Jean Marie Eveillard

Jean-Marie Eveillard's method is best understood as a historically evolving practice, not a timeless checklist. He began with Benjamin Graham's asset-based discipline, added Warren Buffett's emphasis on durable business economics, and after 2008 gave macro risks a larger defensive role. The constant was humility about forecasts: price had to leave room for an uncertain future. Eveillard is living but retired; a March 2026 filing lists him as Trustee Emeritus, a non-voting consultative role rather than investment authority (First Eagle Variable Funds report, 2026; First Eagle SAI, 2026). Accordingly, current First Eagle doctrine and portfolios are treated below as successor evidence, not as decisions by Eveillard.

Core Worldview

Eveillard's starting point was Graham's idea that a security represents partial ownership of a business with an intrinsic value independent of its quoted price. That framework supplied an “order” the forecasting culture of his early career lacked. Because intrinsic value is only an estimate and the future is uncertain, the investor must buy with a margin of safety. In his compact formulation, “Risk to us is absolutely not volatility.” Risk is a permanent impairment of capital caused by paying too much, misreading the business, excessive debt, dishonest reporting, or a change that destroys earning power (Graham & Doddsville interview, 2007/08, pp. 1–6).

Value investing was a “big tent,” with Graham and Buffett representing different but compatible forms. Graham offered a static, quantitative margin of safety: buy assets or normalized earnings at a deep discount. Buffett accepted a smaller apparent discount for a franchise whose moat could grow intrinsic value. Eveillard moved between them. He preferred the Graham end when staff was scarce or business quality was difficult to forecast; a larger analyst team later made Buffett-style qualitative work feasible. He nevertheless approached the Buffett end with trepidation, explicitly denying that he had Buffett's exceptional judgment (Columbia Business School, 2005; Graham & Doddsville, pp. 1–3).

This was an absolute-return orientation, but not indifference to all relative results. In a 2012 interview Eveillard set two objectives over time: beat money-market returns, preferably after inflation, and outperform benchmarks and peers. He rejected a fixed numerical return target and short-term benchmark control. The precise claim is therefore that he accepted long periods of relative lag to protect and compound real capital—not that relative performance never mattered (MOI Global interview, recorded 2012).

The Edge - What Markets Misprice and Why

Eveillard did not claim an informational advantage. “For value investors, the edge is seldom in unusual information,” he said; the advantage was interpreting public evidence differently. Most investors and sell-side analysts worked on six-to-twelve-month horizons, whereas his team underwrote businesses over five years or longer. A patient investor could buy from owners reacting to a weak quarter, an unpopular country, a broken narrative, or a temporary operational cloud (Graham & Doddsville, pp. 3–6).

The behavioral mechanism mattered as much as the arithmetic. Herding feels safer, relative underperformance threatens jobs, consultants reward conformity, and open-end fund investors can redeem before value emerges. During the technology boom Eveillard refused to buy businesses he could not value, yet seven of ten shareholders reportedly left between 1997 and spring 2000. Contemporary reporting captured the dilemma before vindication: either abandon the method or depend on clients willing to tolerate the lag (Journal Record/AP, 1998; Ivey, 2014). The edge thus required both analytical independence and an institution capable of surviving its commercial consequences.

Global breadth widened the opportunity set, particularly where local convention, language, cross-holdings, or neglect obscured value. It did not make every foreign market investable. Eveillard was wary where accounting, political intervention, or minority-shareholder treatment made permanent impairment impossible to appraise. His globalism was selective and industry-led, not a rule that foreign exposure itself diversified risk (Columbia Business School, 2005).

Process

1. Idea Sourcing

Ideas came from continuous reading—newspapers, company material, and incoming research—plus suggestions from portfolio managers and analysts. Sell-side reports could point toward a company, but their conclusions and horizons were not the final work. Analysts also generated ideas independently and had permission to reject a portfolio manager's suggestion after investigation. This decentralized challenge function reduced the risk that Eveillard's status would turn a prompt into a foregone conclusion (Graham & Doddsville, pp. 3–4).

He did not rely on mechanical screens. Reported accounts might omit hidden assets, understate liabilities, capitalize questionable costs, or obscure stock-option expense. A screen could be a lead, but it could not make the adjustments that converted accounting data into economic reality. The investment universe was therefore filtered first by understandability and evidence quality, not merely a low published multiple.

2. Research

Research was performed in-house from public information. The team rebuilt the relevant accounts, then reduced the significant figures to one or two pages and identified roughly three to five characteristics that controlled the business's economics. This was analytical compression, not superficiality: an analyst had to know enough to discard the immaterial. Eveillard spent much of his time reading and interrogating colleagues rather than independently duplicating each model (Graham & Doddsville, pp. 3–4).

Management meetings came after substantial work. Business quality ranked before managerial brilliance because a strong franchise could withstand ordinary managers, but stewardship was not irrelevant. Intellectually dishonest executives, opaque footnotes, dubious tax rates, or aggressive accounting could disqualify an investment. At a 2017 CFA Chicago event, Eveillard retold the Enron lesson: when an analyst could not understand the footnotes, the team moved on (CFA Society Chicago, 2017). The operational circle of competence was simple: understand the business, like its economics, and believe its security was mispriced.

3. Valuation and Entry

Intrinsic value was what a knowledgeable, rational buyer expecting a reasonable return would pay in cash for the whole business. Eveillard favored enterprise value relative to operating earnings because enterprise value incorporated cash and debt, while operating income avoided pretending that depreciation represented no economic cost. He also used adjusted asset values, normalized earnings, comparable transactions, and sum-of-the-parts appraisal when appropriate. The method depended on the business; he explicitly said no valuation method was perfect (MOI Global, 2012; TheStreet Tyco interview, 2003).

Entry required a discount commensurate with uncertainty. A mediocre or asset-heavy company needed a deeper discount; a rare, durable franchise could justify a smaller one because quality itself reduced some impairment risk. Profitable growth was welcome, but the strategy would not pay an extravagant price for extrapolated growth. A March 1999 fund report, preserved by First Eagle, rejected paying for hypothetical growth during the technology boom (First Eagle historical excerpts, 2022). No primary source reviewed established a universal EV/EBIT band, required discount, or exact purchase trigger, so none is imposed here.

4. Sizing

No public primary source located gives a mechanical position-size formula or a maximum weight. Contemporary reporting says Eveillard disliked large bets and preferred many holdings, consistent with his admission that he could not know in advance which ideas would prove best (Kiplinger, 2007). Confidence in appraisal, business durability, balance-sheet strength, and downside plausibly informed size, but published evidence does not support a numerical rule.

A recurrent numerical trap deserves explicit rejection: the 10–15% figure in the 2009 final interview refers to the share of Japanese companies trading below net cash, not a portfolio position limit (Advisor Perspectives interview, 2009, p. 4). Historical allocation percentages likewise describe named funds at particular dates, not target weights.

5. Portfolio Construction

The portfolio diversified across many securities because valuation is fallible. Cash was a residual: it rose when qualifying opportunities were scarce and fell when bargains appeared, with a reserve sometimes retained for redemptions. That is not identical to a tactical cash target. Yet cash also expressed caution in practice. In 1998 Eveillard admitted experience had made him too cautious, an important reminder that “discipline” and failure to search effectively can look identical from a cash balance alone (Los Angeles Times, 1998).

Gold was intentional insurance rather than an appraisable business. Eveillard viewed it as a currency that could not be printed and protection against extreme inflationary, deflationary, or monetary outcomes. In 2012 he described less than 5% as immaterial and more than roughly 10–12% as becoming more than protection; a 2009 interview used a somewhat wider upper boundary. These were dated judgment ranges, not a timeless allocation algorithm (MOI Global, 2012; Advisor Perspectives, 2009). Gold has no cash flow or measurable intrinsic value, and its insurance premium can be costly. In 1997 Eveillard was prepared to close the Gold Fund if the thesis failed for long enough (SFGate, 1997).

Bonds and currency hedges were opportunistic tools, not fixed strategic buckets. Distressed debt could offer equity-like returns when yields compensated for default risk. Currency hedging sought to protect a security-level thesis from a separate exchange-rate loss, but it could forfeit gains: the fund's roughly 70% yen hedge in 2003 surrendered part of that year's yen appreciation (Washington Post, 2003).

6. Sell Discipline

Selling was conditional, not mechanical. If a price fell while the original analysis remained intact, the loss was temporary and might warrant adding. If the analysis had been wrong or the business deteriorated, the investor had to recognize permanent impairment and sell. Patience was not permission to defend a broken thesis (Graham & Doddsville, pp. 6–7).

Valuation created a second decision. A Graham-style security reaching appraised value normally exhausted its margin of safety. A Buffett-style franchise could remain attractive because intrinsic value might keep compounding, even when the current discount vanished. Eveillard acknowledged the conflict and offered judgment rather than a bright line. The sell decision therefore combined thesis validity, updated value, business durability, and opportunity cost. A public universal threshold was not found.

Risk Management

Risk control began before purchase: conservative appraisal, a discount, legible accounting, sound balance sheets, understandable economics, and diversification. Eveillard rejected leverage because it weakened both margin of safety and staying power. He rejected short selling because even a correct valuation thesis could fail on timing and required forecasting market psychology (Graham & Doddsville, pp. 2, 6). Cash, gold, bonds, and selective currency hedges addressed risks that stock selection alone could not remove.

These protections reduced losses; they did not eliminate them. First Eagle Global Class A lost 22.77% in the year through April 2009 versus 39.33% for MSCI World. In the subsequent six-month slice, bullion and four gold-related equities contributed 4.11 percentage points, while the team remained 30% hedged on the yen and 60% on the euro (SEC-filed semiannual report, 2009). This is evidence of relative defense and insurance working, not absolute capital preservation. Nor did broad diversification eliminate common exposure to valuation, currencies, countries, or the same adverse macro regime.

Temperament and Psychology

Eveillard placed temperament above raw intelligence. The investor had to withstand boredom, ridicule, falling prices, and years of lag without changing identity. A stock could remain flat for four years and double in the fifth; a catalyst requirement would reject exactly the patience that allowed the thesis to mature. “We just need to win over time” described the horizon, not certainty of eventual vindication (Graham & Doddsville, p. 7).

Temperament was also institutional. The late-1990s experience shows that a manager needs unlevered staying power, patient clients, honest communication, and an owner willing to survive relative failure. It also exposes a danger: contrarian identity can harden into excessive caution or confirmation bias. In a February 1996 interview, Eveillard openly said of Apple, “So far I'm wrong,” useful contemporaneous evidence against a frictionless hindsight narrative (Los Angeles Times, 1996).

Evolution Over Career

• 1968–late 1970s: Graham supplied intrinsic value, margin of safety, Mr. Market, and humility. Eveillard encountered Berkshire reports around late 1978.
• 1979–early career: With a small staff, he emphasized adjusted assets, net cash, and other Graham-style situations that were especially available abroad.
• Later management years: As the analyst team grew, the method moved toward Buffett-style work on moats and durable earning power without abandoning price. No verified primary source fixes a precise year for the transition.
• Late 1990s: Refusal to join technology momentum tested the philosophy commercially. The later rebound validated the avoidance of speculative prices, but survival depended on remaining shareholders and a supportive acquirer.
• 2007–09: Credit excess increased attention to leverage, currencies, and what macro events could break a bottom-up thesis. Eveillard returned as part of a team and transitioned to senior adviser in March 2009.
• After 2008: He made the top-down overlay explicit: spend most time on businesses, but monitor regimes capable of permanently impairing them. This was scenario defense, not confidence in precise forecasts (MOI Global, 2012).

