board who have owned Berkshire Hathaway long term, my guess is this has been something you have had to grapple with. How did you do it? For Fairfax, some of the "hidden value" is pretty easy to calculate (excess of FV over CV for the publicly traded holdings like Eurobank). Some of it is less precise (like BIAL). I find it a fascinating topic. It is a fact - it is not theoretical. It is a big number: My guess is ~$4.5 billion = $155/diluted share, after tax It is growing by ~+$1 billion per year As we learned the past 2 days, some analysts ignore it completely in their analysis of the company. My guess is "hidden value" at Fairfax is going to continue to increase in size in the coming years. The equity portfolio is growing rapidly. A large and growing part of it is private holdings (with limited disclosures). As a result, how investors think about "hidden value" will be an increasingly important part of their analysis. Thanks in advance! I just want to add that book value is itself a flawed metric as it is a capitalized value. This means that someone made a decision on how much the underlying assets are worth, either based on deterministic formulas (for non-listed assets) or the crowd's collective idea of the value (for publicly listed assets). I think in both cases, that value is likely to be completely wrong relative to your individual "discount rate" based on your individual risk tolerance and IRR expectations. A better way to look at the value of a company (any company) is to go directly to free cash flow, including any look-through cash flow from associates and publicly listed assets partially owned by the company; then capitalize it yourself to arrive at a personalized market value that you are willing to pay.
SafetyinNumbers Posted February 26 Posted February 26 9 hours ago, mananainvesting said: don't you think a premium to NAV is warranted for the culture of the company (which protects the terminal value) and the network effect accumulated over multi-decades (OMERS, Lloyds, India, Greece,Sri Lanka...) that will help compound returns well into the future? BRK trades at a premium to its total investments so its possible but if FFH traded up to its investment value it would be close to 3x book. The key is, we’re still at a very big discount to that so margin of safety is high. 1
sholland Posted February 26 Posted February 26 (edited) 10 hours ago, SafetyinNumbers said: Arguably a fair way to measure the intrinsic value of Fairfax is the value of the investment portfolio. Underwriting more than covers head office costs and financing costs. People would happily pay full NAV for a bond ETF and equity ETF. By buying FFH we get the returns on the full portfolio and get paid to own it. Buying at up to a 40% discount to that number probably offers very good margin of safety. $35.5B market cap divided by $70.0B portfolio minus $10.5B total debt (excluding debt at consolidated non-insurance companies which is non-recourse) = 60% (40% discount). This seems like a decent sanity check with the high quality underwriting that Fairfax has. Edited February 26 by sholland
Phoenix01 Posted February 26 Posted February 26 16 hours ago, Viking said: Earnings Estimate Summary for 2026 and 2027 - Part2 This post updates my 2026 forecast and introduces an initial 2027 estimate for Fairfax. The projections incorporate insights from the Q4 interim report and material developments over recent months. At the beginning of 2025, my EPS estimate for Fairfax was $152 per share. Actual diluted EPS came in at $214. For several years now, I have underestimated Fairfax’s earnings power. That pattern continued in 2025. The central question is whether the higher earnings base established since 2021 is durable. I believe it is. Executive Summary Looked at through the lens of earnings power, Fairfax appears positioned to continue delivering strong results. 