Peter Cundill

As of 2026-07-18.

Guiding questions

1. What did Cundill mean by value once he moved beyond a mechanical net-net screen?
2. Why did he search globally, and what information could only be learned on the ground?
3. How did an idea move from sourcing through research, valuation, entry, sizing, portfolio construction, and sale?
4. How did he distinguish a temporarily unpopular asset from a value trap?
5. Which risk controls protected capital, and where did they fail?
6. How did the method evolve after early success, Japanese stagnation, the technology bubble, and larger mandates?
7. Which claims are Cundill's own, which belong to a team or product, and which are later successor practice?
8. What can an individual investor reproduce without Cundill's organization, access, and unusually patient capital?

Core Worldview

Peter Cundill's organizing belief was that price and realizable business value can diverge sharply because investors extrapolate, follow fashion, avoid embarrassment, and abandon difficult markets. The investor's job was not to forecast the market more accurately. It was to buy a claim on assets or earning power for materially less than a conservative estimate of what an informed owner could realize, while retaining enough financial strength and patience to survive the wait (Ivy Funds semiannual report, 2002; Cundill at Ivey, 2005, 1:05:49).

His late-1973 conversion to Benjamin Graham began with the balance sheet. The authorized biography's preview describes his insight as liquidation value: cash and other current assets, less all liabilities, could sometimes exceed the share price even before assigning much value to the operating business. That discovery made accounting analysis—not a low price/earnings ratio, dividend yield, or exciting growth story—the groundwork of his process (There's Always Something to Do, authorized preview, 2011). In his own 2005 Ivey lecture, Cundill dated the discovery to 1972–73 and said that securities bought below cash had worked with remarkable regularity in that era (Cundill at Ivey, 2005, 1:03).

That was a starting point, not a lifelong frozen formula. By 2005 he described three broad valuation routes: asset-based net asset value, sum-of-the-parts value, and an estimate of future net asset value. He regarded sum-of-the-parts as especially important and the future-value method as the least dependable. The common discipline was to begin with observable facts, use conservative arithmetic, and demand a margin for error (Cundill at Ivey, 2005, 21:30–29:50).

The Edge — What Markets Misprice And Why

Cundill sought places where other investors were least willing or able to look. Every November he traveled to a market that had performed especially badly, then met companies and local observers before committing capital. Sweden in the 1970s and Japan later exemplified this seek-out-trouble habit (Cundill at Ivey, 2005, 5:00; There's Always Something to Do, authorized preview, 2011). A global mandate expanded the candidate set and let him compare an overheated home market with neglected securities elsewhere.

The opportunity usually came from some combination of:

• Aversion and institutional neglect. Recent losses, poor headlines, small size, foreign accounting, bankruptcy, or an unfashionable industry removed buyers before the facts had been fully examined (Cundill at Ivey, 2005, 48:39; Ivy Funds semiannual report, 2002).
• Accounting obscurity. Excess cash, understated property, cross-holdings, or separately valuable divisions could be hidden by consolidated earnings. Conversely, goodwill and recurring operating losses could make apparent book value illusory (Cundill at Ivey, 2005, 11:54).
• Short horizons. Clients and institutions often could not tolerate a position that fell first or remained dormant for years. Cundill expected both (Ivy Funds semiannual report, 2002; Cundill at Ivey, 2005, 1:03:00).
• Control and governance frictions. A discount might persist because insiders could prevent assets from reaching minority owners. The same friction could become an opportunity only when ownership, buybacks, dividends, liquidation, or a credible manager supplied a route to realization (Cundill at Ivey, 2005, 53:14; Cundill at Ivey, 2005, 1:11:39).
• Fashion and extrapolation. His journals, as represented in the authorized biography, treated trends as sources of short-term mispricing rather than foundations for valuation (There's Always Something to Do, authorized preview, 2011).

The edge was therefore investigative and temperamental, not simply quantitative. Screens found oddities; forensic accounting tested whether the discount was real; travel supplied legal, cultural, competitive, and character evidence; and patience gave the thesis time to mature. Cundill sometimes rejected statistically cheap companies after meeting the people or learning the local rules. The numbers established a candidate and downside case, but judgment decided whether the assets were accessible and whether management could be trusted (There's Always Something to Do, authorized preview, 2011; Cundill at Ivey, 2005, 5:03).

Process: Idea Sourcing → Research → Valuation & Entry → Sizing → Portfolio Construction → Sell Discipline

Idea sourcing

Cundill combined systematic and social discovery. He and the team screened large global databases for net-nets, low price/book ratios, low earnings multiples, modest debt, and other balance-sheet anomalies. He also read the Outstanding Investor Digest, followed other respected value managers, watched new-low lists, read broadly, and used a network of analysts and investors (Cundill at Ivey, 2005, 35:00). His screen was deliberately broad: in 2005 he said a global search had found only about 55 classic “magic sixes”—roughly a 6% yield, six times earnings, and 60% of book value—among approximately 55,000 companies [single-source Peter lecture figures] (Cundill at Ivey, 2005, 2:44). The scarcity explained why the method had to adapt beyond pristine net-nets.

Research

Research began with a reconstructed balance sheet. The analyst marked down questionable assets, eliminated goodwill, deducted every liability, normalized working capital, and asked how continuing losses could erode net current asset value. For a going concern, that liquidation calculation was a point of departure rather than an automatic buy. The next questions were whether the business could survive, what each division or investment was worth, who controlled the assets, and what might close the gap (Cundill at Ivey, 2005, 11:54; Ivy Funds semiannual report, 2002).

A directly attributed 2002 Ivy interview adds the organizational routine: detailed balance-sheet and cash-flow analysis, breakup or liquidation appraisal, management interviews, site visits, competitor work, and investigation of events that could change value or liquidity. It says the team often sought a discount of at least 50% to estimated net asset value and expected to hold for three to five years or longer (Ivy Funds semiannual report, 2002). These are Peter-led team practices for one U.S. vehicle, not verified universal thresholds for every mandate.

Amada illustrates the full logic. Cundill began with cash, current assets, debt, and book value, but Japanese cross-shareholdings and weak minority-shareholder incentives made apparent bargains capable of remaining trapped. After having sold the company earlier when losses damaged the balance-sheet thesis, he reconsidered it when business and governance conditions changed. His lesson was not that every net-net works; it was that the same security must be re-underwritten as losses, control, and catalysts change (Cundill at Ivey, 2005, 12:00–21:30).

Company visits were investigative rather than ceremonial. Cundill looked for capable, honest people, conservative accounting, sensitivity to minority shareholders, and a manager, controlling owner, or external actor with reason to unlock value. Local travel tested whether the legal and cultural environment would permit the arithmetic to matter. The authorized biography's early Bethlehem Copper and Credit Foncier cases show both modes: a producing, debt-free mine selling below cash, and a conservatively accounted financial company whose hidden assets required a control or realization analysis (There's Always Something to Do, authorized preview, 2011).

Valuation & entry

Cundill triangulated rather than worshipped one model (Cundill at Ivey, 2005, 28:02):

1. Adjusted net asset value: cash and realizable current assets, plus conservatively marked long-lived assets, less all liabilities and the cost of surviving to realization.
2. Sum of the parts: value divisions, securities, land, or subsidiaries separately, then deduct corporate costs, debt, taxes, and leakage to controlling owners.
3. Earning power or future asset value: use only when the business and accounting made future cash generation defensible, with a wider margin for error.

He was skeptical of precise discounted-cash-flow outputs whose answer changed radically with small discount-rate assumptions, preferring simple arithmetic anchored in facts (Cundill at Ivey, 2005, 51:15). A 2003 Ivy prospectus formalized the team version: the adviser estimated intrinsic value primarily from financial statements, while also considering earnings, dividends, business prospects, management, and potential catalysts, and bought at a significant discount (Ivy Funds prospectus, 2003). This is mandate-level evidence of the Cundill organization, not proof that Peter personally made every decision.

A 2007 prospectus, while Cundill and Hhu Ng were named co-managers, further identifies financial capacity, cash flow, governance, foreign-accounting anomalies, competition, and industry conditions as research inputs (Ivy Funds prospectus, 2007). The increasing list is evidence of a broadened team process, not abandonment of the balance-sheet anchor.

Entry was usually early by design. He accepted that a neglected security could fall after purchase and that a catalyst might take years. But patience was earned only if the revised downside remained acceptable. The JSTOR-controlled chapter record summarizes his broad hurdle as measurable, acceptable downside and a better-than-even chance of a satisfactory profit; it characterizes the posture as calculated risk rather than a gamble (JSTOR record for There's Always Something to Do, 2011). No public source reviewed here establishes one universal discount, target-return formula, or averaging-down schedule across his career.

Sizing

Public sources reviewed do not disclose a universal Cundill sizing formula. Across the cases, discount, liquidity, financial strength, control, and event risk appear to have conditioned conviction; that relationship is an analytical reconstruction, not a stated equation. Cundill expected errors and therefore diversified even after moving toward greater concentration. In the 2005 lecture he described a later portfolio in which roughly 75% of assets sat in about 25 holdings, counting a large cash allocation as one [single-source Peter lecture figures for his then-current practice] (Cundill at Ivey, 2005, 46:30). This is evidence of an evolved portfolio, not a permanent cap or an audited description of every mandate.

Adding after a decline was legitimate only after testing whether price or intrinsic value had changed. The 2002 Ivy annual report shows the team using Japanese weakness to replace names it considered fairly valued with cheaper opportunities and to add to existing positions; it also records that margin-of-safety analysis did not prevent further losses (Ivy Funds annual report, 2002). The distinction is critical: lower price enlarges the opportunity only when solvency, asset value, governance, and the route to realization remain intact.

Portfolio construction

The portfolio was bottom-up and global. The 2003 Ivy mandate disclaimed economic and aggregate-earnings forecasts, market timing, and top-down portfolio construction; it also permitted unlimited defensive cash and high-quality fixed income when bargains were scarce (Ivy Funds prospectus, 2003). Cundill likewise said cash accumulated because he could not find qualifying securities, not because he had predicted an index move (Cundill at Ivey, 2005, 51:15).

Two overlays complicated the “pure stock picker” description. First, the house fully hedged foreign-currency exposure rather than making partial tactical calls. In 2002, that policy hurt the Ivy product as foreign currencies appreciated against the U.S. dollar (Ivy Funds annual report, 2002). Second, Cundill accepted derivatives chiefly as protective tools rather than speculative weapons (Cundill at Ivey, 2005, 37:00). These were mandate and risk decisions around a bottom-up core, not evidence that macro forecasts drove security selection.

Sell discipline

Cundill's memorable operating rule was to sell half after a position doubled, then ask the board for permission to retain the remainder. He linked the rule to the recurring value-investor error of selling winners too early and cited Canadian Marconi as a missed-gain example (Cundill at Ivey, 2005, 33:40). A former Cundill colleague, Tim McElvaine, independently described the discipline as a practical guard against both greed and premature liquidation (MOI Global, 2021). It should be understood as a practiced governance rule during part of his career, not a timeless mathematical law.

The deeper sell logic was prospective. Sell or reduce when price reaches a conservative appraisal; when operating losses, leverage, accounting, control, or regulation breaks the thesis; or when a better risk-adjusted opportunity displaces the holding. A catalyst remains important because statistically cheap assets can stay cheap indefinitely. The 2002 Ivy report documents the team selling holdings it judged fairly valued and recycling capital, while the Amada history shows Cundill exiting when continued losses impaired net asset value and later returning when the facts improved (Ivy Funds annual report, 2002; Cundill at Ivey, 2005, 18:00). Cost basis and past regret did not determine the next decision.

Risk Management

The controls addressed several paths to permanent impairment:

• A balance-sheet margin of safety and survival analysis so continuing losses and liabilities could not quietly consume the discount (Cundill at Ivey, 2005, 11:54).
• Multiple valuation methods rather than dependence on a precise forecast (Cundill at Ivey, 2005, 28:02).
• Governance and catalyst analysis to test whether value could reach minority shareholders (Cundill at Ivey, 2005, 53:14; Cundill at Ivey, 2005, 1:11:39).
• Portfolio diversification because even intensive research could miss accounting, regulatory, or operating damage (Cundill at Ivey, 2005, 1:09:39).
• Cash when standards were not met rather than forced investment (Cundill at Ivey, 2005, 51:17).
• Currency hedging and occasional protective derivatives at the portfolio or mandate level (Ivy Funds annual report, 2002; Cundill at Ivey, 2005, 37:41).

Cable & Wireless is the clearest warning against treating a cash-rich balance sheet as a floor. Cundill valued the cash and Caribbean operations while initially assigning zero value to the fiber-optic network; the network proved to have negative value as liabilities and losses consumed the apparent cushion. He presented it as a major loss and a reason to diversify even a high-conviction portfolio (Cundill at Ivey, 2005, 68:00). The June 2002 U.S. Ivy report captures the error in real time: Cable & Wireless was the fund's worst performer, remained a core position, and the fund bought more while realizing tax losses. Cundill's later lecture—not that filing—explained the network's negative value (Ivy Funds semiannual report, 2002).

Risk controls could also become costs. The full currency hedge hurt the U.S. Ivy vehicle in 2002, and cash could lag in powerful bull markets. Diversification reduced security-specific ruin but could dilute the best ideas; concentration improved impact but magnified thesis errors. Cundill's solution was not to eliminate those tensions, but to make them explicit and remain flexible within a disciplined framework (Ivy Funds annual report, 2002; Cundill at Ivey, 2005, 6:45; Cundill at Ivey, 2005, 1:09:39).

Temperament & Psychology

Patience was an active discipline. It meant tolerating boredom, early losses, and career pressure while repeatedly rechecking the facts. Cundill warned that a manager who launched a net-net strategy in the wrong part of the 1973–75 cycle might lose the client before the bargains paid, even if the securities ultimately worked (Cundill at Ivey, 2005, 63:00). Organizational patience was therefore as important as analytical patience.