What He Explicitly Rejects

• Volatility or short-term tracking error as the definition of risk.
• Leverage that compromises margin of safety and staying power.
• Short selling that makes timing and market psychology decisive.
• Mechanical screens as a substitute for adjusted accounting work.
• Quarterly earnings prediction, precise market forecasts, and near-term catalyst requirements.
• Benchmark hugging and buying what is popular merely to reduce career risk.
• Opaque accounting, suspicious tax economics, dishonest stewardship, and businesses outside the team's understanding.
• Paying for hypothetical growth, momentum, or a narrative unsupported by conservative appraisal.
• Pure country allocation in place of bottom-up industry and company research.

Regimes Where It Thrives Versus Struggles

Tensions Between Stated Philosophy and Actual Behavior

1. Absolute versus relative return. Eveillard rejected short-term benchmark control, yet explicitly wanted to beat benchmarks and peers over time. “Benchmark-agnostic” is directionally useful but incomplete.
2. Bottom-up versus macro. He rejected forecasting, yet cash, gold, currency hedges, credit-cycle caution, and the post-2008 overlay embodied macro judgments. The coherent reading is conditional defense rather than point forecasting.
3. Cash as residual versus caution. Cash arose from scarce bargains, but Eveillard's own 1998 admission shows that high cash could also reflect excessive caution.
4. Management secondary versus disqualifying. He wanted businesses that could survive ordinary managers, but dishonest stewardship and unintelligible accounts were reasons to walk away.
5. Sell at value versus retain quality. Graham favored realization when the discount closed; Buffett-style compounding could justify holding. Eveillard never published a universal reconciliation.
6. Diversification versus conviction. Many holdings reduced single-name error but diluted upside and increased monitoring demands. It was an expression of humility, not proof that concentration was inherently irrational.
7. Capital preservation versus drawdown. The 2008 result was much better than the equity benchmark but still a substantial loss. Permanent impairment is conceptually clean and operationally difficult to diagnose in real time.
8. Personal doctrine versus institutional legacy. First Eagle now formalizes four pillars—security selection, cash, gold, and currency management—and describes 100–150 holdings. Those are successor-team rules. First Eagle announced Eveillard's senior-adviser retirement in 2020, and his current emeritus role has no portfolio authority (First Eagle retirement notice, 2020; current Global Fund guide, 2025).

The durable lesson is not a multiple, allocation band, or number of holdings. It is a system: interpret public facts independently, compress research to the few decisive drivers, insist on a price that admits uncertainty, avoid structures that can force a sale, and build the psychological and institutional endurance to wait. Its weakness is the mirror image of its strength: humility can become over-diversification, patience can become denial, and protection can become costly underinvestment. Eveillard's record matters because the method survived those tensions, not because it solved them.

Jean-Marie Eveillard's public record does not contain an audited trade blotter. The defensible record is a mixture of signed fund reports, later case studies from his book, interviews, transaction documents and market-value snapshots. This chapter therefore ranks seven investments by evidence quality and analytical importance, not by an invented dollar profit. Bank for International Settlements is the single best documented trade: the entry thesis, compulsory exit, legal challenge and final cash payment can all be reconstructed, although the original 1982 cost is still unavailable.

These were fund investments, not Eveillard's personal trades. Charles de Vaulx joined the team in 1987 and was co-manager by 1999, so later cases are attributed to Eveillard, de Vaulx and the First Eagle/SoGen team where appropriate. Current records support treating Eveillard as living but retired: a fund report filed in March 2026 lists him only as Trustee Emeritus, with no legal trustee status, while his current FINRA report says he is not registered and reports no disclosure event (First Eagle Variable Funds report, 2026; FINRA BrokerCheck). No current portfolio is attributed to him.

Measurement and Ranking

Cost and value in a fund schedule produce an unrealized return at one date; they do not prove lifetime proceeds. A takeover premium measures the bid against a reference price, not the fund's return. A fund's total return cannot be assigned to one holding. The ranking preserves those distinctions.

1. Bank for International Settlements - The Single Best Documented Trade

Context and dates. SoGen International acquired privately traded BIS shares in 1982 and held them until the bank compulsorily withdrew all private shares in January 2001. A later case summary, derived from Eveillard's book and approved by him, says the shares were bought at roughly a 60% discount to book value (MOI Global case summaries). The precise weighted cost and dividends were not published.

Thesis and how found. The bank served central banks, carried unusually low commercial risk and held substantial gold. The market discount and dividend supplied value without a visible catalyst. That absence was deliberate: the investment depended on asset value and patience, not a forecast that the discount would close.

Size and structure. The position was ordinary BIS shares across the Global, Overseas and Gold funds. The approved case summary describes BIS as one of Global's largest holdings and 7% of Gold's net asset value near the withdrawal [single-source, book-derived]. No share count by fund was located.

Entry, path and drawdown. There was no public continuous price series or fund-level drawdown. On January 8, 2001, BIS withdrew all 72,648 privately held shares for CHF16,000 each (BIS withdrawal record). First Eagle did not challenge BIS's power to recall the shares; it challenged compensation. Its signed 2002 fund report said CHF16,000 was barely half of net asset value and disclosed the team's disappointment when the partial award selected 70% of NAV rather than full NAV (First Eagle annual report, 2002).

Exit and P&L. The Hague Arbitral Tribunal, administered by the Permanent Court of Arbitration rather than the International Court of Justice, found the recall lawful and compensation should equal per-share NAV less 30% (PCA case record; UN Reports of International Arbitral Awards, partial award). BIS ultimately paid an additional CHF9,052.90, taking total compensation to CHF25,052.90 per share. That is 56.58% above the original CHF16,000 payment, not the investment's total return. The partial award increased reported NAV by 1.6% in Global, 1.3% in Overseas and 0.4% in Gold; dividends and the 1982 cost remain unknown.

What it teaches. Deep value can require legal work when a controlling institution tries to crystallize a discount against minorities. The case also shows intellectual discipline: First Eagle won more money but did not win full NAV, and its own report called the result disappointing. Calling it a complete legal victory would rewrite the record.

2. Shaw Brothers - Twenty Years of Minority-Shareholder Patience

Context and dates. Eveillard found Hong Kong-listed Shaw Brothers in 1986. The approved case summary says SoGen held it for 20 years and earned 15% annually [single-source, book-derived]. Compounding that rate gives about 16.4 times capital, an arithmetic translation rather than a separately reported result.

Thesis and how found. Shaw owned 33% of television broadcaster TVB, a large Chinese-film library and valuable property. Its shares traded at an estimated 40% discount to net asset value. The analytical edge was look-through appraisal of a family holding company in a market that local observers treated as speculative.

Size and structure. Weak minority protections mattered more than confidence in the asset appraisal, so the team capped the position at 1% according to the case summary. That is an unusually useful sizing disclosure: a potentially exceptional return did not justify ignoring jurisdictional risk.

Entry, path and drawdown. No exact 1986 price or continuous drawdown is public. The April 2003 SEC-filed schedule shows why a smooth 15% CAGR should not be imagined: Global carried 21.65 million shares at $21.88 million cost and $19.57 million value, a 10.6% unrealized loss, while Overseas carried 17.44 million shares at $16.48 million cost and $15.76 million value, a 4.4% loss (First Eagle semiannual report, 2003). The same report nevertheless named Shaw as a positive contributor over the six months.

Exit and P&L. The source reports a roughly 2006 endpoint and 15% annualized result, but gives no exit date, proceeds, dividends or dollar P&L. The return is therefore not independently cross-verified and remains explicitly single-source.

What it teaches. Country risk belongs in position size, not only in the discount rate. Shaw also demonstrates the difference between a good long-term result and an easy path: periodic filings can show a loss while the eventual compounded outcome is strong.

3. Buderus - Asset Value Crystallized by Bosch

Context and dates. Buderus, a German heating-systems manufacturer, was among Global's largest holdings at October 2002 and was its largest at April 2003. Bosch announced its public bid in April and acquired control in July; its annual report says ownership reached 97.16% by year-end (Bosch annual report, 2003).

Thesis and how found. Public sources do not preserve Eveillard's original research memo. The defensible inference is that the team appraised an established heating franchise and manufacturing assets below business value; this is consistent with its contemporary asset-oriented process, but the exact valuation and discovery channel are [unverified].

Size and structure. At April 30, 2003, Global held 3.515 million shares worth $115.57 million, 4.43% of the fund, against $58.56 million cost. Overseas held 1.545 million shares worth $50.80 million, 2.76% of that fund, against $26.73 million cost. The two vehicles must not be combined into a personal exposure.

Entry, path and drawdown. The filing does not disclose first purchase or maximum drawdown. The October 2002 report already placed Buderus at 3.92% of Global, so the position predated the takeover. At April 2003 the schedule showed unrealized gains of 97.4% in Global and 90.0% in Overseas. Those are cost-to-market snapshots after trading, not full-cycle returns.

Exit and P&L. The fund report says the Bosch offer drove Buderus sharply higher and that takeovers significantly helped Overseas. It does not state when First Eagle tendered or its realized proceeds. Bosch's acquisition confirms crystallization, but only the filing-date gains above are measurable.

What it teaches. A catalyst can arrive after a value thesis is already well established. Buderus is stronger evidence than a press anecdote because the primary filing supplies shares, cost, market value and fund weight before closing; it is still not a realized-profit ledger.

4. Brau Union and BBAG - A Paired Holding-Company Takeover

Context and dates. The team owned operating brewer Brau Union and parent BBAG when Heineken agreed to acquire the group in 2003. The April fund report says the offer was received and accepted; Austria's Takeover Commission records the later public offer for Brau Union at EUR127.27 per share, open from November 2003 to January 2004 (Austrian Takeover Commission).

Thesis and how found. The two listed layers gave the team exposure to a leading Austrian brewer and its Central European assets through both operating-company and holding-company securities. No primary source states the appraisal or why capital was divided between the two, so a claim that one wrapper was an arbitrage against the other would be [unverified].

Size and structure. Overseas held 185,000 Brau Union shares at $7.21 million cost and $18.19 million value, plus 115,000 BBAG shares at $4.31 million cost and $11.89 million value on April 30, 2003. Combined value was $30.08 million, approximately 1.6% of fund assets [reconstructed from the same filing].

Entry, path and drawdown. Purchase dates and peak-to-trough loss are unavailable. The April marks imply gains of 152.2% on Brau Union, 175.9% on BBAG and 161.0% combined. Because cost incorporates accumulated trading and the schedule preceded the final public-offer period, none is a verified realized return.

Exit and P&L. The regulator reports the EUR127.27 Brau Union offer carried premiums of 68.55%, 89.98% and 95.86% to its three-, six- and twelve-month reference prices. Those are market-reference premiums, not First Eagle returns. The fund report establishes positive contribution and an accepted bid, but not tender dates or final dollar profit.

What it teaches. Holding-company structures can offer multiple routes to the same asset pool, while a strategic buyer can close both discounts at once. The evidence also warns against substituting a bid premium for a manager's P&L.

5. Lindt & Sprungli - From Statistical Cheapness to Quality Compounding

Context and dates. The book-derived case dates the opportunity to about 1991, after management turnover and a roughly 35% share-price decline. It describes Lindt as one of Eveillard's successful calls, but it does not establish that First Eagle held an unchanged position through 2015.