2026 Outlook Diluted EPS: ~$187 Excess of Fair Value over Carrying Value (FV–CV): +$25/share (after tax) Economic EPS: ~$212 2027 Outlook Diluted EPS: ~$204 Excess of FV–CV: +$18/share (after tax) Economic EPS: ~$223 Street Estimates (as of February 25, 2026) Sell-side forecasts for diluted EPS forecasts: 2026: $184 2027: $183 It should be noted, these numbers exclude FV–CV value creation. Analysts understand Fairfax much better than they did three or four years ago. However, they still struggle with Fairfax’s equity holdings and capital allocation and its impact on future results – their forecast for 2027 appears overly pessimistic. Normalized Earnings Power From 2023 through 2025: Average diluted EPS: $183 Average FV–CV gains: $32/share Normalized economic EPS: ~$215 These figures likely represent a reasonable baseline for future modelling. Valuation Snapshot With Fairfax shares trading currently near $1,700: P/BV: 1.35 ($1,700 / $1260 BVPS) PE (economic EPS): ~7.9x ($1,700 / $215) For a business with Fairfax’s earnings durability, decentralized operating model, and demonstrated capital allocation skill, this valuation appears inexpensive. Forecast Framework Fairfax’s earnings have three fundamental drivers: Underwriting profit – How good is the P/C insurance business? Total return on the investment portfolio – How good is the team at Hamblin Watsa? Capital allocation – How good is senior management? High-Level Assumptions Over the past 3 years, Fairfax has delivered an average: Combine ratio (CR): 93.0% Total return on investments: 10.0% 2026 Assumptions CR: 93.5% Total return on investments: 8.5% 2027 Assumptions CR: 93.5% Total return on investments: 8.3% Total return includes annual changes in excess of fair value over carrying value for non-insurance associate and consolidated holdings. Six-Year Snapshot of Operating Earnings: The Transformation Since 2021 The chart below clearly communicates the dramatic transformation that has happened with earnings, beginning in 2021. Importantly, the increase has been driven primarily by operating income: 2021: $1.9B, or $71 per share 2025: $5.5B, or $237 per share (+234% vs 2021) 2026E: $5.8B, or $259 per share (+9% YoY) 2027E: $6.0B, or $275 per share (+6% YoY) The key point: earnings growth has been driven primarily by operating income — not one-time gains. Even as the hard market moderates, the earnings base appears durable. Detailed Assumptions for Each Line-Item in Forecast for 2026 (Preliminary; will be refined upon release of the 2025 Annual Report.) 1. Underwriting Profit Estimate: $1.73B Net premiums written growth of ~2% (moderating hard market) Combined ratio: 93.5% Average catastrophe year Continued reserve releases (following strong 2025 trend) 2. Interest and Dividend Income Estimate: $2.55B Q4 2025 run rate: $646M Tailwinds: Growth in fixed-income portfolio: $50B → $52B Expansion of mortgage loan portfolio via Kennedy Wilson Headwind: Moderation in short-term rates Yield Assumption: ~4.9% (vs ~5.0% in 2025) 3. Share of Profit of Associates Estimate: $1.05B Eurobank, Poseidon steady Improved performance from EXCO Resources (US natural gas producer) Sanmar Chemicals shifting from headwind to a tailwind 4. Non-Insurance Consolidated Operations Estimate: $450M Key contributors: Recipe, Sleep Country, Peak Achievement, AGT Food & Indredients This earnings stream is breaking out and looks positioned for meaningful growth. When building the initial forecast, separate numbers are input for 5. IFRS 17 and 6. Life Insurance/Run-Off. For quarterly actuals, these two items are combined. Fairfax does not provide these separately, but we can back into them (as a total) based on other disclosed data. 5. IFRS 17: Discounting and Risk Adjustment Two variables drive this bucket: Net written premium growth: ~2% Trend in interest rates: assumes stable interest rates (December 31, 2025 levels) This category remains difficult to model; confidence in the estimate is low. 6. Life Insurance and Run-Off Expected adverse reserve development for run-off: ~$250 million, roughly the average of the past 2 years Sale of Eurolife’s life insurance business in Greece is expected to close in Q2 ~$950M proceeds; use of proceeds TBD Models will be updated post-closing. 