His journals, as represented in the authorized book, emphasize thoroughness, measured action, humility, teamwork, and people of competence and integrity (There's Always Something to Do, authorized preview, 2011). His lecture adds curiosity, travel, broad reading, and comfort admitting an error. Intuition mattered, but he described it as accumulated experience operating inside a disciplined framework—not permission to ignore the accounts (Cundill at Ivey, 2005, 59:00). He also rejected insider tips outright: privileged-sounding information was likely wrong, legally dangerous, or both (Cundill at Ivey, 2005, 67:00).

Evolution Over Career

The core sequence remained stable: find neglect, reconstruct value, inspect the people and place, demand a discount, wait, and sell when value or facts changed. What evolved was the breadth of acceptable evidence and instrument. The 1970s supplied plentiful classic Graham bargains. As they became scarce, Cundill incorporated sum-of-the-parts situations, bankruptcies, stronger businesses with defensible franchises, control transactions, and selected protective derivatives. He described the mature style as eclectic rather than a pure collection of “cigar butts” (Cundill at Ivey, 2005, 39:00–44:00).

The 1990s imposed the hardest philosophical correction. He acknowledged exiting European banks too early, moving into Japan too early, and remaining too inflexible as growth and technology dominated. A sponsor chart reports a 9.4% loss in 1990 after 15 positive calendar years; the controlled authorized-book record separately summarizes that loss as about 9% and records only modest gains in 1991 and 1992 [single-source-family/affiliated return figures] (Mackenzie historical chart, 2015; JSTOR record for There's Always Something to Do, 2011). His resulting rule—borrowed from John Templeton—was to change a winning game. In Cundill's formulation, tactics and the opportunity set must evolve, while discipline, valuation, and the margin of safety remain (Cundill at Ivey, 2005, 6:30).

Later Mackenzie material describes a more formal quality-and-valuation scorecard, milestones, and a three-to-five-year horizon in the Cundill tradition (Mackenzie Cundill team profile, 2020). Those are successor-team practices, not rules that should be projected backward onto Peter. The authorized biography itself was assembled by a former fund director from interviews, speeches, and exclusive access to Cundill's journals; the publisher's provenance supports its value while also marking it as an edited, affiliated narrative (McGill-Queen's University Press, 2011).

What He Explicitly Rejected

• Top-down market timing as the stock-selection engine. Ideas should arise from company-specific value, not a heroic economic forecast (Ivy Funds prospectus, 2003).
• Forecast precision. Small changes in a DCF discount rate can manufacture confidence without improving knowledge (Cundill at Ivey, 2005, 51:15).
• Cheapness without survivability. A net-net that keeps losing money can liquidate the investor's margin of safety (Cundill at Ivey, 2005, 11:54).
• Cheapness without a path to realization. Hostile control, weak law, or indifferent management can make an asset discount permanent (Cundill at Ivey, 2005, 1:11:39).
• A frozen formula. Graham's own practice evolved; Cundill believed tactics also had to evolve inside a disciplined framework (Cundill at Ivey, 2005, 24:31; Cundill at Ivey, 2005, 6:45).
• Speculative derivative use. He described derivatives as defensive tools, not an offensive arsenal (Cundill at Ivey, 2005, 37:00).
• Insider tips. They corrupted independent analysis and created legal as well as factual risk (Cundill at Ivey, 2005, 1:07:25).
• Comfort as a prerequisite. If an idea was popular, obvious, and emotionally easy, the discount was likely smaller (Cundill at Ivey, 2005, 1:05:49).

Regimes Where It Thrives vs. Struggles

This regime map is an analytical synthesis, not a table Cundill published.

The method should do best when fear creates measurable discounts and catalysts, while law, financing, and management still permit value to reach shareholders. It should struggle when assets are controlled but not realizable, losses consume book value, or capital providers cannot wait (Cundill at Ivey, 2005, 31:40; Cundill at Ivey, 2005, 1:04:35; Cundill at Ivey, 2005, 1:11:39). Its most dangerous regime is not simply a rising market; it is one in which yesterday's accounting facts cease to describe tomorrow's obligations.

Tensions Between Stated Philosophy And Actual Behavior

Bottom-up investing versus market and currency overlays. The formal Ivy mandate rejected market timing (Ivy Funds prospectus, 2003), yet the organization fully hedged currencies (Ivy Funds annual report, 2002) and Cundill accepted defensive derivatives (Cundill at Ivey, 2005, 37:00). The tension was stronger than hedging alone: a contemporaneous profile reports that after investing in Japan in 1985–87 he shorted Tokyo several years too early and kept the short until 1995 (Forbes, 2004). The prospectus language therefore describes one product's stock-selection engine and policy, not every overlay Cundill used across his career.

Catalyst discipline versus prolonged value traps. Cundill stressed that cheap securities could remain cheap forever, yet Japanese holdings showed how long he could wait for governance to improve. Amada eventually supported the willingness to revisit, but Cable & Wireless showed that waiting while liabilities compound can destroy the cushion (Cundill at Ivey, 2005, 12:00; Cundill at Ivey, 2005, 1:08:34; Cundill at Ivey, 2005, 1:11:39).

Flexibility versus early exits and stubborn entries. He preached tactical adaptation after acknowledging that he sold European banks too soon, entered Japan too soon, and was too rigid in the 1990s. The philosophy's mature flexibility was partly a lesson extracted from behavior that had violated it (Cundill at Ivey, 2005, 6:45; Cundill at Ivey, 2005, 9:02).

Concentration versus diversification. A later portfolio concentrated roughly 75% of assets in about 25 holdings, counting cash as one [single-source Peter lecture figures], while Cable & Wireless taught him that even apparently strong balance sheets require diversification (Cundill at Ivey, 2005, 46:30; Cundill at Ivey, 2005, 1:08:34–1:09:39). The reconciliation is graduated conviction, not certainty: concentrate enough for researched ideas to matter, but retain enough breadth to survive an unknowable liability.

Margin of safety versus averaging down. The co-managed U.S. Ivy product's 2007 Japanese consumer-finance exposure is the hardest adverse case. Cundill and Hhu Ng acknowledged that political risk had impaired their original value estimates, yet the fund continued building Aiful, Takefuji, and Acom. At the March 31, 2007 period-end—not at the start of the year—Japan was 48.28% and financial services 36.23%, while the fund returned 10.71% versus 15.45% for the MSCI World Index [single-source co-managed U.S. product figures] (Ivy Funds annual report, 2007). A 2005 interview attributes the initial Takefuji purchase and addition specifically to Ng, not Peter alone (Washington Post, 2005). This one jointly managed vehicle cannot stand for every Cundill mandate, but it shows how a team using his framework could mistake lower price for wider safety after a regulatory thesis changed.

Mechanical Graham value versus business quality. Cundill began with liquidation arithmetic but later admitted selected franchise quality and future earning power. The controlled book record characterizes the blend as mostly Graham with smaller Buffett and Cundill elements (JSTOR record for There's Always Something to Do, 2011). Evolution broadened the opportunity set, but it also made valuation more dependent on judgment.

The reproducible core for an individual is demanding but compact: search neglected markets; rebuild the balance sheet; triangulate net asset, sum-of-the-parts, and earning-power value; investigate control and catalysts; write what would invalidate the thesis; size for both conviction and unknown liabilities; and let today's prospective payoff—not cost basis—govern sales. What cannot be copied directly is Cundill's global research organization, company access, board oversight, cross-border execution, client patience, or the team contributions embedded in fund results. The durable lesson is not “buy whatever looks statistically cheap.” It is to combine a conservative numerical anchor with field research, a realizable path to value, and the flexibility to admit when the anchor itself has moved.

As of 2026-07-18. This file ranks Peter Cundill-era trades only: Cundill Value Fund, Peter Cundill & Associates / Cundill Investment Research vehicles, and clearly Cundill-led special situations before his 2009 retirement from day-to-day management and 2011 death. Later Mackenzie Cundill team positions are successor evidence, not Peter Cundill trades.

Guiding questions

1. Which trades did Cundill himself, the authorized biography, or contemporaneous fund reports identify as unusually important?
2. Which cases have reconstructable dates, thesis, sizing, drawdown, exit, and P&L rather than just a colorful anecdote?
3. Where do primary filings corroborate the holding, transaction, or exit price?
4. Which figures are repeated across sources but trace to the same Risso-Gill / Cundill source family and therefore need a [single-source cluster] flag?
5. How did the winners express the philosophy documented in the profile and philosophy files: global search, hidden assets, field work, catalysts, patience, and sell discipline?
6. Which apparent winners should be excluded because they were mistakes, successor-team holdings, allocator exposures, or legally unresolved curiosities?

Source stance and ranked map

Cleveland-Cliffs is the single best Cundill-identified trade because the best recovered version of his final interview names it as one of his best investments and explains the hidden-asset thesis, averaging down, partial sale, crash-driven repurchase, and second realization cycle (ValueBuddies repost of David Berman interview, 2011). The exact Cleveland-Cliffs cost, weight, exit, and CAGR figures are still [single-source cluster] because the detailed numbers mostly come through the authorized biography and derivative summaries, not original Canadian fund reports (Morningstar India, 2016; GuruFocus case studies, 2013).

1. Cleveland-Cliffs - single best Cundill-identified trade

Context & dates: In the mid-1980s, Cleveland-Cliffs was a dull iron-ore pellet producer in a market that preferred growth stocks. Cundill later described the stock as more than halving while investors ignored extra assets on the balance sheet (ValueBuddies repost of David Berman interview, 2011). The detailed trade accounts place the holding from roughly 1986 through exits in 1990-1991 (Morningstar India, 2016).

Thesis & how they found it: The thesis was classic Cundill: a neglected cyclical company with a hidden Michigan power-plant asset carried at a low accounting value, plus depressed sentiment in an old-economy industry (ValueBuddies repost of David Berman interview, 2011; GuruFocus case studies, 2013).

Size & structure: Secondary accounts say the fund owned more than 100,000 shares and that the position approached 4% of assets [single-source cluster] (Morningstar India, 2016).

Entry and the path - including drawdown endured: Reported figures are an initial buy near $15, a fall to about $6, and an average cost around $9.75 after further purchases [single-source cluster]. That implies roughly a 60% fall from first purchase to trough and about 39% below average cost at the low (Morningstar India, 2016).

Exit & P&L: The reported path is partial sales around $19-$20 in 1987, repurchases near $10 after the October crash, and a final exit around 1990-1991, producing more than 30% annualized [single-source cluster]. Exact dollar profit was not recovered (Morningstar India, 2016; ValueBuddies repost of David Berman interview, 2011).

What it teaches: This is the cleanest expression of Cundill's edge: hidden asset value, willingness to buy through visible drawdown, and a repeatable sell/rebuy cycle when market panic recreated the discount. The caveat is equally important: confidence comes from Cundill's own retrospective account, while the precise trade math still lacks original fund records.

Sources: ValueBuddies repost of the Berman/Report on Business interview; Morningstar India; GuruFocus case-study summary; authorized-biography source family.

2. Pan-Ocean Energy - field-work oil winner

Context & dates: Pan-Ocean was an oil company with Gabon operations that attracted Cundill's attention before Addax Petroleum's 2006 agreement to acquire Pan-Ocean subsidiaries holding substantially all operations for C$1.605 billion plus assumed net debt (Rigzone/Addax release, 2006; Lexpert transaction note, 2006).

Thesis & how they found it: The biography-derived account describes a cash-rich, debt-light company whose African reserves and production were worth materially more than the market price, with Cundill getting comfortable through on-the-ground diligence in Gabon and Tanzania (There's Always Something to Do, online copy, 2011).

Size & structure: The biography-derived account says Cundill ultimately owned about 6% [single-source cluster]. No contemporaneous Cundill filing was found to verify the exact stake.

Entry and the path - including drawdown endured: Exact entry price and interim drawdown were not recovered. The attractive feature was not a clean public-market chart but the gap between market value and asset value in a politically and operationally difficult region.

Exit & P&L: The authorized-biography source family says Cundill made roughly six times his money in about two and a half years [single-source cluster]. Addax/Pan-Ocean deal records corroborate the 2006 realization event and transaction scale, not Cundill's cost basis or fund-level profit (Rigzone/Addax release, 2006; Lexpert transaction note, 2006).

What it teaches: Pan-Ocean shows Cundill's mature method moving beyond simple net-nets: balance-sheet protection, asset appraisal, field research, and a control transaction that turned appraisal into cash.

Sources: Authorized biography source family; Addax transaction coverage in Rigzone and Lexpert.

3. Fairfax Financial - the Watsa-underwriting and credit-cycle proxy

Context & dates: Cundill's Fairfax investment began around the insurer's early-2000s pressure period and extended through the later credit-cycle payoff. The case needs careful attribution because Prem Watsa and Fairfax generated the famous credit-default-swap gains; Cundill owned Fairfax equity rather than running those hedges directly (Fairfax Q1 2008 release).

Thesis & how they found it: Cundill knew Watsa well and appears to have treated Fairfax as a deeply stressed owner-operator insurer whose book value, investment discipline, and credit-market optionality were underpriced. A Risso-Gill interview says Cundill encouraged Fairfax's CDS program and used Fairfax as a proxy when a mutual-fund mandate constrained direct CDS use (Risso-Gill interview, 2010).

Size & structure: Biography-derived accounts put a spring 2000 purchase at about $122 million with an average cost around $171 [single-source cluster]. A 2005 SEC 13G confirms Cundill-linked entities as passive beneficial owners of Fairfax shares but should not be read as activism (SEC Fairfax 13G, 2005). The Ivy 2009 annual report separately shows 46,000 Fairfax shares worth $11.868 million in that U.S. product (Ivy Funds annual report, 2009).