Thesis and how found. Eveillard knew the product and recognized a premium brand, distribution advantage, pricing power, attractive margins and free cash generation. The importance is evolutionary: the case moved beyond Graham-style assets toward a business capable of increasing intrinsic value.

Size and structure. At April 2003, Overseas held 5,500 participation certificates and 350 registered shares. Their combined $4.48 million cost and $5.78 million value represented roughly 0.3% of fund assets. Different rights and prices make it wrong to collapse them into one per-share cost.

Entry, path and drawdown. The initial weighted price and worst fund drawdown are not public. At April 2003 the two securities together were up 29.0% over reported cost, but the participation certificates were up 51.1% while registered shares were up only 6.5%. That divergence shows why a single remembered stock chart is not an account return.

Exit and P&L. No exit or realized P&L was located. The approved case summary says Lindt net profit grew from CHF38 million in 1992 to CHF343 million in 2014 and its registered share rose from CHF2,500 to CHF74,000 by December 2015. Lindt's audited 2014 report independently confirms CHF342.6 million net income (Lindt & Sprungli financial report, 2014). The share-price comparison remains book-derived and is an issuer outcome, not proof of First Eagle's holding-period return.

What it teaches. Paying for quality can still be value investing when durable economics expand appraisal. The data boundary is equally instructive: a company compounded superbly, but without tax lots and an exit, the fund's exact gain cannot be claimed.

6. Gold Miners to Bullion and Linked Notes - A Successful but Costly Campaign

Context and dates. Eveillard launched the Gold Fund in 1993. It struggled badly: by March 1998 it was down 17.6% since inception and he publicly considered closing it (Fortune, 1998). De Vaulx later recalled persuading him in late 2001 that mining shares were too expensive and the prospectus should allow bullion.

Thesis and how found. Gold was insurance against disorder and monetary error, not a cash-flow asset. The refinement was security selection within that hedge: when miners embedded costly operating leverage, physical bullion and linked notes could express the macro insurance more cleanly.

Size and structure. In fiscal 2002, Global's gold-related stocks contributed a little over three percentage points. At April 2003, Gold held 15.4% in bullion and 12.6% in gold-linked notes with 1.5-to-1 exposure. Its 92,488 bullion ounces had $31.94 million cost and $31.31 million value, a 2.0% unrealized loss at that date. This was a multi-security campaign, not one entry.

Entry, path and drawdown. The path included years of disappointment, central-bank selling and expensive mining equities. In the year ended October 2002, gold rose 12.41%, while Gold Fund Class A rose 70.70% and its gold-mining index rose 27.30%. The fund's larger result included miners, stock selection and leverage and cannot be assigned to bullion. A contemporary account says the fund had lost 9.1% annualized from end-1993 to mid-1998 before gold later rallied from roughly $250 to $350 (Washington Post, 2003).

Exit and P&L. There was no clean exit before Eveillard's December 2004 retirement, and no security-level lifetime P&L. The April 2003 report says Gold rose 12.86% over six months versus 3.22% for its mining index because bullion and notes outperformed expensive miners. That is the best evidence for the portfolio-design decision, not a closed-trade return.

What it teaches. A sound hedge thesis can begin as a poor investment, and the chosen security can matter as much as the asset class. Willingness to abandon expensive miners for bullion improved implementation; willingness to consider closing the fund shows patience was not unlimited.

7. Tyco - Buying Both Equity and Distressed Credit Through an Accounting Crisis

Context and dates. Eveillard and de Vaulx began buying Tyco common near $15 during the 2002 governance and accounting collapse; it subsequently fell to about $8. A March 2003 interview put the average common cost at $13 and contemporaneous appraisal near $25 (TheStreet interview, 2003).

Thesis and how found. Familiarity with Tyco's connector, security and health-care businesses let the team value major divisions separately. The analyst rebuilt the income statement, included refinancing costs omitted by zero-coupon accounting and deducted debt from gross asset value. This was not faith in reported earnings; it was a claim that good cash-generating assets survived bad promotion and governance.

Size and structure. At April 2003, Global held 2.475 million shares plus $25.5 million face value of two Tyco bonds. Common and bonds together were 2.46% of the fund. The stock carried $37.06 million cost and $38.61 million value; the bonds carried $21.04 million cost and $25.65 million value.

Entry, path and drawdown. From the reported $15 starting area to $8, the observed common-stock decline was about 47%; it is not the fund's exact drawdown because buying continued. The team knew accounting was aggressive, yet the subsequent discovery of additional irregularities could still have invalidated its reconstruction. The 2002 annual report said Tyco was down and that intrinsic-value estimates had been reduced.

Exit and P&L. At April 2003, the common showed a 4.2% unrealized gain, the bonds 21.9%, and the combined position 10.6%, calculated from the SEC-filed costs and values. By October, the common traded near $20, about 53.8% over the stated $13 average cost, and the managers still appraised it at $23-$24 (Washington Post, 2003). No complete exit or realized P&L was located, so Tyco ranks last despite its rich process evidence.

What it teaches. Capital-structure breadth creates optionality: the bonds monetized improving solvency before the equity fully recovered. Tyco also marks the dangerous edge of contrarianism—an investor can adjust bad accounts and still underestimate fraud, refinancing needs or governance damage.

What the Ranking Excludes

• Mexican bonds after the 1994 peso crisis were profitable according to the 2003 Washington Post profile, but no instrument, currency, size, entry, exit or P&L was located. Labeling them Tesobonos would be conjecture.
• Hilti had a documented CHF1,150 tender and a team appraisal of CHF1,500-1,600, while the 2003 filing says its going-private transaction helped Overseas. The original cost and realized return remain missing, so it does not displace a measurable case.
• The 2007-2009 Japan campaign included Fanuc, SMC, Keyence and overcapitalized insurers, but the successor boundary is too important: Eveillard returned in 2007, transferred oversight in 2008 and left management in March 2009. Later outcomes cannot be assigned to him.
• McDonald's, Swissair and Secom belong in mistakes or mixed outcomes. A famous thesis is not automatically a greatest trade.

Skill, Luck, Attribution and Confidence

The repeated skill is visible across countries and security types: decompose assets, insist on a discount, size for legal and governance risk, tolerate a long wait, and use corporate action without requiring it at entry. BIS and Shaw show patience; Buderus and Brau Union show strategic buyers validating appraisal; Lindt shows the Buffett-style extension into quality; gold and Tyco show implementation across commodities and credit.

Luck and selection bias remain material. Five of seven cases benefited from a compulsory repurchase, takeover or later control transaction. Public archives favor remembered winners, while the same method produced Swissair. Team attribution also matters: post-1999 filings cover Eveillard and de Vaulx together, and fund returns include many securities, cash, currencies and expenses.

The conclusion is deliberately narrower than a league table of profits. BIS is the best documented full-cycle realization, Shaw has the strongest disclosed long-horizon return but only a book-derived source, and Buderus has the clearest primary cost-and-value evidence. For every case, absolute dollar P&L is unavailable. That limitation is not a defect to conceal; it is the central fact needed to read Eveillard's greatest trades honestly.

Jean-Marie Eveillard's record is famous for avoiding ruin, but the useful lesson is not that value discipline prevented mistakes. It is that several different kinds of error—bad securities, costly omissions, client-flight risk and temporary marks—must not be collapsed into one loss number. The public record is unusually incomplete at the trade level: no full tax-lot ledger or audited security-attribution history was located. This chapter therefore reports exact figures only when a fund filing or attributable account supplies them and labels estimates, unrealized marks and book-derived recollections.

As of 2026-07-20, Eveillard is living and retired. A March 2026 fund filing lists him as Trustee Emeritus without portfolio authority, and FINRA reports him as unregistered with no disclosure event; neither source is universal legal clearance (First Eagle Variable Funds report, 2026; FINRA BrokerCheck). A Delaware court dismissed a 2025 accounting-disclosure suit against First Eagle entities with prejudice on July 9, 2026; Eveillard was not a defendant (Dandini v. First Eagle Funds, 2026). First Eagle and its distributor separately paid nearly $40 million in a 2015 SEC settlement covering improper distribution payments and disclosures from January 2008 through March 2014. The SEC named the firms, not Eveillard; most of the period followed his March 2009 management exit, so it would be wrong to convert the firm finding into a personal investment mistake (SEC, 2015).

Loss map

Major losses

Swissair: assets without durable economics

The clearest realized security loss was Swissair. In the early 1990s, SoGen International Fund bought an apparent asset bargain: a young aircraft fleet, Swissotel properties and hidden balance-sheet reserves after 1990–1992 losses depressed the shares. Eveillard read recession, the Gulf War's tourism shock and higher oil prices as temporary. The error was treating a cyclical explanation and attractive assets as sufficient when the operating franchise was becoming structurally weaker (MOI Global case summary).

The thesis then broke on several fronts. Switzerland's 1992 rejection of closer European integration limited route access. Management pursued a “hunter strategy,” buying minority stakes in other airlines and unrelated services. The Eveillard-approved, book-derived case says that from 1996 to 2000 staff rose from 36,000 to 79,000, liabilities from CHF9.7 billion to CHF19.1 billion, equity fell from CHF2.1 billion to CHF1.2 billion, and CHF700 million of profit became a CHF2.9 billion loss. It also alleges that consolidation choices and undeclared liabilities obscured deterioration. These figures have no second security-level source in the located record and remain single-source. The terminal event is independently confirmed: Switzerland's federal audit office says the fleet was grounded for lack of liquidity in October 2001, prompting a CHF1.45 billion state loan, CHF1.15 billion of which was paid out (Swiss Federal Audit Office).

The same approved account says Eveillard sold the 0.1% position in 2001 at an 80% loss. If both percentages refer to the same capital base, the mechanical fund-level hit was about eight basis points; that is an inference, not disclosed attribution. The security outcome was severe, while position sizing contained the portfolio damage.

The behavioral root was a cluster of false comforts: a national-champion halo, faith in supposedly conservative “Germanic” accounting, and asset value detached from cash burn and leverage. Eveillard's stated lessons were correspondingly concrete: distrust cultural accounting reputations, treat leverage as especially dangerous in cyclicals, and recognize airlines' overcapacity, labor, pension and political-route burdens. Swissair is the best counterexample to the legend that a low price and hard assets automatically create a margin of safety.

The English broadcaster: an unpriced political option

An earlier mistake involved an unnamed English television broadcaster. A summary of Michael Sincere's interview with Eveillard says the stock performed well for several years, then collapsed when the prime minister did not renew its broadcast license. Eveillard's lesson was not to assume that a foreign business shares the legal and regulatory characteristics of its U.S. analogue; overseas investing required more local research (Capital Ideas interview excerpt).

This is credible as an attributed lesson but thin as a case study. The company, purchase date, license decision, position size, cost and realized loss were not identified in the accessible source. It should not be assigned a fabricated P&L or retrofitted to a broadcaster by guesswork. What is supportable is the analytical failure: the team appraised current cash flows without correctly valuing a binary political concession.