7. Interest Expense Estimate: $875M Based on Q4 2025 run rate ($211M) plus expected issuance. 8. Corporate Overhead and Other Estimate: $490M Modest increase vs 2025 ($480M) 9. Net Gains on Investments Estimate: $1.5B Drivers: Mark-to-market gains (FFH-TRS, Orla Mining, others). Realized gains from asset sales A major revaluation event is not built into the base case. As hidden value continues to grow, these become more likely – we just don’t know the timing. 10. Gain on Sale / Deconsolidation of Insurance Subsidiaries This this bucket captures large asset sales, usually insurance. Over the past five years, large one-time gains from asset sales/revaluations have averaged about $400 million per year. Estimate: $350M Reflects expected gain from Eurolife transaction (expected Q2 close) 11. Income Taxes Estimate: ~22% Up from 18% in 2025 Guided range (Q4 call): 22% to 25% Investment gains, taxed at lower rates, are expected to be an important driver of earnings 12. Non-Controlling Interests Estimate: $440M (~9.5% of net earnings, similar to 2025) Potential Allied World minority buyout not yet modeled. 13. Effective Shares Outstanding Estimate: 20.2M Assumes share reduction in 2026: ~650,000, or 3.1% 2025YE: 20.9M (down ~812,000, or 3.7% YoY) Additional Notes: Underwriting profit includes insurance and reinsurance (excludes life/run-off) Interest and dividend income and share of profit of associates includes insurance, reinsurance, and life/run-off @Viking - Thank you for the amazing posts. Please keep them coming... In the spirt of improving the forecast, would it make sense to account for the base CR without an CATs and then layer in a CAT factor. That would allow a better look-through on the insurance operations, and more focus on the risks of potential CATs as we move through the wildfire and hurricane season.
Viking Posted February 26 Posted February 26 39 minutes ago, Phoenix01 said: @Viking - Thank you for the amazing posts. Please keep them coming... In the spirt of improving the forecast, would it make sense to account for the base CR without an CATs and then layer in a CAT factor. That would allow a better look-through on the insurance operations, and more focus on the risks of potential CATs as we move through the wildfire and hurricane season. @Phoenix01, there are lots of different ways to build a forecast. In the beginning (start of each year), my goal is to try and come up with a ‘midpoint’ for each line item. What I would call a ‘normalized’ number for that year based on what we know today. That is what I have done for 2026 and 2027. As the year progresses, actuals will form part of the 2026 forecast. The normalized number for 2027 will get fine tuned. One time items will start to take the forecast off a ‘normalized’ number for 2026. Of course, the wild card is when big one-time events happen. We will get a really bad cat year one of these years. When that happens, it doesn’t mean my forecast was wrong (at least it doesn’t to me). Because I wasn't trying to build a forecast for a really bad year. Same with investment gains. We are going to get a really big one - one of these years. This is going to spike earnings. Anyways, my goal is to be transparent with the build so board members can then adjust as they see fit - make it more conservative or more aggressive.
Txvestor Posted February 26 Posted February 26 17 hours ago, Viking said: Earnings Estimate Summary for 2026 and 2027 - Part2 This post updates my 2026 forecast and introduces an initial 2027 estimate for Fairfax. The projections incorporate insights from the Q4 interim report and material developments over recent months. At the beginning of 2025, my EPS estimate for Fairfax was $152 per share. Actual diluted EPS came in at $214. For several years now, I have underestimated Fairfax’s earnings power. That pattern continued in 2025. The central question is whether the higher earnings base established since 2021 is durable. I believe it is. Executive Summary Looked at through the lens of earnings power, Fairfax appears positioned to continue delivering strong results. 