Entry and the path - including drawdown endured: Fairfax's public stock record was brutal in the early 2000s: the company later reported annual stock-price declines in 1999, 2000, 2001, and 2002 before the 2003 rebound, followed by additional weak years in 2004 and 2005 (Fairfax annual report, 2010).

Exit & P&L: The biography-derived path says Cundill sold about half around $217 nine months after the first purchase, later held and added around the short-seller/credit-stress period [single-source cluster]. Fairfax itself reported major Q1 2008 gains related to credit-default swaps, including $3.8 billion notional sold for $885 million of proceeds and substantial realized/mark-to-market gains (Fairfax Q1 2008 release). The exact Cundill aggregate profit was not recovered.

What it teaches: Fairfax is a great judgment case rather than a clean one-stock P&L ledger: Cundill backed a trusted capital allocator during reputational stress, indirectly captured credit-crisis convexity, and tolerated a path that many clients would have abandoned.

Sources: Risso-Gill interview; SEC 13G; Ivy 2009 annual report; Fairfax Q1 2008 release and 2010 annual report.

4. Sibir Energy - high-return special situation with messy control risk

Context & dates: Sibir Energy was a London AIM-listed Russian oil situation after the post-Soviet and post-Russian-crisis dislocations. It involved operating assets, reserve claims, partner disputes, dilution, governance conflict, and eventual Gazprom Neft interest (RFE/RL, 2009; Jones Day transaction note, 2009).

Thesis & how they found it: Biography-derived summaries present Sibir as a sum-of-the-parts oil asset trading far below reserve value, with legal and partner complexity scaring away conventional investors (GuruFocus case studies, 2013).

Size & structure: The biography source family describes Cundill's recovery vehicle using a convertible debenture and later equity exposure, including a roughly $5 million commitment and a 12% coupon in one phase [single-source cluster] (There's Always Something to Do, online copy, 2011).

Entry and the path - including drawdown endured: The path was ugly: asset disputes, joint-venture dilution, a disputed shareholder loan, an AIM suspension, and leadership turnover. Opened sources support the mess but not a precise Cundill mark-to-market drawdown.

Exit & P&L: Secondary summaries say the investment became a ten-bagger [single-source cluster]. Public transaction sources corroborate a 2009 control event involving Gazprom Neft, but not the exact Cundill exit price or fund-level P&L (RFE/RL, 2009; Jones Day transaction note, 2009).

What it teaches: Sibir was the opposite of a tidy spreadsheet bargain. The skill was recognizing that extreme legal/partner discomfort could coexist with overwhelming asset value; the luck and risk were that a politically complicated exit actually materialized.

Sources: GuruFocus case-study summary; authorized-biography online copy; RFE/RL; Jones Day.

5. Ecuador distressed debt - Brady-era sovereign bargain

Context & dates: Cundill's Ecuador debt trade ran from a first purchase in November 1992 to final sale in April 1995, during the emerging-market debt restructuring era.

Thesis & how they found it: The biography-derived account says Cundill compared Ecuador's debt to the Costa Rica Brady restructuring template, reasoning that deeply discounted sovereign paper could be worth materially more if a restructuring converted defaulted claims into more financeable instruments (There's Always Something to Do, online copy, 2011).

Size & structure: No opened source gave the capital committed, fund weight, or whether all exposure sat in the same public-market vehicle. Treat the size as [not found].

Entry and the path - including drawdown endured: Reported economics are first purchases around 20 cents on the dollar and an average cost just below 34 cents [single-source cluster]. No mark-to-market drawdown series was recovered.

Exit & P&L: The biography-derived account reports final sale in April 1995 and a 42% annualized return [single-source cluster]. Absolute P&L was not recovered.

What it teaches: The trade shows Cundill applying Graham-style appraisal to claims rather than common stocks: price, legal priority, sovereign incentives, and restructuring precedent mattered more than near-term earnings.

Sources: Authorized-biography online copy and derivative Cundill case-study summaries.

6. Tokyo Broadcasting System and the Japan bargain basket

Context & dates: Cundill's Japan work began earlier, but the reconstructable TBS/Japan basket evidence clusters around 1998-2002. Ivy's 2002 filings show Japan at 25.6% of net assets at mid-year and 35% at year-end, with Tokyo Broadcasting System at 3.9% of assets in the top-ten holdings on December 31, 2002 (Ivy semiannual report, 2002; Ivy annual report, 2002).

Thesis & how they found it: The thesis was hidden assets and cross-holdings: TBS's real estate, cash, and securities could cover much of the market price, leaving the broadcast franchise cheap. This was sourced through global screens and Japan-specific field work, matching the process in the philosophy file (GuruFocus case studies, 2013; Ivy annual report, 2002).

Size & structure: TBS was 3.9% of Ivy Cundill Global Value Fund at year-end 2002; Japan was 35% of that product's assets (Ivy annual report, 2002). These filings support the U.S. product holding, not necessarily every Cundill mandate.

Entry and the path - including drawdown endured: Secondary summaries say Cundill bought TBS below Y1,500 and sold most between roughly Y2,500 and Y3,500 [single-source cluster]. The Ivy report confirms Japan sold off sharply in Q4 2002 despite the stated margin of safety, so the basket endured material mark-to-market pressure (Ivy annual report, 2002).

Exit & P&L: The TBS per-share sale range is book-derived; no realized P&L by name was found. The 2002 Ivy product itself returned -12.17% without sales charge, so the right framing is a security-level or basket-level bargain campaign inside a volatile vehicle, not a clean fund-year victory (Ivy annual report, 2002).

What it teaches: Japan shows both sides of Cundill: he could identify balance-sheet bargains in an alien market, but governance, deflation, and timing could keep cheap assets cheap for a long time.

Sources: SEC Ivy 2002 semiannual and annual reports; GuruFocus case-study summary; authorized-biography source family.

7. Credit Foncier Franco-Canadien - hidden-asset origin trade

Context & dates: Credit Foncier was a formative mid-1970s holding during the early Cundill Value Fund period, when classic Graham bargains were still available.

Thesis & how they found it: Cundill found conservative accounting in real estate and mineral rights, estimated liquidation value above $150 per share, and began buying near $43 [single-source cluster] (authorized preview of There's Always Something to Do, 2011; GuruFocus case studies, 2013).

Size & structure: The authorized-biography source family says the group controlled more than 2% within months and that Credit Foncier plus Bethlehem Copper exceeded half of early assets at one point [single-source cluster].

Entry and the path - including drawdown endured: Entry near $43 is the recovered figure. No interim drawdown was found.

Exit & P&L: The position was reportedly sold at a multiple of cost, but no exact exit price or fund-level dollar profit was recovered (authorized preview of There's Always Something to Do, 2011).

What it teaches: Credit Foncier is the template for Cundill's balance-sheet imagination: accounting value was not accepted at face value, but reconstructed into a liquidation or control-case appraisal.

Sources: Authorized preview; GuruFocus case-study summary.

8. Bethlehem Copper - early Graham net-net with fast realization

Context & dates: Bethlehem Copper was another early 1974-era Cundill Value Fund winner, before the later global organization was built out.

Thesis & how they found it: It was a producing, debt-free mining company selling near cash value. That made it close to the pure Graham bargain that Cundill had recently learned to seek (authorized preview of There's Always Something to Do, 2011; Morningstar India, 2016).

Size & structure: The exact share count and fund weight were not recovered; biography-derived accounts place it among the early concentrated holdings.

Entry and the path - including drawdown endured: Reported average cost was about $4.50. Watsa's foreword/authorized preview and later summaries say the stock doubled within months; no major drawdown evidence was recovered (authorized preview of There's Always Something to Do, 2011; Morningstar India, 2016).

Exit & P&L: Partial profits were reportedly taken around $13 within about six months [single-source cluster]. Absolute fund-level P&L was not found.

What it teaches: Bethlehem shows that Cundill's early edge was not mysticism. He bought a solvent asset at a price that made liquidation math do most of the work, then sold when the discount closed.

Sources: Authorized preview; Morningstar India.

9. J. Walter Thompson - hidden real estate in a disliked ad agency

Context & dates: J. Walter Thompson had fallen from its post-IPO highs during a weak advertising and reputational environment in the 1970s.

Thesis & how they found it: Cundill saw an advertising agency whose hard assets and real estate were worth far more than the depressed quote suggested. Secondary summaries cite hard book value around $18 per share before assigning value to key freeholds and leases [single-source cluster] (Morningstar India, 2016; GuruFocus case studies, 2013).

Size & structure: Sources conflict in wording: one account describes the position as around 10% of fund assets, while another says Cundill controlled 4.9% of JWT equity [single-source cluster]. No original filing was recovered.

Entry and the path - including drawdown endured: The stock reportedly fell from above $20 to roughly $4 before Cundill accumulated at a little over $8. The drawdown endured by the fund after entry was not found.

Exit & P&L: Cundill reportedly sold above $20 about a year later, a double-plus on average cost [single-source cluster] (Morningstar India, 2016). Exact absolute P&L is not known.

What it teaches: JWT is a repeatable hidden-asset case: a disliked operating business can hide conservatively carried property and brand-related value, but the investor still needs a route from appraisal to market recognition.

Sources: Morningstar India; GuruFocus case-study summary; authorized-biography source family.

10. American Investment Company - secondary entry, primary exit price

Context & dates: American Investment Company was a consumer-finance company under inflation and funding pressure in the late 1970s. Cundill joined the board during a sale/merger process that ended with Leucadia's 1980 transaction, later litigated in Delaware (Justia, Dalton v. American Inv. Co., 1985).

Thesis & how they found it: The biography-derived account says Cundill bought around 200,000 shares near $3 against hard book value above $12 [single-source cluster]. The thesis fit his pattern: statistically cheap financial assets plus a possible control transaction.

Size & structure: Reported size was 200,000 common shares [single-source cluster]. The Delaware decision confirms Cundill as a named director defendant, but not his cost basis (Justia, 1985).

Entry and the path - including drawdown endured: Entry near $3 is secondary. No interim drawdown was recovered. The company faced real inflation/funding stress, which is why it needed a merger partner.

Exit & P&L: The court record confirms AIC common shareholders were cashed out at $13 per share in the Leucadia merger. If the 200,000 shares at $3 account is correct, the gross pre-cost gain would be about $2.0 million, or more than 4x capital [entry single-source; exit primary] (Justia, 1985).

What it teaches: AIC combines cheap-asset selection with governance and transaction risk. It was profitable for common shareholders, but the litigation reminds the reader that value realization can shift economics among classes of capital and create fairness disputes.

Sources: Justia court record; authorized-biography source family.

Near-winners, exclusions, and caveats

Tiffany was a profitable hidden-asset trade, with secondary accounts reporting an average cost near $8 and sale near $19 within a year; it is best framed as a good trade with a sell-too-early lesson because Avon later offered substantially more, with contemporaneous reporting supporting a $45 offer while some summaries cite roughly $50 (Morningstar India, 2016; TIME archive, 1978). Nikkei put options and the broader Japan short are dramatic but lack enough recovered size, premium, maturity, and P&L data for a top-ten entry (GuruFocus, 2013). BRL Enterprises and Control Data have SEC 13D evidence for holdings or sales, but no cost basis or realized-return reconstruction was found (SEC BRL 13D/A, 1998; SEC Control Data 13D/A, 1995).

Cable & Wireless is explicitly excluded. Cundill's last-interview account treated it as his worst investment, and the 2002 Ivy filing shows it was already the fund's worst performer while the team was still adding to the position (ValueBuddies repost of David Berman interview, 2011; Ivy semiannual report, 2002). Contemporary press reports of repeated warnings, liabilities, and write-downs reinforce why cheapness plus cash was not enough (Guardian, 2002; Guardian, 2002). Mississippi defaulted bonds are also excluded: the Bond Buyer documents Cundill's 1995 claim on 1841 Mississippi debt, but no successful public-market recovery was found (Bond Buyer, 1995). Current Mackenzie Cundill materials are successor-team evidence and should not be treated as Peter Cundill trade data.

Pattern, luck, and skill

The repeatable skill is visible across the set: start with a balance sheet or claim schedule that other investors dislike, reconstruct realizable value, investigate the people and place, insist on survival, and wait for a transaction, cycle turn, buyback, restructuring, or control event. The biggest winners were not all classic net-nets. Pan-Ocean, Fairfax, Sibir, and Ecuador show Cundill adapting the same appraisal discipline to oil assets, insurance capital allocation, Russian special situations, and sovereign debt.

The luck and limitations are just as visible. Many exact returns are [single-source cluster] figures from the authorized biography and its derivative summaries. Original Cundill Value Fund reports, Canadian filings, and complete transaction ledgers were not recovered in this run. Team and product attribution must be handled carefully: Ivy Cundill Global Value Fund was a U.S. subadvised product, Fairfax's CDS gains belonged to Fairfax, and post-2011 Mackenzie holdings belong to successors. As of 2026-07-18, Cundill is deceased; current legal searches did not surface a new posthumous securities-law development, while historical legal context remains the corporate tax appeal and Mississippi bond suit mapped in sources.md (McGill Reporter obituary, 2011; Peter Cundill & Associates Ltd. v. The Queen, 1991; Bond Buyer, 1995).

As of 2026-07-18. This is a study of Peter Cundill-era mistakes, not a complete loss ledger. No audited run of Canadian predecessor-fund reports or transaction-level records was recovered. The vivid position histories usually come from Christopher Risso-Gill's authorized biography, while SEC filings document a separate U.S. Ivy product. Exact vehicles, periods, currencies, manager rosters, and source families are therefore kept visible.