Gold Fund: an asset-class thesis implemented through the wrong instrument

Eveillard launched the Gold Fund in August 1993, but patience nearly exhausted itself before the thesis worked. Fortune reported that the fund was down 17.6% from inception through March 1998 while the S&P 500 had gained 134.5%; Eveillard was prepared to seek closure if gold did not improve during 1998 (Fortune, 1998). A later account gives a -9.1% annualized return from end-1993 to mid-1998, a different interval and return convention that should not be forced into reconciliation (Washington Post, 2003). In January 1997 he said that, if the thesis remained unproductive, he would tell shareholders he had made a mistake and recommend redemption (SFGate, 1997).

This episode mixed thesis error, timing and vehicle design. Gold itself had no cash flow or measurable intrinsic value, while mining equities added operating, political and capital-allocation risks. Charles de Vaulx later persuaded the team that miners were too expensive; in late 2001 the prospectus was changed to permit bullion. The subsequent recovery—Gold Fund Class A rose 70.70% in fiscal 2002—does not retroactively erase five years of poor results (First Eagle annual report, 2002). It does show a genuine process correction: separate the desired monetary hedge from the economics of the companies used to express it.

2007–2009: relative defense still meant a large absolute loss

First Eagle's current historical chart records a 30.95% Global Fund loss from November 2007 through February 2009, compared with -49.63% for the S&P 500 and -53.22% for MSCI World (First Eagle Global Fund investor guide, March 2026). A 30.95% decline requires a 44.82% gain merely to restore the starting capital. Cash, gold and low financial exposure worked relatively; they did not deliver absolute preservation.

American Express shows the distinction between a temporary mark and a proven permanent loss. At April 30, 2009, Global owned 9.253 million shares with $397.95 million of cost and $233.36 million of value, a $164.58 million or 41.36% unrealized decline calculated from the SEC-filed schedule (First Eagle semiannual report, 2009). Eveillard acknowledged the mark but argued that expected future earnings had barely changed. His defense was revealing: the team anticipated zero profitability and nonperforming loans above 10%, but did not anticipate that the market had failed to see the same risk (Advisor Perspectives, 2009). That is not evidence of a realized loss, and later recovery would not prove that the interim risk was imaginary. It is evidence that correct fundamental stress assumptions can coexist with poor entry timing and a much larger mark than intended.

Attribution must also remain bounded. Eveillard returned in March 2007 after Charles de Vaulx's departure, but oversight moved toward Matthew McLennan during 2008 and Eveillard transitioned to senior adviser in March 2009. The drawdown and American Express position therefore belong to the Global Value team and fund, not to a solo personal account.

The deeper admission came later: Eveillard said value investors missed the crisis because they ignored top-down mortgage and credit excesses. His process change was not to become a macro forecaster. It was to retain bottom-up appraisal while monitoring regime-level threats capable of invalidating many company valuations at once (MOI Global interview, 2012).

Errors of omission

Freezing after Black Monday. Eveillard sold nothing during the October 1987 crash, but he also did not buy that day. Once he understood portfolio-insurance selling, the fund bought modestly; decades later he said it bought less than it should have. This is a clean error of omission caused by uncertainty and fear, but no defensible foregone-profit figure exists (Jason Zweig, 2017).

Excess caution in the late 1990s. In August 1998 Eveillard conceded that experience had become a handicap and that he had probably been too cautious for two or three years (Los Angeles Times, 1998). Yet calling the entire refusal to buy technology a mistake would be outcome-biased. Global lost only 0.26% in 1998 while the S&P 500 gained 28.58%, then gained 23.75% during the March 2000–September 2002 technology bust while the S&P lost 38.84%. The investable error was not avoiding unpriceable securities; it was allowing caution, cash and the client proposition to become so difficult to distinguish that shareholders could not stay.

Currency hedging. In 2003 the team hedged roughly 70% of its yen exposure and forfeited part of the currency's 8.5% gain. The hedge reduced one risk but created an opportunity cost when the insured-against outcome did not occur (Washington Post, 2003). This is not proof that hedging was irrational; it is a reminder that insurance must be judged over scenarios, not only by the premium paid in one period.

Selling and holding errors are underdocumented. Eveillard's operating rule was sound: a price decline with intact business value could be temporary, but a broken analysis or deteriorating model required selling. He summarized permanent impairment as “Damn it, I made a mistake” and said the investor should move on (Graham & Doddsville, 2007/08). The archive does not establish how consistently he applied that rule across all holdings. Public profiles disproportionately preserve Swissair, celebrated winners and positions that later recovered; silent exits and stale capital are the missing denominator.

Near-death moments

The late-1990s technology boom was the strategy's principal commercial near-death. Contemporary reporting put redemptions at roughly $100–$200 million per month before easing to about $50 million. Another account shows assets falling from $4.4 billion in October 1997 to $1.46 billion in September 2001, though that combines flows and returns and uses a later trough. Eveillard later said seven of ten shareholders left between autumn 1997 and spring 2000 (Forbes, 1999; Financial Advisor, 2002; MOI Global, 2012). Société Générale chose to sell the operation, and the eventual First Eagle ownership preserved the strategy.

This was not portfolio insolvency: no gate, leverage spiral, forced liquidation or fund closure was located. It was an open-end business-model crisis. The behavioral root was partly admirable—refusal to imitate a mania—and partly managerial: a long-horizon process depended on clients whose tolerance was shorter than the holdings' realization period. The episode's later vindication can obscure the counterfactual. Had the 1999 sale produced an owner unwilling to tolerate the style, Eveillard could have been right about technology and still lost the institutional platform needed to realize that judgment.

The Gold Fund supplied a second, explicit closure risk. Here the driver was not shareholder flight alone but Eveillard's willingness to admit that a five-year experiment might have failed. Its survival owed something to time, a changed implementation and a favorable gold cycle. That mixture is evidence of adaptation, not pure forecasting skill.

What changed—and what did not

The constant was the distinction between quotation loss and impairment. That framework encouraged patience and prevented forced imitation, but it could also rationalize stale positions if the analyst kept declaring the thesis intact. The correction was always judgmental, not algorithmic.

Assessment: skill, luck and the lesson of survivorship

Eveillard's loss record supports three conclusions. First, diversification and small sizing were real skills: an 80% Swissair loss was survivable because the position was reportedly only 0.1%. Second, the method's defenses were relative, not absolute. A 30.95% crisis drawdown and a 41.36% American Express mark are economically meaningful even when benchmarks did worse. Third, several recoveries required favorable external events and enough institutional patience to wait. Gold's rebound, the technology bust and a tolerant acquirer were not wholly under the manager's control.

The strongest evidence of skill is therefore not a claim of mistake-free capital preservation. It is the willingness to identify different error classes, contain individual damage and change a process without abandoning its core. The strongest criticism is symmetrical: the public canon remembers the survivals, while the unreported losing tax lots, foregone returns and client costs remain unknowable. A rigorous reading treats Swissair as a genuine analytical failure, 1987 and the late 1990s as omissions, 2008 as a team drawdown with partial defense, and later vindication as neither exoneration nor proof of foresight.

This is a provenance-first quotation file, not a list of internet aphorisms. The 40 quotations below are each 25 words or fewer, retain the source's wording, and come from an interview, speech report, transcript or contemporaneous publication that explicitly attributes the words to Jean-Marie Eveillard. Repetition across sources is minimized. The dates identify when the words were spoken or published, not when a later webpage was posted.

Eveillard remains living and retired as of 2026-07-20. A March 2026 filing lists him only as Trustee Emeritus; his current FINRA report says he is not registered and reports no disclosure event (First Eagle Variable Funds report, 2026; FINRA BrokerCheck). Those records do not turn current First Eagle positions, policies or statements into his words.

Value, Price and Business Quality

1. “Value investing is a big tent that accommodates many different people.”

   — Graham & Doddsville interview, 2007/08 (Columbia transcript). The metaphor accommodates Graham-style assets and Buffett-style franchises without pretending they are identical.

2. “The idea of margin of safety, the idea of intrinsic value, the idea of Mr. Market, the very humble idea that the future is uncertain.”

   — Graham & Doddsville, 2007/08. This 25-word sequence is the compact core of his Graham conversion.

3. “A business can have value even if it is not growing.”

   — Graham & Doddsville, 2007/08. Growth is neither a prerequisite for value nor a reason to ignore price.

4. “Value investors tend to think like private equity investors.”

   — Graham & Doddsville, 2007/08. His comparison is about appraising the entire business, not using private-equity leverage.

5. “We look for stocks selling at a significant discount.”

   — TheStreet interview, 2003 (transcript). The discount is relative to a reconstructed business value, not a low market multiple alone.

6. “If we are less certain, then we want to see a greater discount.”

   — TheStreet, 2003. The margin of safety expands when appraisal confidence falls.

7. “You look at businesses, you try to figure out what they’re worth.”

   — Forbes video transcript, 2009 (transcript). This is a business-owner definition, not a forecast of the next quarter.

8. “There is no perfect way to place a value on a business.”

   — MOI Global interview, recorded 2012 (edited transcript). His preferred tools were aids to judgment, not precision machines.

9. “We like EV/EBIT because it introduces the balance sheet into the picture.”

   — MOI Global, 2012. The statement explains a preference, not a timeless or universal buy threshold.

10. “With Buffett, the true margin of safety may be more associated with the perception of the quality of the business.”

    — MOI Global, 2012. Eveillard was describing the Buffett side of the value tent, where durable economics can protect value.

Research, Accounting and Interpretation

11. “The work, of course, starts with public information — running numbers.”

    — Graham & Doddsville, 2007/08. Public evidence precedes management access.

12. “The important numbers have more or less to fit on a single page or two pages at the most.”

    — Graham & Doddsville, 2007/08. Compression follows analysis; it does not excuse skipping the footnotes.

13. “For a value investor the devil is in the details.”

    — Graham & Doddsville, 2007/08. His simple framework demanded detailed accounting work.

14. “We don’t do screens because we like to check the accounting carefully and make our own adjustments.”

    — Graham & Doddsville, 2007/08. Historical-cost assets and reporting choices could defeat a mechanical screen.

15. “Wall Street is nothing but a vast promotion machine.”

    — TheStreet, 2003. The hyperbole accompanied a Tyco case in which his analyst rebuilt the accounts rather than trusting promotion.

16. “When it becomes more confusing, we avoid the company.”

    — TheStreet, 2003. Incomprehensibility was a reason to pass, not an invitation to invent assumptions.

17. “And it was incomprehensible on purpose.”

    — TheStreet, 2003, on Enron’s off-balance-sheet footnotes. This is an attributed judgment, not proof that every complex filing is deceptive.

18. “What I love best about Graham and Buffett is the simplicity of their arguments.”

    — Ivey Business School talk, 2014 (official report). Ivey also records his advice to reduce a business to a few decisive strengths and weaknesses.

Risk, Loss and Portfolio Defense

19. “Borrowing works both ways.”

    — Graham & Doddsville, 2007/08. Leverage magnifies error as readily as insight.

20. “Borrowing reduces your staying power.”

    — Graham & Doddsville, 2007/08. This is the institutional reason leverage conflicts with a long realization period.

21. “Risk to us is absolutely not volatility.”

    — Graham & Doddsville, 2007/08. The alternative was permanent impairment; the 2008 fund drawdown shows this definition did not mean an absence of pain.

22. “Permanent impairment of capital, which are fancy words for ‘Damn it, I made a mistake.’”

    — Graham & Doddsville, 2007/08. The phrase distinguishes a broken thesis from a lower quotation.

23. “You have to acknowledge your mistake, sell at a loss, and move on.”