2026 Outlook Diluted EPS: ~$187 Excess of Fair Value over Carrying Value (FV–CV): +$25/share (after tax) Economic EPS: ~$212 2027 Outlook Diluted EPS: ~$204 Excess of FV–CV: +$18/share (after tax) Economic EPS: ~$223 Street Estimates (as of February 25, 2026) Sell-side forecasts for diluted EPS forecasts: 2026: $184 2027: $183 It should be noted, these numbers exclude FV–CV value creation. Analysts understand Fairfax much better than they did three or four years ago. However, they still struggle with Fairfax’s equity holdings and capital allocation and its impact on future results – their forecast for 2027 appears overly pessimistic. Normalized Earnings Power From 2023 through 2025: Average diluted EPS: $183 Average FV–CV gains: $32/share Normalized economic EPS: ~$215 These figures likely represent a reasonable baseline for future modelling. Valuation Snapshot With Fairfax shares trading currently near $1,700: P/BV: 1.35 ($1,700 / $1260 BVPS) PE (economic EPS): ~7.9x ($1,700 / $215) For a business with Fairfax’s earnings durability, decentralized operating model, and demonstrated capital allocation skill, this valuation appears inexpensive. Forecast Framework Fairfax’s earnings have three fundamental drivers: Underwriting profit – How good is the P/C insurance business? Total return on the investment portfolio – How good is the team at Hamblin Watsa? Capital allocation – How good is senior management? High-Level Assumptions Over the past 3 years, Fairfax has delivered an average: Combine ratio (CR): 93.0% Total return on investments: 10.0% 2026 Assumptions CR: 93.5% Total return on investments: 8.5% 2027 Assumptions CR: 93.5% Total return on investments: 8.3% Total return includes annual changes in excess of fair value over carrying value for non-insurance associate and consolidated holdings. Six-Year Snapshot of Operating Earnings: The Transformation Since 2021 The chart below clearly communicates the dramatic transformation that has happened with earnings, beginning in 2021. Importantly, the increase has been driven primarily by operating income: 2021: $1.9B, or $71 per share 2025: $5.5B, or $237 per share (+234% vs 2021) 2026E: $5.8B, or $259 per share (+9% YoY) 2027E: $6.0B, or $275 per share (+6% YoY) The key point: earnings growth has been driven primarily by operating income — not one-time gains. Even as the hard market moderates, the earnings base appears durable. Detailed Assumptions for Each Line-Item in Forecast for 2026 (Preliminary; will be refined upon release of the 2025 Annual Report.) 1. Underwriting Profit Estimate: $1.73B Net premiums written growth of ~2% (moderating hard market) Combined ratio: 93.5% Average catastrophe year Continued reserve releases (following strong 2025 trend) 2. Interest and Dividend Income Estimate: $2.55B Q4 2025 run rate: $646M Tailwinds: Growth in fixed-income portfolio: $50B → $52B Expansion of mortgage loan portfolio via Kennedy Wilson Headwind: Moderation in short-term rates Yield Assumption: ~4.9% (vs ~5.0% in 2025) 3. Share of Profit of Associates Estimate: $1.05B Eurobank, Poseidon steady Improved performance from EXCO Resources (US natural gas producer) Sanmar Chemicals shifting from headwind to a tailwind 4. Non-Insurance Consolidated Operations Estimate: $450M Key contributors: Recipe, Sleep Country, Peak Achievement, AGT Food & Indredients This earnings stream is breaking out and looks positioned for meaningful growth. When building the initial forecast, separate numbers are input for 5. IFRS 17 and 6. Life Insurance/Run-Off. For quarterly actuals, these two items are combined. Fairfax does not provide these separately, but we can back into them (as a total) based on other disclosed data. 5. IFRS 17: Discounting and Risk Adjustment Two variables drive this bucket: Net written premium growth: ~2% Trend in interest rates: assumes stable interest rates (December 31, 2025 levels) This category remains difficult to model; confidence in the estimate is low. 6. Life Insurance and Run-Off Expected adverse reserve development for run-off: ~$250 million, roughly the average of the past 2 years Sale of Eurolife’s life insurance business in Greece is expected to close in Q2 ~$950M proceeds; use of proceeds TBD Models will be updated post-closing. 