Guiding questions

1. Which losses can be quantified, and which are only self-identified mistakes or opportunity costs?
2. What did Cundill say went wrong, rather than what can be inferred with hindsight?
3. When did patience become stubbornness, and when did pressure cause premature capitulation?
4. Which errors came from leverage, cash burn, accounting, regulation, governance, timing, or legal complexity?
5. What changed in the process after each mistake?
6. Which fund drawdowns belong to Peter, a co-managed team, or successors?
7. Where do return series, trade figures, or contemporaneous and retrospective accounts conflict?
8. What do the incomplete record and celebratory source base prevent us from concluding?

Evidence and attribution stance

The best documented Peter-specific loss is Cable & Wireless (C&W). The strongest contemporaneous repeated-error case is Japanese consumer finance, but it belongs to a Peter Cundill/Hiok Hhu Ng team and a distinct U.S. vehicle. Westar, LTV, BC Capital, and the 1991 process reset are chiefly an authorized-biography source cluster. The official sponsor chart is useful for calendar returns, but its historical Series A bridge is not reconciled to the current legal series inception; annual observations also cannot establish an intrayear peak-to-trough drawdown (Mackenzie/Morningstar historical chart; current Mackenzie MRFP).

1. Cable & Wireless - the worst documented loss

The apparent margin of safety

Cundill began buying C&W in March 2000 at just over £4 after asset sales had transformed approximately £4 billion of net debt into £2.6 billion of net cash. His team estimated net asset value at £4.92 a share and treated £3.25 billion of cash and marketable securities as protection. The shares kept falling as the company issued four profit warnings in 18 months, yet the fund added. On 13 November 2002, after an emergency restructuring and a reported £4.4 billion loss, Cundill ordered the Canadian Value Fund position sold (authorized-biography online copy).

The authorized biography says just under $100 million was invested, the position was roughly 6% of the fund, and the loss approached $59 million. The reproduced text does not identify the dollar currency. These figures are [single-source cluster]: no original Canadian fund statement was found to verify the cost, weight, or loss. The book also says C&W largely caused an approximately 11% Canadian-fund decline in 2002, while Mackenzie's historical calendar chart reports -13.8%. The difference may reflect vehicle, series, measurement period, or rounding, but the public record does not reconcile it (authorized biography provenance; Mackenzie/Morningstar historical chart).

The contemporaneous record

A separate U.S. filing confirms the behavior, not the Canadian trade arithmetic. At 30 June 2002, C&W was the Ivy Cundill Global Value Fund's worst performer. The managers still called it a core holding and bought more while realizing tax losses; that U.S. portfolio held 66,000 shares worth $168,011, or 2.4% of net assets (SEC Ivy semiannual report, 2002). The U.S. fund later reported a -12.17% Class A NAV return in 2002, but its commentary also blamed currency hedging and Japanese holdings. It is not the Canadian flagship and is not a C&W-only result (SEC Ivy annual report, 2002).

Contemporaneous company evidence explains why cash did not create a liquidation floor. C&W recorded repeated warnings, large operating losses and impairments, restructuring obligations, network and property commitments, and a possible £1.5 billion tax exposure after a credit downgrade. Its later filing reported a £6.0 billion operating loss for 2003, including a £787 million September 2002 impairment, a further £1.594 billion March 2003 impairment, and £442 million of restructuring costs (Guardian, September 2002; Guardian, November 2002; Guardian, December 2002; C&W Form 20-F).

What Cundill said, and what changed

In his 2005 Ivey lecture, Cundill called C&W the team's worst investment. The decisive mistake was assigning zero value to the fiber operation when its cash burn and obligations made its contribution deeply negative. He also identified weak accounting, management ambition, emotional attachment to the thesis, and the absence of internal dissent. His explicit response was to appoint a devil's advocate when no natural skeptic emerges (Ivey lecture, about 1:08:34-1:10:30).

The case exposes a chain rather than a single forecasting error: gross cash was mistaken for distributable value; operating and contingent liabilities were insufficiently capitalized; a static appraisal anchored the team while facts deteriorated; averaging down amplified exposure; and unanimity removed a brake. The process lesson is not “never average down.” It is to recompute intrinsic value after adverse evidence, model loss-making operations below zero, trace restrictions on cash, stress leases and contingent claims, and require a written dissent before adding.

2. Westar - when client pressure defeated the thesis

Westar was a debt-heavy Canadian conglomerate that appeared to qualify as a net-net around $2. The biography reports a $1.63 average cost, says it became the fund's largest position, and describes the price falling below $1. Advisers and clients treated the holding as proof of stubbornness. Cundill sold most of it below $0.50; within months, the shares exceeded $2. Exact weight, share count, and dollar loss were not recovered, so the trade arithmetic remains [single-source cluster] (authorized-biography online copy).

The original underwriting and the exit both failed in different ways. High leverage weakened the apparent net-current-asset protection and made time costly. Then marketing and career pressure displaced a fresh appraisal at the point of sale. Cundill learned not to abandon a sound thesis merely because clients and price action are hostile. Yet the biography suggests regret over Westar later contributed to excessive persistence at C&W.

That pairing is the useful lesson. Westar was capitulation; C&W was escalation. “Be patient” cannot solve both. A durable process must distinguish thesis impairment from price pressure and must pre-commit review criteria for both selling and adding. Capital-holder patience, position size, leverage, and time to realization are part of security analysis, not external nuisances.

3. The 1989 catalogue and the 1991 process reset

Cundill's New Year self-audit named Southmark, Crazy Eddie, Japanese warrants, Westar, and BC Capital as mistakes. For BC Capital, he acknowledged being fascinated by cash flow and overlooking high debt—an explicit breach of his own leverage discipline. No reliable security-level P&Ls were recovered, so these are self-identified analytical errors rather than quantified major losses (authorized-biography online copy).

The official historical chart records 1990 at -9.4%, Cundill's first negative calendar year in that presentation, followed by +5.5% in 1991, +7.1% in 1992, and +43.1% in 1993. Compounded, 1990-1992 produced only about +2.4%; including 1993 lifted the four-year result to about +46.5%. These are point-to-point calendar calculations, not maximum drawdowns (Mackenzie/Morningstar historical chart).

The biography's reproduced internal material describes a concrete 1991 reset: reduce holdings by about one-third, increase average position size roughly 20%, start small and add only after the details satisfy, and extend the balance-sheet screen to future earning power, business quality, and management. It also records a sharper behavioral rule: do not keep reformulating a question until it yields the desired answer. Concentration itself was not the cure; the intended cure was fewer positions supported by better underwriting and repeated scrutiny.

4. LTV - correct assets, poor time appraisal

LTV demonstrates why nominal recovery and investment success are not synonyms. Cundill correctly saw valuable real estate that could support the distressed securities, but he underestimated the labor, pension, court, and negotiation complexity. LTV entered bankruptcy in 1986 and did not emerge until 1993. The biography calls the outcome dismal in time-value terms after years of analytical attention and performance drag, not a verified permanent loss of principal (authorized-biography online copy).

External records validate the omitted complexity. LTV's bankruptcy shifted enormous underfunded-pension claims to the Pension Benefit Guaranty Corporation and triggered labor, benefit, and bankruptcy litigation; one contemporaneous account put PBGC claims above $2.3 billion for 108,600 workers and retirees (Los Angeles Times, 1989; Second Circuit LTV pension litigation). The lesson is to model priority, stakeholders, legal duration, and opportunity cost alongside liquidation value. Patience without recurring time-adjusted re-underwriting is merely stubbornness with a longer clock.

5. Japan - right direction, early rotation

Cundill later acknowledged selling successful European banks too soon, entering Japan too early, and being too inflexible. By the end of 1997, the biography says the fund had more than 40% in Japan. The official chart shows -10.7% in 1998 and +33.4% in 1999; the biography instead describes approximately -9% and +16%. Those pairs are [disputed/different series or convention] and should not be silently combined (Ivey lecture; authorized-biography online copy; Mackenzie/Morningstar historical chart).

The official annual figures imply that the 1998 loss was recovered by year-end 1999. That does not erase the omission cost of selling European winners, the multi-year wait, or the career risk of a concentrated country rotation. The Nikkei short provides the same tension: Cundill was early and repeatedly paid to maintain it before Japan finally broke. The biography says those hedge losses brought him close to quitting in late 1989; because no trade ledger or independent account was recovered, this is a [single-source cluster] career-risk moment, not proof that a fund was near insolvency or closure. Being ultimately right does not make early timing free.

Tiffany is the cleanest premature-sale example. The biography says the fund accumulated roughly 3% near $8 and sold near $19 before Avon offered about $50; contemporaneous TIME reported a $45 offer. The discrepancy makes the offer price [disputed], but the behavioral result is clear: Cundill adopted a rule of selling half after a double (TIME, 1978; authorized-biography online copy). That rule reduced the regret of all-or-nothing exits, but it could also force sales unrelated to remaining discount. It is a regret-control device, not a substitute for valuation.

6. Japanese consumer finance - a co-managed regulatory miss

The U.S. Ivy record shows a miss that persisted after the team recognized it. For the fiscal year ended 31 March 2007, the Peter Cundill/Hiok Hhu Ng-managed fund returned +10.71% versus +15.45% for MSCI World. Aiful, Takefuji, and Acom were its three worst performers. Management acknowledged that political and regulatory risk had impaired the original value estimates, yet kept adding. At period-end, the three positions totaled $141.94 million by arithmetic, about 14.22% of $997.96 million net assets; Japan was 48.28% and financial services 36.23% (SEC Ivy annual report, 2007).

The next fiscal year made the repeated error clearer. The fund fell -12.07% versus -3.26% for MSCI World; Aiful and Takefuji were described as major causes of underperformance. The team still believed the worst legal issues were behind the companies. At 31 March 2008, those two positions were worth $49.51 million, a derived 8.38% of $591.03 million of net assets. Acom no longer appeared, but the filings do not disclose a security-level realized loss, so its disappearance cannot be converted into one (SEC Ivy annual report, 2008).

Attribution matters. A contemporaneous profile associates the original Takefuji work and additions particularly with Ng, who had co-managed since 2001. The case belongs to the Cundill/Ng/team process and the U.S. Ivy product—not Peter alone (Washington Post, 2005).

The regulatory change was structural. Japan's legislature passed the reform in December 2006; the regime later came fully into force after reducing the maximum rate, tightening repayment-capacity rules, and exposing lenders to vast excess-interest reimbursement claims. The Bank of Japan reported that the five largest consumer-finance companies all recorded fiscal-2006 net losses because of refund provisions and loan losses. Japan's FSA later reported refunds of ¥360.6 billion in fiscal 2007, ¥370.0 billion in 2008, and ¥388.9 billion in 2009 for three major lenders. Takefuji entered reorganization in 2010, while Aiful pursued a business-revitalization ADR plan, so outcomes were not uniform (Japan FSA, 2006 bill design; Japan FSA, 2010 enactment and implementation; BOJ Financial System Report, 2008; Japan FSA, February 2011; Japan FSA, March 2011; Aiful business-revitalization plan).

Book value, dividend yield, and insider ownership offered little protection against compulsory refunds, a lower legal price ceiling, volume constraints, and funding stress. The process failure was treating a changed legal and social compact like a temporary sentiment shock. Later outcomes confirm the impairment's severity; they do not prove that every consequence was foreseeable in 2007.

7. 2008 - a severe transition-era drawdown with shared attribution

Mackenzie's historical chart shows -25.1% for calendar 2008, the worst annual return in its 1975-2014 presentation. Combined with -1.7% in 2007, the two-year decline was about 26.4%. Even after +15.1% in 2009 and +9.9% in 2010, the series remained about 6.9% below its end-2006 level; subsequent results belong to a successor team, not a personal Cundill recovery (Mackenzie/Morningstar historical chart).

Peter's role was already transitioning. By 2008, the U.S. fund listed Cundill, Ng, Wade Burton, and Andrew Massie. Its 2009 filing presents the discussion under Massie and James Thompson, but says Massie managed the fund during the fiscal year and Thompson became co-manager in April 2009, after period-end. Contemporary team reporting says Massie had held portfolio-management responsibility for the Canadian Value Fund since 2004. The U.S. product fell -33.87%, not the index's -42.58%, and its relative defense does not cancel the absolute loss. Nor can it explain the Canadian -25.1% without a Canadian holdings ledger (SEC Ivy annual report, 2009; Advisor.ca team report).

The defensible conclusion is narrow: the Cundill franchise suffered a severe absolute drawdown during a planned team transition. It is evidence that a diversified deep-value process can still expose capital holders to long recovery periods. It is not a clean Peter-only P&L or proof of a specific security-selection error. No verified evidence that the fund or adviser was near insolvency or closure was recovered.

Behavioral root causes across the record

The cases reveal recurring tensions rather than one stable flaw:

• Anchoring and escalation: C&W's net-cash appraisal survived repeated warnings; Japanese lenders were enlarged after original estimates were impaired.
• Leverage blindness: cash flow or net-current assets distracted from debt at BC Capital and Westar.
• Patience versus stubbornness: LTV consumed years; C&W and the lenders kept attracting capital; Westar was sold too soon.
• Group and career pressure: no natural dissenter challenged C&W, while clients and marketing pushed the Westar exit.
• Factor concentration: security-level diversification did not neutralize shared Japan, financial-sector, regulatory, currency, or liquidity risks.
• Timing and regret: Japan was entered early, European banks and Tiffany were sold early, and the half-after-a-double rule managed emotion only imperfectly.
• Macro inconsistency: a bottom-up identity coexisted with large cash positions, Nikkei shorts, country rotations, and currency hedges whose effect changed by regime.

These are not proof that the method failed. They show that a cheap balance sheet is an input, not a safeguard, when the business consumes cash, law rewrites economics, leverage shortens the clock, or the team cannot challenge its own narrative.