    — Graham & Doddsville, 2007/08. Patience was conditional on the business and analysis remaining intact.

24. “Did I miss something?”

    — Graham & Doddsville, 2007/08. This is the first question after an adverse price move, not a presumption that the market is wrong.

25. “We’re beginning to ask if there’s something we missed.”

    — TheStreet, 2003, discussing McDonald’s. The remark is useful contrary evidence against portraying him as automatically doubling down.

26. “I believe a small position in gold always makes sense.”

    — TheStreet, 2003. He immediately framed gold as insurance rather than a normal cash-flow investment.

27. “Gold is the only currency that cannot be printed.”

    — MOI Global, 2012. This explains his monetary-insurance thesis; it is not proof that gold has an intrinsic value.

28. “Cash is a residual.”

    — MOI Global, 2012. Cash rose when securities failed the valuation test, subject to liquidity needs; it was not a fixed macro allocation.

29. “Money is not supposed to be free.”

    — MOI Global, 2012. The remark anchors his concern that near-zero rates distorted prices and credit behavior.

30. “A credit burst would eventually follow a credit boom, just like night follows day.”

    — MOI Global, 2012. The wording is categorical; his own admission that value investors missed 2008 limits any claim of precise foresight.

Patience, Temperament and Clients

31. “If you are a value investor, you are a long-term investor.”

    — Graham & Doddsville, 2007/08. The horizon separates appraisal from short-term market psychology.

32. “We don’t need to win every day. We just need to win over time.”

    — Graham & Doddsville, 2007/08. This is an endurance rule, not permission to ignore permanent deterioration.

33. “One is not entitled to be happy every day.”

    — Graham & Doddsville, 2007/08. Eveillard connects emotional discomfort to the conditions under which contrarian investing works.

34. “To lag is to suffer. Suffer psychologically.”

    — MOI Global, 2012. Relative underperformance imposed personal, commercial and career costs even when the portfolio thesis survived.

35. “After one year, investors in our funds were upset, after two years they were mad, after three years they were gone.”

    — MOI Global, 2012. This is his retrospective description of the technology-boom redemption cycle.

36. “It takes perseverance to be a successful investor over the long term.”

    — Ivey, 2014. Perseverance is presented as a behavioral requirement, not a substitute for research.

37. “Value investing works over time and it makes sense.”

    — Ivey, 2014. The claim expresses his conviction; his surviving record alone cannot eliminate manager-selection or survivorship bias.

38. “Our time horizon for a security is five years, not five months.”

    — The Wall Street Transcript, 2002 (interview page). The line is a horizon, not a mandatory holding period.

39. “I was helped more by what I did not own than what I did own.”

    — MOI Global Asian Investing Summit, 2014 (selected transcript quotations). Avoiding late-1980s Japan, late-1990s TMT and early-2000s financials also means the archive highlights successful omissions.

40. “Join a value shop.”

    — Graham & Doddsville, 2007/08. His shortest career advice assumes the institution can tolerate the method’s inevitable periods of lag.

Annotated Index of Primary and Near-Primary Materials

The archive is rich in interviews and poor in personal letters. No public, systematic archive of shareholder letters signed by Eveillard alone was located. Fund reports were corporate documents, often team-authored or signed in an officer capacity. They are indexed separately so institutional prose is not silently converted into personal authorship.

Authored books and fund materials

• 2016 French book; 2017/18 German edition — Value Investing Makes Sense. The lawful German sample includes an Eveillard-authored preface, memoir framing and table of contents; it is the closest thing to a sustained first-person primary text, but quotations in this chapter remain in verified English sources rather than back-translating the German.
• March 1999 — SoGen Funds annual-report passage. First Eagle’s official historical excerpt preserves a warning against overpaying for hypothetical growth. The archive credits “SoGen Funds, Inc.,” not Eveillard alone, so it is evidence of team doctrine rather than an individual quotation.
• 2000 and 2002 — First Eagle/SoGen annual reports. Reports include portfolio commentary and officer signatures, but multiple managers and corporate authorship make them primary fund records, not a personal-letter series.

Major interviews and conversations

• March 2002 — The Wall Street Transcript. A direct manager interview on the funds, long horizon and then-current process; only the public excerpt is indexed because the full transcript is paywalled.
• March 2003 — TheStreet, “10 Questions.” A long, freely accessible Q&A on valuation uncertainty, Tyco accounting, McDonald’s, gold and Japan; unusually useful because questions and answers are clearly separated.
• 2007 — Financial Advisor, “The World According to Eveillard.” The profile/interview captures credit-boom anxiety, opaque banks and leverage shortly after his return to management.
• Winter 2007/08, updated October 2008 — Columbia Graham & Doddsville. The most comprehensive public English interview: intellectual history, research workflow, risk, temperament, client redemptions and career advice. The PDF was text-extracted and visually checked for layout and speaker attribution during this task.
• March 31, 2009 — Advisor Perspectives, final interview. The six-page PDF covers diversification, gold, currency hedges, Japan and American Express. It interviews Eveillard together with Abhay Deshpande and does not label each response by speaker; it is therefore indexed but excluded from the individual quote corpus above.
• September 2009 — Forbes, Intelligent Investing. The direct video transcript defines value investing and discusses shorting, Japan and macro awareness. A Forbes briefing book preserves part of the interview even though the article endpoint may return a paywall response.
• 29 November 2011 — CFA Institute, “Conversation with a Veteran Value Investor.” The official multimedia page confirms the event and topic but exposes no transcript; it is an audiovisual primary record, not a text-quotation source here.
• Recorded 2012 — MOI Global, “Art of Global Value Investing.” The edited transcript covers the post-2008 top-down overlay, gold, cash, EV/EBIT, sell decisions, client psychology and Europe. MOI states explicitly that the recording occurred in New York in 2012 despite the page’s 2017 publication date.

Speeches, classes and conference sessions

• 18 October 2005 — 15th Graham & Dodd Breakfast. Columbia’s official recap places Eveillard between Graham and Buffett and summarizes his international-accounting cautions; it is a reported speech, not a full transcript.
• 13 March 2014 — Ivey value-investing class. The school’s report preserves several direct quotations on benchmark lag, research simplification, perseverance and the late-1990s client exodus.
• 1 April 2014 — MOI Asian Investing Summit. Selected direct quotations cover Korea, Japan, India, Russia, Bitcoin and the importance of what he avoided; the full session is access-controlled.
• 7 April 2015 — MOI Asian Investing Summit. Selected direct quotations cover Tokyo, China, India, the Philippines, Indonesia, Brazil and commodity-linked markets; useful for dated views, not timeless allocation rules.
• 1 June 2016 — London Value Investor Conference. Public notes preserve themes but not a speaker-verified transcript, so apparent quotations require caution and are omitted from the corpus.
• 9 August 2017 — CFA Society Chicago. The society recap covers Graham, Buffett, Hayek, accounting and the Enron footnotes; it is an official event report, not verbatim proceedings.

Podcasts and recorded interviews

• Fall 2007 — Morningstar conversation with Martin Whitman. The archived video index identifies seven segments, including “No Pain, No Gain” and redemption-period lessons; the surviving page does not provide a reliable transcript.
• 12 July 2019 — Value Investing with Legends, episode 6. The one-hour Apple episode supplies timestamps for Graham, Buffett, Lindt, humility, client management, BIS, macro awareness and accounting. Because the page offers metadata rather than a verified transcript, it is indexed but not mined for quotation text.

Provenance Boundaries, Criticism and Current Context

Three boundaries keep this archive honest. First, an interviewer’s paraphrase is not put in quotation marks. Second, a fund’s “we” may represent Eveillard, Charles de Vaulx, Abhay Deshpande, Matthew McLennan or an institutional team; the joint 2009 interview demonstrates why attribution cannot be assumed. Third, a later compilation being “approved by” Eveillard makes it valuable secondary evidence, not a transcript or signed original.

The words also deserve resistance, not reverence. “Risk is not volatility” is analytically useful but can obscure the real client harm of drawdowns and withdrawals. Patience can become denial; the archive does not expose every stale holding or losing tax lot. Gold and macro remarks were scenario judgments whose success varied by interval. The record overrepresents celebrated omissions and eventual recoveries. Team members, fund ownership and favorable market reversals helped the method survive, while public quotations cannot allocate luck versus skill.

Finally, the current legal boundary remains unchanged. First Eagle and its distributor settled a firm-level SEC matter in 2015; the release did not name Eveillard (SEC settlement). A Delaware court dismissed Dandini v. First Eagle Funds with prejudice on July 9, 2026; Eveillard was not a defendant (Delaware Superior Court). First Eagle announced his senior-adviser retirement in 2020 and described later portfolio oversight as belonging to his successors (retirement notice). None of those developments changes the historical words above, and none supports attributing present-day First Eagle views to him.

Research task: F — key writings As of: 2026-07-20

Corpus verdict and reading rules

Jean-Marie Eveillard has one verified, conventional investment book: Value Investing Makes Sense, also issued as En Bourse, investissez dans la valeur ! and in translated editions. No second Eveillard-authored investing book, academic paper, systematic personally signed shareholder-letter archive, or stable corpus of bylined articles was located. That scarcity is substantive: interviews and fund reports are often presented online as “his writings,” but direct speech, officer signatures and institutional “we” do not establish individual authorship.

The present record supports treating Eveillard as living and retired. A fund report filed in March 2026 lists him only as Trustee Emeritus and says that status is not a legal trusteeship; it does not give him current portfolio authority (First Eagle Variable Funds report, 2026). FINRA's current individual report is unusually useful bibliographic evidence: it records “Author of book Value Investing Makes Sense,” with author activity beginning April 15, 2015, while also warning that employment data need not update after registration ends (FINRA BrokerCheck). Current First Eagle publications and portfolios therefore belong to successor teams, not to an implied continuing author-manager.

The sole core work by Eveillard

1. Value Investing Makes Sense / En Bourse, investissez dans la valeur ! (copyright 2016; editions from 2016–18)

Access, editions and authorship. Valor Editions identifies Eveillard as the sole author and gives the French print edition's 2017 publication date, 152 pages, ISBN 9782361170233 and full contents (French publisher). The lawful German preview supplies the strongest open rights-page evidence: it says the work was first published by Valor in 2016, names Sascha Mattke as translator, and gives the German print and electronic ISBNs (German publisher preview). The Spanish publisher lists a 143-page translation with the same 13-chapter and four-appendix architecture (Spanish edition). Page counts and dates vary by language and format; these are editions of one work, not three separate books.

Central thesis. Value investing is logically sound because a security can trade below a conservatively appraised business value, and it works over sufficiently long horizons because fear, fashion, incentives and short time horizons repeatedly create those gaps. The method still requires judgment, patience, skill, institutional staying power and luck. Eveillard presents that thesis as memoir and casebook rather than as a formal valuation manual.

Ten key ideas.