7. Interest Expense Estimate: $875M Based on Q4 2025 run rate ($211M) plus expected issuance. 8. Corporate Overhead and Other Estimate: $490M Modest increase vs 2025 ($480M) 9. Net Gains on Investments Estimate: $1.5B Drivers: Mark-to-market gains (FFH-TRS, Orla Mining, others). Realized gains from asset sales A major revaluation event is not built into the base case. As hidden value continues to grow, these become more likely – we just don’t know the timing. 10. Gain on Sale / Deconsolidation of Insurance Subsidiaries This this bucket captures large asset sales, usually insurance. Over the past five years, large one-time gains from asset sales/revaluations have averaged about $400 million per year. Estimate: $350M Reflects expected gain from Eurolife transaction (expected Q2 close) 11. Income Taxes Estimate: ~22% Up from 18% in 2025 Guided range (Q4 call): 22% to 25% Investment gains, taxed at lower rates, are expected to be an important driver of earnings 12. Non-Controlling Interests Estimate: $440M (~9.5% of net earnings, similar to 2025) Potential Allied World minority buyout not yet modeled. 13. Effective Shares Outstanding Estimate: 20.2M Assumes share reduction in 2026: ~650,000, or 3.1% 2025YE: 20.9M (down ~812,000, or 3.7% YoY) Additional Notes: Underwriting profit includes insurance and reinsurance (excludes life/run-off) Interest and dividend income and share of profit of associates includes insurance, reinsurance, and life/run-off With respect to capital allocation, it looks like about ~40% is coming back to shareholders via share buyback and dividends. Additionally, they have been making some acquisitions to enlarge their non insurance consolidated operations eg Peak, sleep country, Keg, redemption of preferreds etc. I'm not sure if their insurance operations have needed capital last couple of years as growth has slowed. As others have said before, theirs is a 360° Capital Allocation model. They are likely deciding by the various movements within their portfolio on the location of the best value. It doesn't even have to be outside of. Even within the company which they presumably know fairly well, they have options. I am sure they are looking at their desired hurdle rate based on strategic importance and share buybacks as a baseline. With the number of such opportunities around, including arguably, depressed stock price, insurance sub. minority buy back options, internal source transactions, insurance sub. capital when growth is available, redemption of various debt and pref vehicles. and so on. It's a wide canvas even before even looking at external equity investments.
MMM20 Posted February 26 Posted February 26 (edited) On 1/23/2026 at 6:36 AM, MMM20 said: Reminder that ~10% drawdowns are still common... Price of admission for ~40% compounding for ~5 years... And the stock is still cheap however you cut it... Especially if you apply Buffett's logic on the float... A) book value + float methodology B) two-pillar methodology I'm a buyer. I'd be a seller if it doubled overnight. Hey @djokovic1 I saw you posting on the recent Fairfax writeup on VIC. Maybe suggest @eigenvalue check Buffett's book+float or two-pillar methodologies if he's so sure ~20x P/E isn't fair for this thing? Edited February 26 by MMM20
Marco Van Basten Posted February 26 Posted February 26 1 hour ago, MMM20 said: Hey @djokovic1 I saw you posting on the recent Fairfax writeup on VIC. Maybe suggest @eigenvalue check Buffett's book+float or two-pillar methodologies if he's so sure ~20x P/E isn't fair for this thing? Given that eigenvalue posted the stock when it was at USD 900 two years ago, you might want to re-read what he posted before giving him advice...