Process changes that are actually documented

Three post-error changes have direct support rather than retrospective inference:

1. After early selling errors, sell half after a double rather than making an all-or-nothing exit (authorized-biography online copy; TIME, 1978).
2. After the 1990-1992 slump, reduce the number of holdings, increase average position size only with deeper work, start small, and add earnings quality and management to the balance-sheet screen (authorized-biography online copy).
3. After C&W, appoint a devil's advocate when dissent does not arise naturally and value loss-making operations and liabilities explicitly (Ivey lecture, about 1:08:34-1:10:30; authorized-biography online copy).

Separately, the journals and lecture support a standing discipline—keep the original analytical question fixed, revisit facts continuously, and distinguish price decline from thesis impairment—but the public record does not establish that discipline as a dated response to one loss (authorized-biography online copy; Ivey lecture).

The Japanese consumer-finance record does not reveal a Peter-specific postmortem or process change. Inventing one would sanitize the evidence. The documented sequence instead shows recognition of impairment, continued additions, another year of losses, and later industry restructuring.

Legal, governance, and record limits

Bounded searches through 2026-07-18 found no credible new personal securities-enforcement action, but that is not a lifetime clean-record claim. The Federal Court of Appeal dismissed a corporate tax appeal and treated Cundill as the directing mind of Canadian and Bermuda entities in the transaction; it was not a securities-fraud finding (Peter Cundill & Associates Ltd. v. The Queen). In an American Investment Company merger case, preferred holders alleged fiduciary breach against directors including Cundill, but Delaware Chancery entered judgment for the defendants; the allegation must not be restated as a finding (Dalton v. American Investment Co.). A 2006 adviser Form ADV contained no filed disciplinary-response pages as of that entity and date—useful but jurisdiction- and time-bounded (SEC/IAPD Form ADV).

An Ontario Securities Commission bulletin separately records a “Consent to Suspension” under a rule governing surrender of registration for Peter Cundill & Associates (Bermuda) Ltd., effective 18 September 2006. It was an administrative registration surrender contemporaneous with the Mackenzie acquisition/transition; without an adverse enforcement order, it should not be recast as a sanction (OSC Bulletin, 2006; Ivey advisory-board biography and transition context).

Finally, the record is selected. The biography was authorized and written by a former Cundill director using journals, interviews, and speeches. Sponsor materials foreground the surviving flagship, and derivative articles recycle the same attractive cases. Without a complete trade ledger, audited predecessor-series bridge, monthly drawdowns, and comparable records for all vehicles, this chapter can identify failure mechanisms but cannot calculate a comprehensive hit rate, loss ratio, or Peter-only track record.

As of 2026-07-18. These 31 short fragments preserve Peter Cundill's public voice without pretending that the surviving record is a complete transcript of his career. They include direct interviews and a recording, contemporaneous fund commentary, institutional statements, and journal material selected by his authorized biographer. The labels are part of the evidence: a Peter-only interview is not the same thing as an unsegmented discussion signed by two or four managers.

Provenance and Editorial Method

• Every displayed fragment is 25 words or fewer. Across each underlying source family—not each mirror, repost, excerpt, or URL—the combined quotation is also capped at 25 words. The largest family total is 25.
• Words are reproduced without ellipses, bracketed repairs, or silent substitutions. Initial capitalization and terminal punctuation may be normalized when a clause is lifted from a longer sentence.
• Direct interview means the publisher visibly identifies Cundill as the interviewee. Direct recording; caption-exact means two Ivey encodings expose the same caption wording; it is not represented as an audio-forensic transcript.
• Joint manager interview/discussion means the filing names every manager but does not identify the author of individual answers. The collective “we” is retained and never converted into “Cundill said.”
• Journal excerpt via authorized editor separates the underlying composition date from Christopher Risso-Gill's 2011 or 2014 publication date. The raw journals are not publicly available for comparison.
• Press-attributed remark, institutional statement, reported speech, and affiliated attributed adage describe weaker or more mediated chains. Context follows each fragment so that a memorable line is not mistaken for a timeless rule.

Value, Selection, and Cash

1. “Securities trading below cash” - Ivey Value Investing class, 2005. Cundill was recalling the unusually fertile net-net opportunity set of the 1970s, not promising that every later cash discount would work. [direct recording; caption-exact across two Ivey encodings] (Official video, 01:43).
2. “Cable & Wireless remains undervalued based on our intrinsic estimated valuations” - Ivy semiannual report, June 2002. The sentence records confidence before Cundill later called the position his worst investment. [direct Peter interview, speaking for the fund team] (SEC filing).
3. “The Fund bought more of Cable & Wireless at current levels” - Ivy semiannual report, June 2002. Read beside the later loss, the fragment shows why contemporaneous material matters: hindsight had not yet repaired the thesis. [direct Peter interview, speaking for the fund team] (SEC filing).
4. “we were fully hedged on international currency exposure” - Ivy annual report for 2002. The hedge hurt the U.S. product as foreign currencies appreciated; this is a period policy, not a universal personal rule. [direct Peter interview, speaking for the fund team] (SEC filing).
5. “we took the opportunity to upgrade the Fund's Japanese holdings” - Ivy annual report for 2002. Weak markets were used to replace positions judged fairly valued with cheaper candidates. [direct Peter interview, speaking for the fund team] (SEC filing).
6. “market volatility was our friend during the fiscal year” - Ivy annual report for the year ended March 2005. Volatility supplied both purchases and sales, while high cash still detracted from performance. [joint manager interview: Peter Cundill and Hhu Ng] (SEC filing).
7. “mergers and acquisitions remain a vital catalyst for unlocking hidden shareholder value in Japan” - Ivy annual report for the year ended March 2005. The claim belongs to a period of slowly changing Japanese governance. [joint manager interview: Peter Cundill and Hhu Ng] (SEC filing).
8. “We will not invest for the sake of being fully invested” - Ivy annual report for the year ended March 2006. Cash was the consequence of unmet standards, not a requirement to forecast the market. [joint manager interview: Peter Cundill and Hhu Ng] (SEC filing).
9. “uncovering value ideas remains challenging” - Ivy annual report for the year ended March 2006. The phrase qualifies the report's 79%-of-estimated-intrinsic-value portfolio figure and its willingness to hold more cash. [joint manager interview: Peter Cundill and Hhu Ng] (SEC filing).

Risk, Error, and Flexibility

10. “We were blindsided by political risk in the Japanese consumer finance sector” - Ivy annual report for the year ended March 2007. This is an explicit joint admission; separate reporting particularly associates the initial Takefuji work with Ng. [joint manager discussion: Peter Cundill and Hhu Ng] (SEC filing).
11. “we have continued to build our positions in the Japanese consumer finance sector” - Ivy annual report for the year ended March 2007. The report pairs recognition of the miss with continued averaging down, preserving the unresolved judgment at that date. [joint manager discussion: Peter Cundill and Hhu Ng] (SEC filing).
12. “we feel it is very difficult to fully understand all the risks inherent in their balance sheets” - Ivy annual report for the year ended March 2008, on global banks. [joint discussion: Peter Cundill, Hhu Ng, Wade Burton, and Andrew Massie] (SEC filing).
13. “We do not pretend to know” - Ivy annual report for the year ended March 2008, answering whether the global correction was ending or beginning. The team treated bottom-up valuation as a way to avoid a forced macro answer. [joint four-manager discussion] (SEC filing).
14. “1991 was the year of self-inflicted torture. So what did I do, or learn? I relearned my craft.” - retrospective reflection on 1991, reproduced in Routines and Orgies (2014), chapter 24, p.173. The visible source does not disclose the diary entry's composition date. [Cundill reflection via authorized biographer] (Google Books search record).
15. “I didn't fall under the spell of a beautiful Japanese maiden and then get going on Japan.” - Forbes, May 2004. The joke rejected a romanticized explanation for his early Japanese buying; the same profile records that his market short began years too soon. [direct quote in a reported profile] (Article).
16. “Be flexible.” - Ivey class, 2005. This was part of a compact closing sequence about adapting tactics within a disciplined framework. [direct recording; caption-exact] (Official video, 29:29).
17. “Be international.” - Ivey class, 2005. It summarizes a global search practice, not a mechanical diversification quota. [direct recording; caption-exact] (Official video, 29:34).
18. “On a measured basis, be patient.” - Ivey class, 2005. “Measured” matters: Cundill's own loss history shows that waiting without re-underwriting can consume the margin of safety. [direct recording; caption-exact] (Official video, 29:43).

Career, History, and Criteria

19. “I have been a practicing portfolio manager since the beginning of 1975.” - The Wall Street Transcript, listed October 2001. The publisher exposes an excerpt but not the full interview, so this supports the career boundary rather than a complete chronology. [direct edited interview; excerpt only] (Publisher excerpt).
20. “I think there's an analogy between the two disciplines – both study the past to understand the present and predict the future.” - McGill Reporter, April 2008, on finance and history. [direct interview] (Article).
21. “I don't care what a company does or where it does it as long as it meets my criteria.” - undated adage preserved by the Peter Cundill Foundation. Its date and original document are not supplied. [affiliated attributed adage] (Foundation biography).

Patience, Opportunity, and Time

22. “No fortunes are made in prosperity. Ours is a marathon without end. Enjoy the passing moments.” - Career Achievement Award acceptance, December 2001, reproduced in a 2002 Mackenzie/Fund Library feature. [reported reproduction of direct speech] (Feature mirror).
23. “It has been increasingly difficult to find bargain issues on a global basis that have made the fund so successful.” - March 2006 remark explaining why the Cundill Value Fund would restrict new purchases. [press-attributed remark] (Advisor.ca report).
24. “The mantra is patience, patience and more patience.” - Cundill's final interview, published by The Globe and Mail on January 27, 2011. The interview occurred before his January 24 death; its precise recording date is not disclosed. [direct edited Q&A] (Original interview).
25. “Think long-term and remember that the big rewards accrue with compound annual rates of return.” - final Globe and Mail interview, published January 2011. It was advice about choosing researched managers and remaining with them, not a guarantee about every manager or holding. [direct edited Q&A] (Original interview).
26. “We are having a difficult time finding anything we want to buy. Don't send me your money!” - 1985 fund annual meeting, preserved in Michael Meighen's 2011 Senate memorial. No original meeting transcript was recovered. [reported direct speech via memorial] (Official Senate Debates, p.1705).
27. “this is what I want to do for the rest of my life!” - Cundill's December 1973 journal entry after discovering Graham-style liquidation value, reproduced in There's Always Something to Do (2011), chapter 1, p.4. [journal excerpt via authorized editor; 1973 composition / 2011 publication] (Google Books page).
28. “Curiosity is the engine of civilization.” - reproduced in the authorized book's “What Makes a Great Investor?” chapter, p.185. The visible text does not identify a source document or composition date. [Cundill fragment via authorized editor; medium/date unknown / 2011 publication] (Google Books search record).

Team and Stewardship

29. “one of the most successful partnerships in Canadian financial services” - Mackenzie acquisition announcement, August 2006. Cundill described the transaction as continuity rather than an abrupt change of investment identity. [direct institutional statement] (News release).
30. “focus on doing what we do best” - Mackenzie acquisition announcement, August 2006. The phrase introduces security selection as the team's continuing work after the transaction. [direct institutional statement] (News release).
31. “selecting securities where financial returns can be realized” - Mackenzie acquisition announcement, August 2006. Cundill connected the organizational transaction to outcomes for clients and fund investors. [direct institutional statement] (News release).

Annotated Index of Primary and Near-Primary Materials

Recordings and speeches

• March 2005 - Ivey Value Investing class. The sole verified full direct recording found. Ivey's current guest-speaker index establishes institutional provenance and links the canonical 74-minute lecture; an alternate Ivey upload is the same source family. The talk covers net-nets, Sweden and Japan, screening, fieldwork, valuation, selling, cash, derivatives, Cable & Wireless, and catalysts. Automatic captions are useful locators but remain imperfect (Ivey index; canonical video; alternate Ivey encoding).
• December 2001 - Career Achievement Award acceptance. A Mackenzie/Fund Library feature reproduces Cundill's short acceptance poem and excerpts from an earlier internal short-selling memo. The original event recording and memo were not recovered, so the feature is an evidentiary bridge rather than a raw archive (Feature mirror).
• 1985 - Cundill Value Fund annual meeting. The “Don't send me your money” warning survives through Michael Meighen's official 2011 memorial statement. No original audio, transcript, or annual-meeting booklet was located (Senate Debates).

Interviews and institutional statements

• October 23, 1980 - Peter C. Newman interview. McMaster University's catalogue records a 22-page transcript and audio cassette 32 in the Newman fonds. Neither is digitized; the entry is the earliest located interview record and is indexed without quotation (Catalogue).
• October 2001 - The Wall Street Transcript. The publisher exposes only a short excerpt and bibliographic date. It establishes a direct interview but cannot support claims about the inaccessible remainder (Excerpt).
• June and December 2002 - Ivy manager interviews. Both reports explicitly identify Peter as the interviewee and management-team leader. They provide rare contemporaneous Peter-attributed wording about selection, currency hedging, Japan, and Cable & Wireless in one U.S. product (Semiannual; annual).
• May 2004 - Forbes, “Found in Translation.” The original profile contains clearly attributed remarks about Japan and the early market short. It is reported journalism, not a trade ledger (Article).
• March and August 2006 - fund closing and Mackenzie acquisition. These period statements explain capacity pressure and Cundill's rationale for formalizing the Mackenzie relationship (Closing report; acquisition release).
• April 2008 - McGill Reporter. A short direct interview connects Cundill's interest in history with investment research and documents the new history prize (Article).
• January 2011 - final Globe and Mail interview. David Berman's Q&A supplies Cundill's last published advice to ordinary investors. Publication followed his death; the precise interview date remains undisclosed (Interview).