1. Graham and Buffett define a continuum, not opposing camps. Graham's asset and normalized-earnings discipline protects against overpayment; Buffett's qualitative extension recognizes that a durable franchise can expand intrinsic value.
2. A margin of safety compensates for an unknowable future. Valuation is an estimate, so greater uncertainty should require a larger discount rather than a more elaborate forecast.
3. Accounting interpretation precedes a multiple. Hidden assets, leverage, pension burdens, capitalized costs and opaque footnotes can make a superficially cheap security expensive or unintelligible.
4. Special situations widen the opportunity set. Preferred shares, holding-company discounts, spin-offs, closed-end funds and high-yield bonds can offer value outside ordinary common-stock selection.
5. Avoidance contributes as much as selection. The late-1980s Japanese bubble, late-1990s technology mania and pre-2008 financial excess illustrate that not owning an unappraisable security can be a positive decision.
6. Benchmarking creates a commercial fragility. A sound long-horizon process may lag for years, while open-end clients can redeem before the thesis works and force sales at the worst time.
7. Gold is insurance, not an appraisable business. Eveillard treats bullion as protection against monetary disorder and distinguishes it from mining companies, whose operations, costs and capital allocation can defeat the hedge.
8. Country familiarity is not risk analysis. BIS, Shaw Brothers, Legrand and Kohler show that legal rights, controllers and minority treatment matter; a discount is not protection if the holder cannot realize fair value.
9. Quality cannot excuse price, and assets cannot excuse bad economics. Lindt and Shimano illustrate the Buffett extension; Swissair shows how leverage and structural deterioration can destroy apparent asset value.
10. Humility is operational. Diversification, cash, rejection of leverage and willingness to sell a broken thesis are acknowledgments that the investor will be wrong, not decorative personality traits.

Best chapters and sections. Page locators below follow the 192-page German edition because its full contents are publicly inspectable; chapter numbers transfer across the French and Spanish editions, but page numbers do not.

• Read first: Chapter 2, “Value Investing Makes Sense — From Graham to Buffett” (German pp. 45–72). This is the conceptual center: Graham, Buffett, value traps, moats, valuation, international complications and the tension between owning quality and selling at value.
• Then Chapter 4, “Value Investing Works Over Time” (pp. 81–92). It supplies the missing institutional mechanism—why few managers persist, why lag hurts, and why omissions shaped long-run results. The late-1990s account should be read as retrospective testimony, not as proof that every contrarian position eventually wins.
• Use Chapter 3, “Special Situations” (pp. 75–78), as a map rather than a manual. It is the shortest practically useful section. A contemporary review praises the coverage but calls it too brief; the same review says the book can read like an interview transcript and needed tighter editing (Strictly Value review, 2016).
• Pair Chapter 5, “On Gold” (pp. 95–108), with Chapter 10, “Swissair and Other Lessons” (pp. 139–46). The pairing prevents defense from becoming dogma: gold includes timing and instrument-choice errors, while Swissair is the clearest case of assets, leverage and governance defeating a cheapness narrative.
• Read Chapters 6–9 (pp. 111–38) as comparative cases. BIS and Kohler are Graham-style legal/asset situations; Lindt is the Buffett-style franchise; Shimano and Shaw add Japanese governance and Hong Kong minority-risk context. The point is not to copy old securities but to see how the type of uncertainty changes the required discount and size.
• Finish with “Last Words” (pp. 159–63) and Appendices A–D (pp. 166–75). These compress margin of safety, pain, leverage, diversification, cash and gold, then add Legrand, benchmark tyranny and the reported long-term result. Appendix C is a manager's selected presentation, not an independently examined personal-account record.

Evidence and criticism. The book's strength is unusual candor about Swissair, client flight, timing and colleagues; its weakness is retrospective selection. Remembered winners, avoided bubbles and later vindication receive more space than the denominator of ordinary holdings and silent exits. Case returns often lack tax lots, full drawdowns or audited security attribution. The fund record is also collaborative, particularly with Charles de Vaulx and later teams. The book is best read as Eveillard's explanation of a practice, not as a complete trade ledger or proof that the stated rule was followed consistently.

Authorship traps: what is not a second Eveillard work

• Fund reports are institutional records. First Eagle's archive reproduces a March 1999 SoGen Funds passage under the vehicle's name, not Eveillard's individual byline (First Eagle historical excerpt). A report signed in an officer role can establish adoption or responsibility without proving solo drafting.
• Approved compilations remain compilations. MOI Global says its extensive case-and-quotation resource was authored by research associate Alex Gilchrist and read and approved by Eveillard (MOI Global compilation). Approval raises its evidentiary value; it does not transfer the byline.
• Interviews, talks and podcasts are direct voice, not authored prose. They are primary evidence for ideas but should not inflate the bibliography.
• Translations and the German-edition foreword are parts of the same book. Different titles, ISBNs and pagination do not create additional works. Retailer listings that count languages as separate “works” are bibliographically misleading.
• Endorsements are not forewords. A cover quotation for another author's book shows recommendation, not authorship or a substantive contribution.

Best primary companions, ranked

These items are not works “by” Eveillard in the strict bibliographic sense. They are the best direct materials for testing and extending the book.

1. Columbia, “Staying Power: Jean-Marie Eveillard” (2007/08). The complete interview PDF is the best open companion: it covers intellectual chronology, research workflow, analysts, leverage, shorting, risk, sell discipline, client redemptions and career advice. Read it before accepting a later summary.
2. MOI Global, “The Art of Global Value Investing” (recorded 2012). This edited transcript is best for post-2008 revision: EV/EBIT, cash as residual, gold, sell tension, credit excess, macro awareness and the psychology of lag. Its 2017 webpage date is not the recording date.
3. TheStreet, “10 Questions” (2003). The direct Q&A is the strongest live case demonstration of adjusted accounting, variable discounts, Tyco, McDonald's and gold. It reveals uncertainty in real time rather than after the outcome.
4. Columbia's 15th Graham & Dodd Breakfast recap (2005). The official account efficiently locates Eveillard between Graham and Buffett and records international-accounting cautions. It is a reported speech, not a verbatim transcript.
5. Ivey value-investing class report (2014). The school's event account is strongest on analytical compression, benchmark pain, perseverance and the client exodus.
6. Forbes Intelligent Investing interview (2009). The transcript gives the cleanest business-owner definition of value and captures the newly explicit top-down overlay after the crisis. It is useful precisely because it marks evolution rather than a timeless rulebook.

Best works about Eveillard, ranked

1. Ronald W. Chan, The Value Investors, Chapter 6 (second edition)

The publisher's contents identify “A Journey to the Center of Value” and its sections “Valley of Tears,” “The Inefficient Market,” “The Meaning of Value,” “The Courage to Say No” and “Seeking Protection” (Wiley). This is the best dedicated book chapter because it combines biography and direct interview material in a comparative global-value project. A CFA Institute review confirms the Eveillard coverage but exposes a useful limitation: the chapter compresses his early career into a growth-versus-value conversion narrative and necessarily lacks the later detail of his own book (CFA Institute review).

2. William Green and Michael O'Brien, The Great Minds of Investing (2015)

This is the strongest concise character study. The publisher describes 33 investor portraits paired with profiles drawn from Green's long interview archive (publisher). Green's accessible Eveillard profile excerpt centers the late-1990s near-death and capacity to endure pain. Its virtue is institutional fragility; its limitation is brevity and a focus on the dramatic redemption episode.

3. Michael Sincere, 101 Investment Lessons from the Wizards of Wall Street (2000)

The available interview excerpt preserves an early-career profile and the unnamed English-broadcaster license error. It is valuable because it predates the book and much of the later legend. The company, dates and loss remain unidentified, so the lesson is stronger than the case reconstruction.

4. Alex Gilchrist, “Learning From Jean-Marie Eveillard” (2021)

The MOI resource ranked above under authorship traps is the most detailed case-and-quotation index and was read and approved by Eveillard. It is the fastest case locator and page-reference bridge to the book. Its weakness is exactly why it cannot rank first: approval is not independent verification, and many figures remain book-derived.

5. Aaron Pressman, “For Eagle, Value Is the Prey” (2003)

The Washington Post profile is the best independent contemporaneous control for the 24-year fund result, team scale, retirement plan, Mexican bonds, Tyco and currency hedging. It prevents the later memoir from swallowing de Vaulx and the fund structure, although its performance figures are rounded and not a personal-account audit.

6. “The World According to Eveillard” (2007)

The Financial Advisor profile is best for the credit-cycle boundary just before the crisis: opaque banks, leverage, securitization and the unexpected return to management. It is a profile/interview hybrid and should not be cited as an authored essay.

7. William Green, Richer, Wiser, Happier (2021)

The publisher describes a synthesis drawn from more than 40 investor interviews. Eveillard appears chiefly as evidence about resilience, survival and the capacity to endure underperformance. Read it for cross-investor context, not for biography, bibliography or security-level facts; the thematic synthesis deliberately sacrifices attribution detail.

Recommended reading path

Start with Chapters 2 and 4 of Value Investing Makes Sense for philosophy and institutional endurance. Read the Columbia interview next to see the live operating process. Return to the book's Chapters 3 and 5, then the BIS/Kohler, Lindt, Shimano, Shaw and Swissair cases. Pair those retrospectives with the 2003 Washington Post profile and TheStreet interview. Use Chan for a compact outside biography, Green for temperament and the 2012 MOI transcript for the post-crisis revision. Finish with the book's last words and appendices.

That sequence moves from doctrine to contemporaneous evidence, then to cases and criticism. It is less likely to mistake one brief memoir for a complete archive or to attribute a team's fund record, institutional prose and successor doctrine to a single author.

Current legal and institutional boundary

First Eagle announced Eveillard's senior-adviser retirement in 2020 and assigned later portfolio oversight to successors (First Eagle retirement notice). A Delaware court dismissed Dandini v. First Eagle Funds with prejudice on July 9, 2026; Eveillard was not a defendant (Delaware Superior Court). The SEC's 2015 distribution-fee settlement named First Eagle entities, not Eveillard (SEC). These firm-level controls belong in the critical record but do not establish a personal finding, current management authority or a new Eveillard-authored work.

Task A Source Map - Profile (T0721)

As of 2026-07-20. These sources are ranked for profile work. Eveillard-direct, fund, team, adviser, successor-platform and secondary evidence are separated; fund returns and AUM are not treated as personal results.