MMM20 Posted February 26 Posted February 26 (edited) 2 hours ago, Marco Van Basten said: Given that eigenvalue posted the stock when it was at USD 900 two years ago, you might want to re-read what he posted before giving him advice... You might want to re-read what I’ve posted bc I’ve been saying here for more than 4 years that even bulls like him are too conservative and intrinsic value is US$2K -> $3-4K. Yes, even if that’s 20x P/E now and so are some other great compounders. It was priced to compound at 40% then, it’s still 20% now and that’s still too cheap by half. Edited February 26 by MMM20
TwoCitiesCapital Posted February 26 Posted February 26 (edited) I think the problem with pricing a company "correctly" is the margin you have to be right by. Two companies are identical. They both deliver a 20% RoE and both have a 7% cost of capital/discount rate. If company #2 misses in year 20 and has 0 earnings for that one year, the impact to the DCF to today's valuation is ~13%. If that miss occurs in year 1 instead of 20, the difference in valuation today is ~16%. And that's if the company compounds at 20% per year in 95% of observations. What if it only happens 90% of the time? Or 80%? Both still result in phenomenal track records of compounding revenue/profit/book, but dramatically change your returns as an investor if you paid up for the 95% and only got 80%. If Fairfax was valued at 3x book, I'd be a seller all day. Maybe it looks stupid up front, particularly if they keep compounding at 20% for 1, 3, or 5 years. But all it takes is 1-2 bad years in 20 of not hitting your 20% goal to screw your returns if it was priced at perfection (and this is BEFORE considering multiple contraction that would also likely occur). Just look no further than CSU - was a market darling the last 5-years and has a phenomenal track record. Is now down more than 50% from its highs despite still hitting its return goals. People just got scared that it may not and now there is an active debate back and forth in that thread if it's STILL too expensive or not after the 50% drawdown. If you bought 1-year ago, you paid for 100% execution and you may only get 80%. Edited February 26 by TwoCitiesCapital
Munger_Disciple Posted February 26 Posted February 26 (edited) 29 minutes ago, TwoCitiesCapital said: I think the problem with pricing a company "correctly" is the margin you have to be right by. Two companies are identical. They both deliver a 20% RoE and both have a 7% cost of capital/discount rate. If company #2 misses in year 20 and has 0 earnings for that one year, the impact to the DCF to today's valuation is ~13%. If that miss occurs in year 1 instead of 20, the difference in valuation today is ~16%. And that's if the company compounds at 20% per year in 95% of observations. What if it only happens 90% of the time? Or 80%? Both still result in phenomenal track records of compounding revenue/profit/book, but dramatically change your returns as an investor if you paid up for the 95% and only got 80%. If Fairfax was valued at 3x book, I'd be a seller all day. Maybe it looks stupid up front, particularly if they keep compounding at 20% for 1, 3, or 5 years. But all it takes is 1-2 bad years in 20 of not hitting your 20% goal to screw your returns if it was priced at perfection (and this is BEFORE considering multiple contraction that would also likely occur). Just look no further than CSU - was a market darling the last 5-years and has a phenomenal track record. Is now down more than 50% from its highs despite still hitting its return goals. People just got scared that it may not and now there is an active debate back and forth in that thread if it's STILL too expensive or not after the 50% drawdown. If you bought 1-year ago, you paid for 100% execution and you may only get 80%. Good post! Adds a much needed dose of reality to this bulled-up board. Edited February 26 by Munger_Disciple
73 Reds Posted February 26 Posted February 26 25 minutes ago, TwoCitiesCapital said: I think the problem with pricing a company "correctly" is the margin you have to be right by. Two companies are identical. They both deliver a 20% RoE and both have a 7% cost of capital/discount rate. If company #2 misses in year 20 and has 0 earnings for that one year, the impact to the DCF to today's valuation is ~13%. If that miss occurs in year 1 instead of 20, the difference in valuation today is ~16%. And that's if the company compounds at 20% per year in 95% of observations. What if it only happens 90% of the time? Or 80%? Both still result in phenomenal track records of compounding revenue/profit/book, but dramatically change your returns as an investor if you paid up for the 95% and only got 80%. If Fairfax was valued at 3x book, I'd be a seller all day. Maybe it looks stupid up front, particularly if they keep compounding at 20% for 1, 3, or 5 years. But all it takes is 1-2 bad years in 20 of not hitting your 20% goal to screw your returns if it was priced at perfection (and this is BEFORE considering multiple contraction that would also likely occur). Just look no further than CSU - was a market darling the last 5-years and has a phenomenal track record. Is now down more than 50% from its highs despite still hitting its return goals. People just got scared that it may not and now there is an active debate back and forth in that thread if it's STILL too expensive or not after the 50% drawdown. If you bought 1-year ago, you paid for 100% execution and you may only get 80%. Markets are forward looking(?) Bring new variables into the mix and expect things to remain the same?