Named-manager fund commentary

• March 2005 and March 2006 annual reports. Each section is an unsegmented joint interview with Peter Cundill and Hhu Ng. They document scarcity, cash, concentration, Japanese governance, and consumer finance, but no individual sentence can be assigned to either manager (2005; 2006).
• March 2007 annual report. The amended filing identifies Cundill and Ng jointly and captures both the Japanese regulatory-risk miss and continued additions. Separate contemporaneous reporting associates the original Takefuji decision particularly with Ng, making Peter-only attribution especially unsafe (Report).
• March 2008 annual report. The discussion names Cundill, Ng, Wade Burton, and Andrew Massie and says Burton and Massie joined in December 2007. It is a transition-team document, not a personal letter (Report).

Authorized books and controlled archives

• 2011 - There's Always Something to Do. McGill-Queen's University Press says Christopher Risso-Gill assembled the narrative from interviews, speeches, and exclusive access to Cundill's 45-year daily journal. The PagePlace preview and Google Books pages expose limited exact material; the JSTOR edition is controlled. Cundill's journal date must be separated from the book's publication date (Publisher; authorized preview; JSTOR).
• 2014 - Routines and Orgies. The fuller biography is also authored by Risso-Gill and copyrighted by the Cundill Foundation. Its authorized preview says the editor used 44 years of journals and worked daily with Cundill on both books. Publisher and catalogue page totals vary by format, so printed chapter/page anchors are preferable (Distributor; authorized preview; JSTOR).
• Private journals and notebooks. No public fonds or raw electronic run was found. The 2011 book describes 45 years, the 2014 book 44, and a facsimile caption spans 1963–2007. Those descriptions are reported rather than forced into an audited inventory.

Located but not quote-ready

• A 1989 Forbes remark, 1996 short-selling memo, 2004 Canadian Business interview, Barron's profiles, The Province material, multiple Outstanding Investor Digest appearances, and the search-indexed 2009 chairman-emeritus announcement were located only as bibliographic records, partial excerpts, later reproductions, or a currently unavailable article. They remain index leads, not quotation sources (1989 syndication; Investment Executive author archive).
• No authentic Peter Cundill podcast appearance was found. Later podcasts and videos feature biographers, protégés, investors discussing him, or posthumous narration.

Reading Boundaries

This is a changing and sometimes uncomfortable voice. The 2002 Cable & Wireless conviction precedes Cundill's later admission that the investment was his worst. The 2007 joint report says the managers were blindsided and still added. The 2008 transition team declined to forecast the correction while acknowledging that bank risks were hard to measure. Those tensions are evidence, not defects to edit away.

The public record also has sharp limits. The Canadian flagship's complete manager-letter run and trade ledger were not recovered. The Ivy reports describe a distinct U.S. vehicle with changing co-managers. Edited interviews select; institutional releases polish; memorials remember; automatic captions mishear; Risso-Gill chose and arranged private material. A source can be authorized and still mediated.

Finally, repetition is not provenance. Quote sites and derivative articles frequently assign Cundill sayings associated with Buffett or Templeton. Even the title phrase “there's always something to do” is credited in the authorized 2011 book to Irving Kahn's advice, not presented as an original Cundill maxim. Those formulations were excluded unless an attributable source-visible chain survived review.

As of: 2026-07-18T04:34:48Z
Task: T0475 | Investor: 059-peter-cundill | Code: F-key-writings

Bottom line

Peter Cundill did not leave the kind of public authored corpus that Warren Buffett, Howard Marks, or Seth Klarman did. The practical Cundill reading list is a mediated corpus: fund commentary and interviews where Cundill or the Cundill team spoke directly, an official Ivey lecture video, final press Q&A material, a 1976 investor-letter checklist known through secondary recovery, a net-net worksheet preserved as an appendix, and two Christopher Risso-Gill books built from Cundill's journals, speeches, interviews, and papers. The two major books are therefore near-primary in parts, but they are not books by Cundill. The safest classification is: Cundill's own written/oral record is fragmentary; Risso-Gill's books are the best organized guide to that record.

The most useful first reading is still There's Always Something to Do: The Peter Cundill Investment Approach by Christopher Risso-Gill. McGill-Queen's University Press describes it as a narrative drawn from interviews, speeches, and exclusive access to Cundill's 45-year daily journal, and its table of contents is explicitly organized around Cundill's investing approach, cases, mistakes, and appendices including a net-net worksheet and fund holdings MQUP. The best compact direct-source complement is the June 30, 2002 SEC-hosted Ivy Funds semiannual report, whose "Zigging While Others Are Zagging: The Cundill Story" section summarizes the process in Cundill/team language: deep value, margin of safety, 50% or greater discount to net asset value, sum-of-the-parts work, management checks, competitor checks, catalysts, and three-to-five-year patience SEC 2002 semiannual.

Works by Cundill or closest direct corpus

1. The 1976 All-Canadian Venture Fund letter to unitholders

Classification: Cundill-authored investor communication, but the original public filing or letter scan was not recovered in this run. The usable evidence is secondary reproduction of the criteria, principally Net Net Hunter's article attributing the checklist to the letter after Cundill took over the All-Canadian Venture Fund Net Net Hunter. Treat this as a high-value lead and as a core statement only with the caveat that the original letter still needs archival recovery.

Central thesis: the fund would pursue a Graham-style discipline anchored in balance-sheet cheapness, earnings durability, dividends, and conservative leverage, rather than stories, macro forecasts, or popular themes.

Key ideas: first, price had to be below book value and preferably below net working capital after long-term debt. Second, price mattered relative to history: ideally less than half the former high and near an all-time low. Third, earnings quality still mattered, so the approach was not "cheap at any price." Fourth, dividends were a behavioral and financial sanity check. Fifth, debt had to be judicious rather than merely available. Sixth, the letter gives the earliest clean public version of Cundill's operating formula after he acquired the fund. Seventh, the checklist helps prevent later romanticized accounts from turning him into a pure adventurer; his public framework began as very dry accounting work.

Best sections to recover: the opening statement of mandate, the checklist itself, and any language on liquidation value, sell discipline, and fund risk. Until recovered, use the reproduced criteria as a lead, not as definitive archival text.

2. The net-net worksheet, designed in 1975

Classification: Cundill-designed analytical work product preserved in book appendices. There's Always Something to Do lists "Appendix 3: Net-net work sheet" at pages 225-226 MQUP, while Routines and Orgies lists "Appendix A: Net-Net Work sheet" at page 395 MQUP. Search results from the De Gruyter/JSTOR family also describe it as a worksheet designed by Peter in 1975, but full appendix text was not available through clean public access in this run.

Central thesis: cheapness had to be measured through a repeatable asset test, not merely felt through low price multiples.

Key ideas: the worksheet turns Graham's net-current-asset concept into an operating checklist for a fund. It forces the analyst to inventory realizable current assets, liabilities, debt, and working-capital protection before moving to softer judgments. It links Cundill's public persona to a reproducible research artifact. It also shows how "global value" began with a portable accounting template: the same core calculation could be carried from Canada to Japan, Europe, emerging markets, distressed debt, and special situations. The main weakness is that the public record tells us the appendix exists and where to find it, but does not provide a clean, citable full transcription.

Best sections: Appendix 3 in There's Always Something to Do and Appendix A in Routines and Orgies. These should be page-checked in a physical, library, JSTOR, or publisher copy before reproducing the worksheet in any future canon artifact.

3. "Zigging While Others Are Zagging: The Cundill Story" and the 2002 Ivy interview

Classification: SEC-filed fund communication containing Cundill/team explanations, likely edited for fund-report compliance. It is not an academic article and not a private investor letter, but it is the cleanest official public compact of the process SEC 2002 semiannual.

Central thesis: Cundill's team tried to buy securities at a large discount to underlying value, with risk reduced by asset backing and patience rather than by forecasting.

Key ideas: the target was often a discount to net asset value of 50% or more. The process was bottom-up and security-specific, not based on economic forecasting, market timing, or benchmark country/sector weights. The team built sum-of-the-parts, break-up, and liquidation analyses. Research included financial statements, management visits, competitor conversations, catalysts, and market/liquidity events. Holding periods were normally measured in years. The report also makes attribution boundaries clear: Ivy Cundill Global Value Fund was managed by the Cundill & Associates team, with Peter as team leader, not by Peter alone.

Best sections: the paragraphs headed "Buying a Dollar Bill for 50 Cents," "Stock Selection: Priority #1," and "Not Following the Crowd." These are the most efficient direct reading for a new researcher trying to understand Cundill in ten minutes.

4. Ivy Cundill Global Value Fund annual-report interviews, especially 2002

Classification: SEC-filed Q&A and commentary attributed to Peter Cundill as management team leader of Ivy Cundill Global Value Fund SEC 2002 annual.

Central thesis: the process was durable enough to discuss in a down year, but it did not eliminate drawdowns, currency costs, or country-specific pain.

Key ideas: the 2002 annual report is valuable because it shows live adversity rather than victory-lap biography. The fund returned negative absolute results in 2002, even while outperforming broad comparisons; currency hedging hurt results when foreign currencies appreciated; Japanese holdings were not spared when fourth-quarter sentiment deteriorated; and the team framed weakness as a chance to upgrade the portfolio. The same Q&A restates the core 2003 strategy: bottom-up stock selection, deep value, Benjamin Graham principles, and global opportunism. For the canon, this is a useful antidote to the simplified "net-net equals low risk" myth: margin of safety can widen while quoted performance is still ugly.

Best sections: the 2002 performance Q&A, the discussion of Japan and currency hedging, and the "strategy for 2003" answer. Later SEC reports, especially 2009 and 2012, should be labeled as co-manager or successor-team material rather than Cundill's own current words.

5. 2005 Ivey Value Investing Classes lecture

Classification: direct public lecture, officially indexed by Ivey's Ben Graham Centre. Ivey lists Peter Cundill, Principal of The Cundill Group, as the March 28, 2005 Ivey Value Investing Classes speaker and links the video Ivey video library; the guest-speaker page also records that Cundill founded Peter Cundill & Associates in 1977 and identifies the 2005 presentation Ivey guest speakers.

Central thesis: Cundill presented value investing as a practical craft of independent valuation, patience, and historical perspective.

Key ideas: this is the best direct oral source because it lets researchers hear Cundill rather than read later narration. It anchors his public teaching role at a Graham-focused institution. It should be used to verify tone, emphasis, and examples from the books. It also helps distinguish Cundill's own speaking style from Risso-Gill's narrative architecture. The limitation is important: no official transcript was recovered in this run, so exact quotation requires manual timestamping against the video. Until then, cite the video for existence and topic, not for precise wording.

Best sections: opening framing of Graham/value discipline, any case examples, and Q&A. A future task should produce a timestamped transcript map rather than relying on auto-captions.

6. "Notes from a Battle-Scarred Warrior"

Classification: Cundill-authored philosophy document preserved as Appendix B in Routines and Orgies. JSTOR's visible chapter metadata says chapter 27, "Now We Begin Again," describes Cundill writing it after an Aspen Institute session as the most defining restatement of his investment philosophy since the 1976 letter, incorporating the innovations he had developed from Graham over roughly twenty years JSTOR. MQUP's table of contents places "Notes from a Battle-Scarred Warrior" in Appendix B at page 399 MQUP.

Central thesis: Cundill's mature value investing had moved beyond static Graham net-nets into a broader, battle-tested process while retaining the same margin-of-safety foundation.

Key ideas: this appears to be the most important single Cundill-authored investment document after the 1976 letter. It likely restates process after two decades of mistakes, organizational change, international work, and market cycles. Its placement in Routines matters: the biography treats the document as a hinge between early formula and mature judgment. For canon purposes it should become a priority archival target. The current limitation is access: the public JSTOR/MQUP pages verify existence, significance, and location, but not the full text.

Best sections: Appendix B in Routines and Orgies and the surrounding chapter 27 discussion. Do not cite long excerpts from podcast transcripts or highlight pages purporting to quote the appendix.

7. The daily journals, 1963-2007, as edited corpus

Classification: private Cundill writing mediated through Risso-Gill. Rational Walk reports that Cundill kept a daily journal from 1963 to 2007 and that Risso-Gill, a Cundill Value Fund director for ten years, had exclusive access Rational Walk. MQUP similarly says There's Always Something to Do draws on a daily journal kept for forty-five years MQUP; MQUP says Routines and Orgies is supported by four decades of meticulously kept daily journals MQUP.

Central thesis: the journals are the closest thing to Cundill's intellectual autobiography, but they are not publicly available as a standalone archive.

Key ideas: the journals are unusually valuable because they were contemporaneous, not retrospective. They preserve his thinking about mistakes, travel, illness, relationships, markets, reading, and temperament. They explain why Cundill's process cannot be reduced to a screen: the books repeatedly connect accounting discipline to curiosity, fieldwork, and personal routines. They also create a bias risk, because the reader sees the journals through an editor who was close to the foundation and the fund. Future work should always distinguish "Cundill wrote in his journal" from "Risso-Gill narrates that Cundill believed."

Best sections: There's Always Something to Do chapters 1, 3, 4, 5, 10, 11, 12, 16, 18, and 21; Routines and Orgies chapter 4, chapters 10-16, chapters 26-30, chapters 35-36, chapters 37-41, and the appendices.

8. David Berman / Report on Business final interview

Classification: direct interview, but currently best accessed through a ValueBuddies repost that says the article was taken from CTV News and that David Berman interviewed Cundill for the February issue of Report on Business magazine ValueBuddies. The Senate of Canada memorial statement also quotes the final-investor-advice answer, adding official corroboration for at least that portion Senate of Canada.