1. Jean-Marie Eveillard memoir sample - Eveillard's own account of his exact birth date, wartime childhood, family, education and early intellectual development; translated book sample, used sparingly.
2. First Eagle Funds SAI, 2026 - Latest primary role control: management periods, Trustee Emeritus powers, non-voting legal status and compensation boundary.
3. First Eagle senior-adviser retirement notice, 2020 - Official retirement date, transfer of oversight and distinction between adviser and portfolio management.
4. Columbia Graham & Doddsville interview, 2007 - Best direct interview for early career, Graham conversion, process, commercial adversity and the Morningstar-attributed first-term return.
5. First Eagle Global Fund investor guide, September 2025 - Current institutional process and the crucial 1970 legal-fund versus 1979 manager/strategy inception footnote.
6. First Eagle official history and current scale - 1979 strategy, 1999 acquisition, later firm chronology and carefully defined March 2026 pro-forma AUM/AUA.
7. First Eagle semiannual report, 2009 - Primary transition record for Eveillard's final portfolio-management exit and senior-adviser role.
8. Washington Post, “For Eagle, Value Is the Prey,” 2003 - Contemporaneous independent 24-year return, Global Fund assets, team assets, retirement plan and vehicle examples.
9. Washington Post, “With Eveillard's Exit,” 2009 - Final-exit chronology, 1979–2009 vehicle result and the explicit two-year retirement interruption.
10. Forbes on SoGen's acquisition and 1998, 1999 - Contemporary 1998 return and monthly-redemption evidence plus transaction context.
11. Financial Advisor profile, 2002 - Independent asset peak/trough observations, direct process context and cycle evidence.
12. Morningstar Hall of Fame award archive - Official confirmation that the 2001 International Stock award was joint with Charles de Vaulx.
13. Kiplinger on Eveillard's return, 2007 - Contemporary 2007 return-to-management report, long-period result, rounded rebound record and single-source fund scale.
14. HEC Paris Eveillard endowed chair - Primary alumni status and institutional legacy; its $15 million-to-$50 billion strategy statement extends through successors and is not used as personal AUM.
15. Forbes Eveillard interview, 2009 - Direct account of Graham/Buffett value, business appraisal, horizon, Japan and monetary-gold rationale.
16. Ivey value-investing talk, 2014 - Direct later-career discussion of benchmarks, herd pressure and why value discipline is hard to sustain.
17. Los Angeles Times fund profile, 1991 - Contemporaneous portfolio-allocation snapshot demonstrating equities, cash, gold and bonds in one named vehicle.
18. Los Angeles Times fund profile, 1996 - Later cash, gold, bond and equity snapshot; useful evidence against treating allocations as fixed rules.
19. First Eagle Variable Funds annual report, 2023 - Primary January 1940 birth-month and former-role disclosure, plus Trustee Emeritus status.
20. First Eagle Variable Funds report filed March 2026 - Strongest current primary evidence that Eveillard remains living and listed as Trustee Emeritus; not a vital record.
21. SEC First Eagle distribution-fee settlement, 2015 - Primary firm-level regulatory counterevidence; the release does not name Eveillard personally.
22. Dandini v. First Eagle Funds, Delaware Superior Court, 2026 - Current firm-level litigation control; dismissal with prejudice, with Eveillard not a defendant.
23. First Eagle Global Fund current page - Current successor vehicle, share-class and net-asset context; not an Eveillard record.
24. First Eagle Global Value strategy page - Current official strategy benchmark and successor-team context; later rules are not backdated to Eveillard.

Task A evidence limitations

No audited Eveillard personal account, complete trade ledger, independently examined 1979–2004 monthly series, consistent peak-AUM denominator, citizenship record, current decision-right map or universal legal-clearance source was located. The 15.8%/13.7% headline is secondary and Morningstar-attributed; the 13.8% 1979–2009 vehicle result includes a two-year retirement; reported AUM mixes funds, dates, share classes and later platforms. Current First Eagle holdings and results belong to successor teams. Targeted negative legal searches bound the public record but cannot prove the absence of every proceeding.

Task B Source Map - Investment Philosophy (T0722)

As of 2026-07-20. Sources are listed in first-use order. Eveillard's own historical statements are separated from fund/team records and successor doctrine; dated ranges and portfolio snapshots are not treated as universal rules.

1. First Eagle Variable Funds report filed March 2026 - Current primary evidence that Eveillard is living and listed as Trustee Emeritus; not evidence of portfolio authority.
2. First Eagle Funds SAI, 2026 - Primary definition of the emeritus role as unpaid, non-voting and consultative rather than a legal trustee or portfolio manager.
3. Columbia Graham & Doddsville interview, 2007/08 - Principal direct source for worldview, research workflow, horizon, risk, temperament, leverage, shorting, and Graham-to-Buffett evolution; visually checked during T0722.
4. Columbia Business School, global value speech, 2005 - Official contemporaneous recap of Eveillard's place between Graham and Buffett, global opportunity boundaries and accounting concerns.
5. MOI Global interview, recorded 2012 - Direct retrospective source for absolute and relative objectives, EV/EBIT, cash, gold, sell tension, catalyst rejection and post-2008 macro overlay.
6. Journal Record/AP on value investors, 1998 - Contemporaneous statement of the commercial dilemma created by refusing momentum.
7. Ivey value-investing talk, 2014 - Official direct-speech report on perseverance, benchmark lag and analytical compression.
8. CFA Society Chicago event recap, 2017 - Eveillard's Enron-footnote example and boundary for incomprehensible accounting.
9. TheStreet Tyco interview, 2003 - Direct case study in sum-of-the-parts work, adjusted debt and income statements, and variable discount requirements.
10. First Eagle historical excerpts, 2022 - Official archive reproducing a March 1999 fund-report statement against hypothetical growth; attributed to the vehicle rather than Eveillard alone.
11. Kiplinger on Eveillard's return, 2007 - Contemporary evidence of broad diversification, cash when bargains were absent, gold, and credit-cycle caution.
12. Advisor Perspectives final interview, 2009 - Direct source for dated gold ranges, currency hedges and top-down scenario awareness; visually checked during T0722.
13. Los Angeles Times on cash and caution, 1998 - Contemporaneous counterevidence that experience could produce excessive caution and cash could reflect more than discipline.
14. SFGate on Eveillard and the Gold Fund, 1997 - Direct contemporaneous evidence that patience had limits and Eveillard was prepared to concede a failed gold thesis.
15. Washington Post, “For Eagle, Value Is the Prey,” 2003 - Independent evidence on the yen hedge's opportunity cost, team attribution, and security examples.
16. SEC-filed First Eagle semiannual report, 2009 - Primary fund record for drawdown, gold contribution, cash, currency hedges and Eveillard's senior-adviser boundary.
17. Los Angeles Times on falling knives, 1996 - Direct contemporaneous evidence of uncertainty, error admission and the discomfort of contrarian investing.
18. First Eagle senior-adviser retirement notice, 2020 - Official retirement and succession boundary; successor positions are not attributed to Eveillard.
19. First Eagle Global Fund investor guide, 2025 - Current four-pillar and portfolio framework, used only to distinguish successor doctrine; its prose also conflicts with its 1979 manager-inception footnote.

Task B evidence limitations

No Eveillard-signed shareholder-letter archive, complete trade ledger, verified primary-source position-sizing formula, universal sell threshold, or timeless gold band was located. The full text of Value Investing Makes Sense was not lawfully available, so book-derived EV/EBIT bands, exact sizing language and portfolio-count rules were excluded. Post-2009 comments are retrospective advice, not proof of positions Eveillard personally controlled. Four agent workstreams completed 174 live searches/retrievals; a fifth distinct audit agent was rejected repeatedly by the platform's hard thread limit, so the main agent completed the independent audit sequentially with 30 further targeted searches and 12 pivotal-source reopenings. PDF pages from the 2007/08 and 2009 interviews were visually checked. Targeted current SEC, FINRA and legal searches found no Eveillard-specific current enforcement matter, but that bounded negative result is documented in Task A and is not universal legal clearance.

Task C Source Map - Greatest Trades (T0723)

As of 2026-07-20. Sources are listed in first-use order. Fund holdings and results are not personal-account returns; post-1999 decisions are team-attributed where the record requires it.

1. First Eagle Variable Funds report filed March 2026 - Current primary evidence that Eveillard is living and Trustee Emeritus without portfolio or legal-trustee authority.
2. FINRA BrokerCheck report - Current registration and bounded disclosure-event check; visually checked during T0723.
3. MOI Global Eveillard case summaries - Eveillard-approved, book-derived source for BIS, Shaw and Lindt chronology, thesis, sizing and outcomes; single-source figures remain labeled; visually checked during T0723.
4. BIS shares-withdrawal record - Primary compulsory-withdrawal date, share count, initial payment and final additional compensation.
5. First Eagle annual report, 2002 - Signed primary fund account of the BIS partial award, NAV effects, gold contribution and fiscal-2002 results.
6. Permanent Court of Arbitration BIS case - Primary institutional chronology, parties, partial and final awards, correcting the recurring ICJ misattribution.
7. UN Reports of International Arbitral Awards, BIS partial award - Primary legal reasoning for NAV less 30%, First Eagle's position and the lawful-recall finding; visually checked during T0723.
8. First Eagle semiannual report, 2003 - Primary shares, costs, values, weights, contribution commentary and fund results for Shaw, Buderus, Brau Union/BBAG, Lindt, gold and Tyco.
9. Bosch annual report, 2003 - Primary acquisition completion and year-end Buderus ownership.
10. Austrian Takeover Commission, Brau Union offer - Primary offer price, period, volume and reference-price premiums.
11. Lindt & Sprungli financial report, 2014 - Primary confirmation of 2014 net income; visually checked during T0723.
12. Fortune, “Burned by Gold,” 1998 - Contemporary adverse evidence on Gold Fund inception-to-date loss and contemplated closure.
13. Washington Post, “Patience Is Golden,” 2003 - Independent gold drawdown, recovery and attribution evidence.
14. TheStreet Tyco interview, 2003 - Direct account of Tyco discovery, sum-of-parts appraisal, average equity cost and bond yield.
15. Washington Post, “For Eagle, Value Is the Prey,” 2003 - Independent evidence for Tyco's path, team attribution, Mexican-bond limitation and October 2003 price/appraisal.

Task C evidence limitations

No complete Eveillard or fund trade ledger, original tax lots, continuous position-level drawdowns, absolute realized P&L or independently audited security-attribution series was located. BIS is the only case with a primary-source compulsory exit price; even there, original cost and dividends are missing. Shaw's 15% CAGR and Lindt's early chronology are Eveillard-approved but book-derived and not independently audited. The two required subagent spawn attempts were rejected by the platform's hard thread limit, so the main agent completed all five primary-document, track-record/arithmetic, direct-voice, adverse-evidence and independent-audit workstreams sequentially with 48 targeted searches or retrievals plus pivotal-source reopenings. Current living, registration and legal-status boundaries were refreshed through March/July 2026 SEC and FINRA material plus a July 2026 obituary search; this bounded negative review is not universal legal clearance.

Task D Source Map - Mistakes and Losses (T0724)

As of 2026-07-20. Sources are listed in first-use order. Realized security losses, unrealized marks, vehicle returns, relative lag, opportunity costs and commercial losses are kept separate; fund and team outcomes are not personal-account P&L.

1. First Eagle Variable Funds report filed March 2026 - Current primary evidence that Eveillard is living and listed only as Trustee Emeritus, without current portfolio authority.
2. FINRA BrokerCheck - Current bounded registration and disclosure-event check; not universal legal clearance.
3. Dandini v. First Eagle Funds, Delaware Superior Court, 2026 - Current firm-level litigation control; July 2026 dismissal with prejudice, with Eveillard not a defendant.
4. SEC First Eagle distribution-fee settlement, 2015 - Primary firm-level regulatory counterevidence and exact conduct period; the release does not name Eveillard.
5. MOI Global Eveillard case summaries - Eveillard-approved, book-derived Swissair thesis, deterioration, reported size/loss and lessons; security figures remain labeled single-source; visually checked during T0723 and reopened during T0724.
6. Swiss Federal Audit Office, Swissair loan audit - Independent primary confirmation of the October 2001 liquidity grounding and government-loan amounts.
7. Capital Ideas, Sincere interview excerpt - Attributed English-broadcaster licensing mistake and foreign-regulatory lesson; company and P&L remain unidentified.
8. Fortune, “Burned by Gold,” 1998 - Contemporary Gold Fund inception-to-date loss, relative result and contemplated closure.
9. Washington Post, “Patience Is Golden,” 2003 - Independent alternative gold drawdown interval, later recovery and miner-to-bullion change.
10. SFGate on Eveillard and the Gold Fund, 1997 - Direct contemporaneous closure threshold and willingness to tell shareholders the thesis had failed.
11. First Eagle annual report, 2002 - Primary fiscal-2002 Gold Fund result after the earlier multi-year slump.
12. First Eagle Global Fund investor guide, March 2026 - Official Global Fund results across Black Monday, the technology bust and the financial crisis; visually checked during T0724.
13. SEC-filed First Eagle semiannual report, 2009 - Primary American Express shares, cost and value; fund return, unrealized depreciation, gold contribution and transition boundary.
14. Advisor Perspectives final interview, 2009 - Direct American Express assessment, crisis positioning and succession boundary; extracted PDF text checked during T0724 after direct download returned 403.
15. MOI Global interview, recorded 2012 - Direct retrospective admission that value investors underweighted credit-regime and mortgage excesses after 2008.
16. Jason Zweig on Black Monday, 2017 - Direct retrospective account of freezing, later modest buying and acknowledged under-buying after the 1987 crash.
17. Los Angeles Times on excessive caution, 1998 - Contemporaneous self-criticism that experience had made Eveillard too cautious.
18. Washington Post, “For Eagle, Value Is the Prey,” 2003 - Independent yen-hedge opportunity cost and late-1990s relative-performance evidence.
19. Columbia Graham & Doddsville interview, 2007/08 - Direct temporary-loss versus permanent-impairment rule and sell discipline; visually checked during T0724.
20. Forbes on SoGen's acquisition and 1998, 1999 - Contemporary Global Fund return, monthly-redemption evidence and sale context.
21. Financial Advisor profile, 2002 - Independent asset peak/trough observations and late-1990s commercial context.