Txvestor Posted February 26 Posted February 26 21 minutes ago, Munger_Disciple said: Good post! Adds a much needed dose of reality to this bulled-up board. But in Fairfax's case they will be both exceeding and missing those ROE goals based on muliple variables which are discussed here. Its almost better to take a rolling 5yr earnings average to get a better understand with this company. And even then you'd be missing the increase in intrinsic value increasingly building outside of what Book value captures. However for long term holders that hardly matter as it both adds an option in capital allocation decisions and as long as the average continues at a reasonable number it will be fine over a 10-20yr horizon. Now like you said if the growth was discounted far out at a 3x BV then sure you could make a case.
Munger_Disciple Posted February 26 Posted February 26 5 minutes ago, Txvestor said: But in Fairfax's case they will be both exceeding and missing those ROE goals based on muliple variables which are discussed here. Its almost better to take a rolling 5yr earnings average to get a better understand with this company. And even then you'd be missing the increase in intrinsic value increasingly building outside of what Book value captures. However for long term holders that hardly matter as it both adds an option in capital allocation decisions and as long as the average continues at a reasonable number it will be fine over a 10-20yr horizon. Now like you said if the growth was discounted far out at a 3x BV then sure you could make a case. I particularly agree with this part of the post by @TwoCitiesCapital: If Fairfax was valued at 3x book, I'd be a seller all day.
MMM20 Posted February 26 Posted February 26 35 minutes ago, Munger_Disciple said: I particularly agree with this part of the post by @TwoCitiesCapital: If Fairfax was valued at 3x book, I'd be a seller all day. I agree - I'd sell if it doubled overnight on no news.
Marco Van Basten Posted February 26 Posted February 26 1 hour ago, MMM20 said: You might want to re-read what I’ve posted bc I’ve been saying here for more than 5 years that even bulls like him are too conservative and intrinsic value is US$2K -> $3-4K. Yes, even if that’s 20x P/E now and so are some other great compounders. It was priced to compound at 40% then, it’s still 20% now and that’s still too cheap by half. A five year old bull? In commercial agriculture, they are usually culled at 4-5 years due to fertility... He said probably sell at 2x book value or 15x EPS, which is north of $3100 today. That's 80% above today's stock price, so it is not terribly relevant now... Kind of medieval philosophers debating how many angels fit on a head of a pin... By the way, I am heading to your favorite place in a few months, any intelligence you'd like me to gather?
LC Posted February 27 Posted February 27 Do we have some estimate of western wildfire exposure to Fairfax? I imagine they sell policies via Northbridge.
SafetyinNumbers Posted February 27 Posted February 27 10 hours ago, Munger_Disciple said: I particularly agree with this part of the post by @TwoCitiesCapital: If Fairfax was valued at 3x book, I'd be a seller all day. What are the chances you would still own any at 3x book?
djokovic1 Posted February 27 Posted February 27 (edited) @Viking I think about 2026 and 2027 similar to you. But I am higher for diluted EPS, $200 for 2026 and $230 for 2027. Similar to you I think operating profit will be flattish and investment gains will be more moderate (relative to 2025), however, if you have no premium growth you can assume all the profits will be funnelled into buybacks. You will get shares outstanding go down by much more than 3-4%, closer to 8-10%. Conversely they could use that capital to do M&A (or buy out minorities), in that case top-pline will not be flat and grow more than we have assumed. Which explains my delta to your numbers. @MMM20 I will let Eigen come to his own conclusions. Personally, we have discussed on the board here, I think Fairfax should trade at ~2x book+ if you expect 15% ROE+. Right now it's closer to 1x book. Of course if it triples tomorrow and gets close to 3x book, most of us will be silly to not think of selling! But if it doubles tomorrow to 2x book I am comfortable holding on and gain from the EPS compounding. We have discussed before, it trades at much lower multiples than peers who have lower ROE's and most of them shorter track records. Bullish or not, I care about being accurate, rather than conservative or optimistic. To me it's pretty clear Fairfax is significantly mispriced relative to peers and on an absolute basis and shareholders will do very well from today with a 5 year + horizon. Edited February 27 by djokovic1 1
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now