Central thesis: late in life, Cundill described his own investing through formative loss, best/worst cases, manager selection, margin of safety, patience, and long-term compounding.

Key ideas: the interview is a compact self-portrait. It records his first investment loss as a McGill-era speculative mining purchase. It frames Cleveland-Cliffs as a model of extra assets ignored by the market. It names Cable & Wireless as a severe mistake, a useful direct warning that low price, no debt, cash, and established networks did not prevent things from going wrong. It shows that semi-retired Cundill shifted from stock-picking to selecting value-oriented managers. It also contains practical advice to ordinary investors: do homework on managers, stick with them, and think long-term. Because the available text is a repost with encoding damage and the original Globe/CTV pages were not recovered, use it for paraphrase and verify exact wording before quoting.

Best sections: first investment, best investment, worst investment, current opportunities, and advice for regular investors.

9. Cundill firm-level proxy voting policy

Classification: firm policy, not necessarily Peter-authored prose, but it is SEC-filed and useful for governance posture. The 2006 Ivy filing includes a "Peter Cundill and Associates Proxy Voting Policy" stating that the firm votes proxies in clients' best economic interest over the long term, while considering governance, disclosure, and conflicts SEC SAI 2006.

Central thesis: portfolio ownership carried fiduciary voting duties tied to long-term client economics.

Key ideas: the policy defines client interest economically rather than politically. It supports voting with management unless doing so would impair investment merits. It recognizes conflicts and case-by-case judgment. It is useful because Cundill often invested in companies with hidden assets, control issues, liquidation value, or governance frictions. Its limitation is authorship: cite it as firm policy and compliance evidence, not as Cundill's personal manifesto.

Best sections: compliance objective, general principles, client best interest, material conflicts, and case-by-case voting policy.

Best works about Cundill, ranked

1. Christopher Risso-Gill, There's Always Something to Do

Why it ranks first: this is the core Cundill investment book. It is official-publisher material, published by McGill-Queen's University Press in 2011, and explicitly framed as the Peter Cundill investment approach MQUP. Google Books confirms the same title, author, publisher, and 2011 metadata Google Books. Internet Archive confirms a controlled-lending bibliographic record, but the item is access-restricted Internet Archive.

Central thesis: Cundill converted Graham's balance-sheet logic into a global, opportunistic, research-heavy value practice. The book's strength is not only the record but the sequence: discovery of Graham, fund launch, cases, crashes, distressed debt, Russia, Japan, retirement, and reflections on what makes a great investor.

Key ideas: margin of safety is the spine. The net-net worksheet is the method's earliest hard tool. The travel habit widened the opportunity set. The best cases often involved ignored assets, not glamorous earnings growth. The mistakes are essential reading because they test the method's limits. Cash and sell discipline came from valuation work, not pure market timing. Cundill's edge combined accounting with curiosity. The book is also affiliated: Risso-Gill was a Cundill Value Fund director for ten years, and the preview copyright page shows the Cundill Foundation connection, so it is informed but not detached.

Best chapters: Foreword, 1 "The Eureka Moment," 3 "Launching a Mutual Fund on Value Principles," 4 "Value Investment in Action," 5 "Going Global," 10 "Surviving a Crash," 11 "Distressed Corporate Securities and Defaulted Sovereign Debt," 12 "Dealing with Adversity," 16 "There's Always Something Left to Learn," 18 "The Russian Bear," 21 "What Makes a Great Investor?," and appendices 2-5.

2. Christopher Risso-Gill, Routines and Orgies

Why it ranks second: this is the best fuller biography and temperament source. MQUP describes it as a biography supported by four decades of Cundill's daily journals and says Risso-Gill was a close friend and confidant MQUP. Google Books confirms publication by McGill-Queen's University Press in 2014, with Risso-Gill as author Google Books. JSTOR provides the table of contents and visible chapter snippets, including the importance of "Notes from a Battle-Scarred Warrior" JSTOR.

Central thesis: Cundill's investment record cannot be separated from the habits, appetites, discipline, journaling, friendships, physical challenges, illness, and intellectual restlessness that shaped his judgment.

Key ideas: routine and intense experience coexisted. Journaling was both self-discipline and decision evidence. Physical fitness, travel, and reading were part of the investment engine, not lifestyle decoration. The book helps explain why Cundill could travel to hated markets and still remain methodical. It also gives more texture to organizational tensions, late-career transition, and the post-illness years. Its weakness for Task F is that it is less an investment manual than a life biography, so its investment sections must be selected deliberately.

Best chapters: Introduction, 4 "The Early Journals: A Sampler," 10 "Epiphany and a Fallen Idol," 11 "The Cundill Value Fund," 16 "Crisis and Opportunity," 26 "Reality Checks," 27 "Now We Begin Again," 29 "The Perfect Balance," 30 "The Turn of the Tide," 35 "Triumph and Disaster," 36 "The Turn of the Screw," 37-41 for late-life context, Appendix A, and Appendix B.

3. Ivey Ben Graham Centre materials

Why it ranks third: Ivey provides the cleanest official institutional access point to Cundill speaking directly. The guest-speaker page establishes his role and the 2005 presentation; the video library places him in the March 28, 2005 Value Investing Classes series Ivey guest speakers Ivey video library.

Best use: cite the Ivey pages for provenance, then use the video only after timestamp verification. For future work, a timestamped transcript would be more valuable than another summary of the Risso-Gill books.

4. SEC-hosted Ivy/Cundill filings

Why it ranks fourth: SEC filings are not literary sources, but they are official public records and the best antidote to anecdotal biography. The 2002 semiannual report summarizes the process; the 2002 annual report shows live performance stress; the 2006 SAI documents entity history, subadviser status, and the firm proxy policy SEC 2002 semiannual SEC 2002 annual SEC SAI 2006.

Best use: extract process statements, attribution boundaries, risk disclosures, manager/team separation, and live-period comments. Do not turn U.S. Ivy fund performance into Cundill Value Fund performance.

5. Rational Walk review of There's Always Something to Do

Why it ranks fifth: Rational Walk is the best independent practitioner review recovered in this run. It identifies Risso-Gill's access, highlights the daily journals, and stresses case-study usefulness while also noting behavioral issues such as sell discipline and cash generated by valuation-driven sales rather than market timing Rational Walk.

Best use: use for independent framing and reading guidance, not as the source of exact Cundill quotations. It quotes long book excerpts; future canon work should page-check the book directly.

6. Institutional memorial and philanthropy sources

Why it ranks sixth: the Peter Cundill Foundation biography is useful for official biographical facts, track-record framing, and personality attributes, while the Senate memorial provides a dated public record of death, philanthropy, and final-interview advice Peter Cundill Foundation Senate of Canada. The Cundill Prize press archive is useful for source discovery, including links to obituaries and the Globe final-interview item Cundill Prize archive.

Best use: background and current-status corroboration. These are affiliated or memorial sources, so they should not be used for critique without independent support.

7. Business Insider / Daily Reckoning, Net Net Hunter, Morningstar India, GuruFocus, and other practitioner summaries

Why they rank lower: these sources are useful reader guides and source-discovery maps, but most recycle the Risso-Gill books, the final interview, or each other. Business Insider's Chris Mayer piece usefully frames the book and identifies Risso-Gill as a Cundill Value Fund director for ten years Business Insider. Net Net Hunter is particularly useful for the reproduced 1976 checklist Net Net Hunter. But none should outrank MQUP, JSTOR, SEC filings, Ivey, or the book itself.

Best use: as pointers to original passages, candidate claims, and practical explanations for readers. Avoid using quote aggregators, Goodreads, podcast transcript pages, Substack summaries, Scribd mirrors, or bootleg full-text copies as citation anchors.

Reading order for a new researcher

1. Read the 2002 SEC semiannual "Cundill Story" first for the process skeleton: discount to NAV, sum-of-the-parts, liquidation value, fieldwork, catalysts, and patience SEC 2002 semiannual.
2. Read There's Always Something to Do chapters 1, 3-5, 10-12, 16, 18, and 21, then appendices 2-5 MQUP.
3. Read Routines and Orgies chapters 10-16, 26-30, 35-36, and appendices A-B to see how the process changed through organizational and personal stress MQUP.
4. Watch the 2005 Ivey lecture and build a timestamped note map before quoting Ivey video library.
5. Read the final Berman/Report on Business Q&A through the ValueBuddies repost only as a provisional text, using Senate as corroboration for the final-advice segment ValueBuddies Senate of Canada.

Evidence gaps and cautions

The main archival gap is the original 1976 All-Canadian Venture Fund letter. Its checklist is widely repeated, and the secondary attribution is plausible, but the original should be recovered before future work treats it as fully verified. The second gap is Routines and Orgies Appendix B, "Notes from a Battle-Scarred Warrior," which visible JSTOR and MQUP metadata establish as central but do not expose in clean full text. The third gap is the 1998 Outstanding Investor Digest Cundill material: public search found the issue, but this run did not recover a clean readable copy, so it should remain a lead rather than a cited source. The fourth gap is the original Globe/CTV final interview page; the ValueBuddies repost is useful but encoding-damaged and derivative.

Legal and regulatory context should be kept narrow. The public-source legal review found a 1991 Canadian tax appeal involving fees paid to a Bermuda company, where the appeal from withholding-tax assessments was dismissed, but this was a tax/arm's-length case rather than securities misconduct Tax Interpretations. The Bond Buyer reported a 1995 suit brought by Cundill as plaintiff over old defaulted Mississippi bonds, which is evidence of distressed-credit style, not a scandal Bond Buyer. The 2006 Form ADV for Cundill Investment Research Ltd. showed "No Information Filed" on criminal, regulatory-action, civil-judicial, bond, judgment/lien, and arbitration disclosure pages SEC/IAPD Form ADV. Bounded searches as of 2026-07-18 did not surface a primary SEC/OSC/CSA enforcement action against Peter Cundill personally, but that is negative public-source evidence, not a legal clearance.

The most important attribution rule: never use current Mackenzie Cundill fund materials, post-2009 Ivy successor reports, or later Mackenzie product performance as if Peter Cundill personally authored or generated it. Those materials can describe lineage or continuation, but not Cundill's own writings.

As of 2026-07-18. Peter Cundill died in 2011; these are historical models reconstructed from his record, not the current process of the successor Mackenzie fund (McGill Reporter, 2011; Mackenzie successor profile, 2026).

Evidence Labels And Guiding Questions

A model is direct when Cundill explained the rule in the official 2005 Ivey recording or a Peter-only interview. Adapted means he used a Graham, Templeton, or other inherited idea rather than originating it. Reconstructed means the Canon turns several documented practices into an operational test he did not name. Team/product identifies one Ivy vehicle or named co-manager group. Successor rules are excluded from Peter's model set unless earlier evidence independently supports them.

1. Which models are genuinely Cundill's, and which are borrowed, team-based, or later Mackenzie codifications?
2. How did an accounting screen become a buy, position, review, and sale?
3. Which numerical thresholds were aspirations or point-in-time observations rather than universal rules?
4. What distinguished patience from stubbornness and averaging down from escalation?
5. How did control, law, cash burn, and time change apparent asset value?
6. Which failures actually altered the process?
7. What can an individual reproduce without Cundill's team, access, vehicles, board, and patient clients?

Direct, Adapted, And Reconstructed Models

1. Net-net as a survival test — adapted from Graham, operationalized by Cundill

Cundill began with Graham's net-current-asset logic: conservatively mark current assets, subtract every liability, ignore goodwill, and compare the residue with market value. The controlled book record also confirms that Appendix 3 is a net-net worksheet designed by Cundill in 1975, although its full fields were not recovered through clean public access (JSTOR controlled record, 2011; Ivey lecture, 2005, 12:30–12:51).

Operational rule: reconstruct rather than copy the balance sheet. Deduct restricted cash, doubtful receivables, obsolete inventory, debt, leases, pensions, tax claims, restructuring costs, and expected cash burn. A loss-making division can be worth less than zero.

Falsifier: the model fails if the company cannot survive to realization or if liabilities, losses, or control prevent the assets from reaching shareholders. Cable & Wireless is the decisive counterexample: reported cash did not protect the investment from a cash-consuming network and large obligations (C&W Form 20-F).

2. “Magic Sixes” as a scarcity gauge — direct use of a borrowed screen

In 2005 Cundill described a screen combining roughly a 6% dividend yield, six times earnings, and 60% of book value. He reported only about 55 matches among approximately 55,000 global companies [single-source lecture figures] (Ivey lecture, 2005, 3:06–4:02). The historical origin of the screen was not established from a primary source, so it should not be called a Cundill invention.

Operational rule: use the screen to measure how plentiful classic bargains are and to create a research list. Never buy from the ratios alone. Earnings can be cyclical or false, book value inaccessible, the dividend unsustainable, and debt fatal.

Falsifier: track screened candidates rejected for insolvency, governance, or structural decline. If the screen cannot separate survivors from traps after full underwriting, it is a search convenience rather than an edge.

3. A valuation ladder, not one master formula — direct

Cundill described three lenses: adjusted net asset value, sum of the parts, and future net asset value or earning power. He treated sum-of-the-parts work as especially important and future value as the least dependable. He objected to false DCF precision, not to all forecasting or cash-flow analysis (Ivey lecture, 2005, 27:40–28:30; 52:20–53:10).

Operational rule: begin with observable assets and claims, value separable businesses, deduct corporate leakage and all senior claims, then use earning power only where durability can be defended. Archive sensitivities instead of hiding them in a single precise output.

The Peter-led 2002 U.S. Ivy process often sought a discount of at least 50% to estimated NAV and expected a three-to-five-year-or-longer hold (SEC Ivy semiannual report, 2002). Those are product/team aspirations, not universal career rules. The co-managed portfolio was later described at 79% of estimated intrinsic value in 2006 and 76% in 2007, demonstrating that actual opportunity sets did not always meet the memorable 50-cent ideal (SEC Ivy annual report, 2006; 2007).