Task D evidence limitations

No complete trade ledger, original tax lots, continuous security drawdowns, realized-loss schedule or audited opportunity-cost series was located. Swissair's size and loss and the English broadcaster case are attributable but book/interview-derived; the latter cannot be identified safely. Gold Fund sources use different intervals and return conventions. American Express is an April 2009 unrealized mark, not realized P&L. The two required subagent spawn attempts were rejected by the platform's hard thread limit, so the main agent completed all five primary-document/chronology, track-record/arithmetic, direct-voice/process-change, adverse/current-legal and independent-audit workstreams sequentially with 52 targeted searches plus direct retrievals and pivotal-source reopenings. Five citations were reopened and confirmed; the 2007/08 interview and 2026 First Eagle guide were visually checked. Current legal searches are bounded negative evidence, not universal clearance.

Task E Source Map - In Their Own Words (T0725)

As of 2026-07-20. Sources are listed in first-use order. The quote corpus accepts only explicitly attributable Eveillard speech or authored text; joint interviews, team documents, event summaries and audio pages without verified transcripts are labeled and indexed but not silently converted into individual quotations.

1. First Eagle Variable Funds report filed March 2026 - Current primary evidence that Eveillard remains living and is listed only as Trustee Emeritus, without current portfolio authority.
2. FINRA BrokerCheck - Current registration and bounded disclosure-event check; not universal legal clearance.
3. Columbia Graham & Doddsville interview, 2007/08 - Principal direct source for 18 of the 40 quotations, intellectual chronology, research, risk, temperament and career advice; text-extracted and visually checked during T0725.
4. TheStreet interview, 2003 - Direct Q&A source for valuation, accounting, error recognition and gold.
5. Forbes video transcript, 2009 - Direct interview source for the business-owner definition of value; article endpoint is sometimes paywalled, with an archival Forbes PDF separately indexed.
6. MOI Global interview, recorded 2012 - Edited direct transcript for EV/EBIT, quality, gold, cash, credit and client psychology; the page date is 2017, but MOI identifies the 2012 recording date.
7. Ivey Business School talk, 2014 - Official report with directly attributed quotations on simplicity, perseverance, lag and the late-1990s experience.
8. The Wall Street Transcript, 2002 - Direct long-horizon quotation and interview metadata; most of the transcript is paywalled.
9. MOI Global Asian Investing Summit, 2014 - Selected direct transcript quotations, including the importance of avoided holdings; full session is access-controlled.
10. German sample of Value Investing Makes Sense - Lawful author-authored sample and metadata; visually checked, but not back-translated for the English quote corpus.
11. First Eagle historical excerpts, 2022 - Official reproduction of a March 1999 SoGen Funds passage; attributed to the fund entity, not Eveillard individually.
12. Financial Advisor, “The World According to Eveillard,” 2007 - Direct profile/interview source on credit, banks, leverage and the 2007 investment environment.
13. Advisor Perspectives final interview, 2009 - Joint Eveillard/Deshpande interview on portfolio construction, gold, Japan and American Express; answers lack individual speaker labels and are excluded from the personal quote corpus.
14. Forbes Intelligent Investing anniversary briefing book, 2009 - Archival Forbes PDF preserving the Japan portion of Eveillard's interview.
15. CFA Institute, “Conversation with a Veteran Value Investor,” 2011 - Official audiovisual event page; no surviving transcript was used for quotations.
16. Columbia, 15th Graham & Dodd Breakfast recap, 2005 - Official speech recap with reported insights, not full verbatim proceedings.
17. MOI Global Asian Investing Summit, 2015 - Selected direct quotations on Asian and commodity-linked markets; full session is access-controlled.
18. CFA Society Chicago event recap, 2017 - Official event report on intellectual influences and accounting; not a verbatim transcript.
19. Morningstar conversation with Martin Whitman, 2007 - Surviving seven-segment video index; no verified transcript was used.
20. Value Investing with Legends podcast, 2019 - Primary audio episode and detailed timestamp index; metadata rather than a verified transcript.
21. SEC First Eagle distribution-fee settlement, 2015 - Primary firm-level regulatory counterevidence; the release does not name Eveillard.
22. Dandini v. First Eagle Funds, Delaware Superior Court, 2026 - Current firm-level litigation control; dismissal with prejudice, with Eveillard not a defendant.
23. First Eagle senior-adviser retirement notice, 2020 - Official boundary between Eveillard's historical voice and successor-team portfolio authority.

Task E evidence limitations

No public systematic archive of Eveillard-signed shareholder letters, complete transcript of the 2002 Wall Street Transcript interview, verbatim proceedings for the 2005/2016/2017 speeches, or verified text transcript of the 2011 and 2019 audio programs was located. The 2009 Advisor Perspectives answers are jointly unattributed, and the March 1999 fund prose is institutional. They were indexed but excluded from the 40-quotation personal corpus. Quotes were checked at 25 words or fewer, and the Columbia transcript plus German book sample were visually inspected. The two required subagent spawn attempts were rejected by the platform's hard thread limit, so the main agent completed all five primary-material/chronology, quote-count/provenance, speeches-podcasts-letters, adverse/current-status and independent-source-audit workstreams sequentially with 32 meaningful searches plus direct retrievals. Five randomized citations were reopened and confirmed. Endpoint QA returned 14 direct HTTP successes, six access-controlled responses and three dynamic or transient command-line results that were independently retrievable in the browser, with no confirmed dead source. Current legal searches are bounded negative evidence, not universal clearance.

Task F Source Map - Key Writings (T0726)

As of 2026-07-20. Sources are listed in first-use order. Sole authorship, translation, institutional prose, direct speech, approved compilation and outside profile are kept distinct; translated editions of the same book are not counted as separate works.

1. First Eagle Variable Funds report filed March 2026 - Current primary evidence that Eveillard is living and listed only as Trustee Emeritus, without legal-trustee or portfolio authority.
2. FINRA BrokerCheck - Current registration boundary and unusually strong self-reported authorship evidence naming Value Investing Makes Sense and the April 2015 start of author activity.
3. Valor Editions, French publisher page - Sole-author credit, French title, publication metadata, ISBN and complete chapter/appendix list.
4. German publisher preview - Primary rights page, 2016 original-publication statement, translator, edition ISBNs, exact section-level contents and lawful preface sample; visually checked during T0726.
5. Spanish publisher page - Independent publisher-level translation metadata and full chapter/appendix architecture.
6. Strictly Value book review, 2016 - Contemporary independent appraisal of the book's scope, special-situations material, humility, direct style and editing limitations.
7. First Eagle historical excerpt - Official reproduction of March 1999 institutional fund prose, used to show that vehicle attribution is not an Eveillard personal byline.
8. MOI Global, “Learning From Jean-Marie Eveillard” - Alex Gilchrist-authored, Eveillard-read-and-approved case and quotation compilation; valuable bridge to the book but not an authored Eveillard work.
9. Columbia Graham & Doddsville interview, 2007/08 - Best complete open direct interview on intellectual history, process, risk, temperament and sell discipline.
10. MOI Global interview, recorded 2012 - Edited direct transcript for post-crisis evolution, valuation, cash, gold, credit and client psychology.
11. TheStreet “10 Questions” interview, 2003 - Contemporaneous direct case evidence on accounting reconstruction, variable discounts, Tyco, McDonald's and gold.
12. Columbia global-value speech recap, 2005 - Official reported speech placing Eveillard between Graham and Buffett and recording international-accounting cautions.
13. Ivey value-investing class report, 2014 - Official event account on analytical compression, perseverance, lag and the late-1990s client exodus.
14. Forbes Intelligent Investing interview, 2009 - Direct transcript for business-owner value and the newly explicit post-crisis macro overlay.
15. Wiley, The Value Investors, second edition - Publisher contents and section locators for Ronald Chan's dedicated Eveillard chapter.
16. CFA Institute review of The Value Investors - Independent professional review confirming the Eveillard chapter's scope and revealing its necessary biographical compression.
17. Publisher page for The Great Minds of Investing - Primary book metadata and editorial method for the 33 investor portraits and profiles.
18. William Green's Eveillard profile excerpt - Accessible book-profile text centered on late-1990s institutional near-death and the capacity to endure pain.
19. Michael Sincere interview excerpt - Early interview-derived biography and English-broadcaster licensing error; company and P&L remain unidentified.
20. Washington Post, “For Eagle, Value Is the Prey,” 2003 - Best independent contemporaneous control for team attribution, fund result, scale, security cases and currency hedging.
21. Financial Advisor, “The World According to Eveillard,” 2007 - Profile/interview on opaque banks, leverage, credit and the return to management just before the financial crisis.
22. Simon & Schuster, Richer, Wiser, Happier - Publisher record for Green's cross-investor synthesis; useful for resilience context rather than Eveillard bibliography or security facts.
23. First Eagle senior-adviser retirement notice, 2020 - Official retirement and successor-oversight boundary.
24. Dandini v. First Eagle Funds, Delaware Superior Court, 2026 - Current firm-level litigation control; dismissal with prejudice, with Eveillard absent as a defendant.
25. SEC First Eagle distribution-fee settlement, 2015 - Primary firm-level regulatory counterevidence; the release does not name Eveillard.

Task F evidence limitations

Only one conventional Eveillard-authored investment book was verified. No separate bylined article corpus, academic paper, systematic personally signed shareholder-letter archive, full trade ledger or English publisher archive with stable open access was located. Translated editions have different dates, page counts and ISBNs but the same chapter architecture. Book cases are retrospective and selected; security-level cost, drawdown, exit and attribution remain incomplete, and fund results belong to vehicles and teams. Both required subagent spawn attempts were rejected by the platform's hard thread limit, so the main agent completed all five authored-corpus/bibliographic, primary-work deep-reading, works-about/adverse, current-status/legal and independent-source-audit workstreams sequentially with 34 meaningful live searches plus direct retrievals. The German rights page, contents and preface were visually checked. Current legal searches are bounded negative evidence, not universal clearance.