4. Asset access plus a catalyst clock — direct components, reconstructed gate

Accounting value is useful only if minority owners can reach it. Cundill's team examined management, competitors, control, trading and liquidity events, and possible catalysts. His lecture emphasized that hostile control or weak incentives can leave a security cheap indefinitely (SEC Ivy semiannual report, 2002; Ivey lecture, 2005, 1:11:39).

Operational rule [Canon reconstruction]: identify who controls the assets, the legal rights of each claim, and one or more plausible realization paths—sale, liquidation, restructuring, buyback, dividend, refinancing, normalized earnings, or governance change. Estimate both probability and time; no public source provides a universal Cundill catalyst hurdle.

Falsifier: recalculate the annualized prospective return after delay, cash burn, and legal cost. LTV shows that a nominally protected recovery can still be poor when pensions, labor claims, court process, and seven years consume the time value (U.S. Supreme Court LTV opinion; Canon loss audit).

5. Search where aversion is concentrated, then go and see — direct habit

Cundill said that each November he investigated the year's worst-performing national market. Screens generated names; travel, company visits, local accounting, competitors, managers, and culture tested whether the discount was real. Sweden and Japan illustrate the method (Ivey lecture, 2005, 5:00; authorized preview, 2011).

Operational rule: use extreme aversion as an idea funnel, never as a buy signal. Ask what local law, capital structure, owner behavior, or financing condition a remote screen misses. Pan-Ocean shows the benefit of delaying purchase until fieldwork improved Cundill's assessment of management and political/operational risk (authorized-biography online text, paraphrase only; Lexpert transaction record, 2006). Japan shows that many cheap names can share one governance and deflation trap (Forbes profile, 2004).

6. Re-underwrite from today's facts — direct principle, reconstructed procedure

Cundill's C&W postmortem framed each day as a new decision rather than validation of the original purchase. The operational implication is prospective: cost basis, prior work, and a lower price do not establish a wider margin of safety (Ivey lecture, 2005, 1:08:34).

Operational rule [Canon reconstruction]: before every addition, rewrite adjusted value, cash runway, debt/refinancing, law, control, catalyst, and the strongest contrary case. Name the new evidence that justifies a larger position. If value fell faster than price, averaging down is escalation.

The joint Cundill/Hiok Hhu Ng 2007 discussion is the hardest falsifier: it said political risk impaired original Japanese consumer-finance estimates and also said the fund kept building the positions. Aiful, Takefuji, and Acom were a derived 14.22% of that U.S. product at period-end, while Japan was 48.28% and financial services 36.23% [single-product figures] (SEC Ivy annual report, 2007). Separate reporting particularly associates Takefuji initiation and additions with Ng, so the episode cannot be assigned to Peter alone (Washington Post, 2005).

7. Cash is the residual of standards — direct and team/product

Cundill said cash accumulated because qualifying ideas were scarce, not because he forecast an index decline. The 2003 prospectus likewise described bottom-up selection, no reliance on market timing, and unlimited defensive cash; the 2006 Cundill/Ng discussion said the fund would not invest merely to become fully invested (Ivey lecture, 2005, 51:17; SEC prospectus, 2003; SEC annual report, 2006).

Operational rule: define the valuation and quality conditions that permit deployment. Cash without a written opportunity-cost and deployment rule is undocumented market timing. Cash can protect optionality but can also drag for years and provoke client pressure.

8. Concentrate attention; diversify failure mechanisms — direct observations, reconstructed portfolio test

After an over-diversified early-1990s portfolio, Cundill described a 2005 portfolio with roughly 75% in about 25 holdings, counting a large cash balance as one [single-source point-in-time figures]. C&W later supported diversification because even intensive work can miss a negative-value liability (Ivey lecture, 2005, 46:13; 1:09:39). No universal position limit or sizing equation was found.

Operational rule [Canon reconstruction]: size for downside, leverage, liquidity, cash burn, legal complexity, control, and catalyst uncertainty, then aggregate country, industry, regulation, currency, and funding exposure. Twenty-five securities can still be one Japan/financial/regulatory bet.

Full currency hedging was a house/product overlay, not a timeless personal law. It hurt the 2002 U.S. product when foreign currencies rose, while later language described the portfolio as predominantly rather than universally hedged (SEC Ivy annual report, 2002; SEC Ivy annual report, 2007).

9. Patience needs a clock and a kill test — reconstructed from direct practice

Patience meant surviving early losses and career pressure while repeatedly testing the facts. It did not mean that time was free. Westar and C&W pull in opposite directions: client pressure helped force a premature Westar sale, while attachment and continued additions worsened C&W. The authorized biography is the main source family for Westar; exact weight and loss remain unavailable (McGill-Queen's University Press book provenance, 2011; Canon loss audit).

Operational rule [Canon reconstruction]: precommit separate conditions for price-only adversity and thesis impairment. Review solvency, value decay, catalyst probability, annualized return, and capital-provider runway. A patient thesis without a kill test is unfalsifiable; a sound thesis without patient capital is unownable.

10. Sell half after a double — direct behavioral rule

Cundill described selling half after a position doubled and requiring a governance review before retaining the rest. A former colleague independently remembered it as a process that reduced both attachment and the relief-driven urge to sell everything (Ivey lecture, 2005, 33:40; Tim McElvaine, 2021).

This is a regret-control and review rule, not valuation. The remaining position is not economically free: its current value is still at risk. Tiffany shows why the rule emerged, but its later offer is disputed—about $50 in the authorized-book family versus $45 in contemporaneous reporting (TIME, 1978).

Falsifier: record remaining discount and expected return at the sale date. Repeatedly trimming the best prospective holdings merely because cost doubled would invalidate the rule as a return-maximizing discipline.

11. Institutionalize dissent after failure — Cundill reflection via an edited source

The public Ivey captions support C&W's negative-value and diversification lessons but do not contain the often-assigned devil's-advocate remedy. A controlled book page preserves Cundill's reflection that the team lacked a natural skeptic and proposed appointing one (Google Books, 2011, p.158). This is editor-selected book material, not a caption-exact lecture fragment or dated original memo.

Operational rule [Canon adaptation]: require an independent downside memo before a large addition, and make the advocate state what evidence would veto the trade. A ceremonial skeptic who cannot change size or demand new evidence does not solve groupthink.

12. Preserve the framework; change the tactics — direct lesson, source-conflicted maxim

In the 2005 recording, Cundill said Templeton had advised him to “always change a winning game” and acknowledged failing to adapt during the 1990s. A later edited reproduction attributes the dictum to Peter Robertson. The operative lesson is direct even though the maxim's origin is source-conflicted: retain discipline and the margin of safety while changing search fields, instruments, and tactics as the opportunity set changes (Ivey lecture, 2005, 6:30; BusinessDay, 2015).

Falsifier: specify which observable breakpoint justifies the tactical change and which core rule remains fixed. Otherwise flexibility becomes style drift and every mistake can be excused after the fact.

13. Turn observation and error into a cumulative case library — direct and editor-mediated

Cundill first observed that below-cash securities worked, then used Graham to explain the pattern. His private journals, selected by Christopher Risso-Gill, recorded cases and errors across decades; the raw journal run is not public. Broad reading, travel, history, and curiosity supplied comparisons that a screen could not (authorized preview, 2011; McGill Reporter interview, 2008).

Operational rule [Canon reconstruction]: retain the original thesis, dated facts, additions, disconfirming evidence, exit decision, and postmortem. Classify errors as asset, liability, time, governance, regulation, sizing, factor, behavior, or omission. Do not reformulate the question until the preferred answer appears.

This learning model does not make every saying his own. The authorized book credits its title advice to Irving Kahn, and literary or investing aphorisms repeated in derivative summaries should not be laundered into Cundill-authored doctrine.

Reconstructed Decision Checklist

A. Search and mandate

1. What capital horizon, liquidity, currency, leverage, and client constraints govern the position?
2. Is the candidate a net-net, hidden-asset/SOTP, distressed claim, control event, or earning-power bargain?
3. Why is it neglected, and what would make that neglect rational rather than temporary?
4. What does the global screen omit about local law, accounting, funding, and minority-owner rights?

B. Rebuild value and survival

1. Mark every asset to conservative realizable value; identify location, restriction, tax, and control.
2. Deduct debt, leases, pensions, litigation, restructuring, dilution, and contingent claims.
3. Value loss-making operations below zero when their obligations exceed closure or sale value.
4. Stress cash burn and refinancing through the expected realization period.
5. Triangulate adjusted NAV, SOTP, and defensible earning power; expose sensitivities.

C. People, access, and catalyst

1. Test management and controlling-owner incentives against filings, competitors, customers, and site evidence.
2. Can minority holders legally and practically receive the appraised value?
3. Name the realization path, its probability, and its clock.
4. Write the strongest contrary case and assign an independent advocate for high-conviction additions.

D. Entry, size, and portfolio

1. State the margin of safety and which assumptions consume it.
2. Start small when evidence is incomplete; every add must cite new dated evidence and a fresh valuation.
3. Size for permanent-loss severity, liquidity, leverage, law, control, and catalyst uncertainty—not conviction rhetoric.
4. Aggregate country, industry, regulation, currency, funding, and liquidity factors across names.
5. Hold cash only under a written scarcity/deployment rule.

E. Monitor, wait, and sell

1. Separate a lower price from lower value. Which changed?
2. Recompute annualized prospective return after time, cash burn, and legal cost.
3. Sell when value fills, the thesis breaks, the catalyst decays, survival weakens, or a superior risk-adjusted opportunity appears.
4. Treat selling half after a double as a mandatory review, not proof that the remainder is free.
5. Do not let client discomfort alone force sale, but ensure the capital can actually survive the wait.

F. Learn

1. Preserve the original question and dated evidence.
2. Separate Peter, team, product, issuer, co-manager, and successor attribution.
3. Audit misses and omissions using information available at the time.
4. Change one relevant check after a failure; do not retrofit a complete new philosophy.

Model-To-Case Audit

Failure Modes And Falsifiers

• Static-book-value trap: log restricted cash, negative operations, contingent claims, and value decay. Repeated loss surprises mean the adjusted balance sheet is incomplete.
• Confirmation-driven adding: every addition must cite new evidence, not lower price. Adds after documented thesis impairment falsify fresh-page discipline.
• Catalyst theater: state whether the catalyst is required or optional. Repeated deadline extensions without higher value or probability make patience unfalsifiable.
• Time blindness: calculate expected annualized recovery after legal delay and opportunity cost. Principal preservation can still be failure.
• Governance hindsight: define minority-rights and controller-behavior tests before purchase. A governance concern discovered only after loss has no predictive content.
• Shared-factor diversification: aggregate country, sector, law, currency, funding, and liquidity. Many tickers that fail together are one exposure.
• Mechanical selling: compare retained expected return with the half sold and its replacement. Cost-based trimming that repeatedly sacrifices the best remaining asymmetry is a bad rule.
• Ceremonial dissent: archive the advocate's downside case, requested evidence, and sizing effect. A critic with no impact is theatre.
• Flexibility as drift: record the breakpoint, tactical change, and invariant principle. If every regime excuses a new style, the framework has disappeared.
• Attribution leakage: tag each statement and outcome as Peter-only, team/product, co-manager, issuer, or successor. Fund holdings do not prove sole origination.

Skill, Luck, And Team Attribution

The strongest evidence supports durable skill in global candidate generation, balance-sheet reconstruction, field research, uncomfortable holding periods, special-situation structure, and learning after error. It does not establish a lone-manager or factor-independent record. The U.S. Ivy documents describe a Peter-led team in 2002, joint Cundill/Ng work in 2005–07, and four managers in 2008. Fairfax's credit-default-swap gains belonged to Fairfax, and post-transition results belong to successors (SEC Ivy prospectus, 2007; Fairfax Q1 2008 release).

Luck and regime mattered. The 1970s offered abundant classic bargains; later control transactions and restructurings required counterparties, courts, financing, and timing outside Cundill's control. No complete Canadian trade ledger, monthly drawdown series, or public factor attribution separates value, size, country, currency, team, platform, timing, and personal security selection.

Transferability To An Individual Investor

What can be replicated

An individual can reproduce the net-net and SOTP logic, read filings first, mark assets and liabilities conservatively, write a cash-burn runway, define minority access and catalysts, start small, require new evidence for additions, aggregate common factors, hold cash when standards fail, maintain a thesis ledger, and conduct a devil's-advocate review. Smaller capital can exploit securities too illiquid for an institutional fund and can avoid benchmark pressure.

The sensible adaptation is a narrower, documented portfolio—not a miniature globe-spanning Cundill organization. An individual can replace some travel with local filings, calls, industry records, and explicit uncertainty discounts, while declining situations that cannot be verified.

What cannot be cleanly replicated

Cundill's organization supplied analysts, global databases, management and competitor access, site visits, legal and accounting support, cross-border trading, currencies, private claim structures, a board, and clients able—sometimes—to wait. An individual cannot responsibly follow hundreds of securities or assume identical access, liquidity, taxation, or governance influence.

Nor should an individual copy a nominal 50% discount, 25-name snapshot, full-currency hedge, sell-half rule, or successor scorecard as a universal formula. Current Mackenzie quality ranks, milestones, and catalyst monitoring are successor codifications and must not be projected backward onto Peter (Mackenzie successor-team profile, 2020).

Evidence Boundaries And Open Questions

• The Ivey recording is the sole verified full Peter-only recording. Its automatic captions are navigational; the document paraphrases except for tiny named phrases.
• The 1975 net-net worksheet exists in controlled book metadata, but its full public fields were not recovered. This checklist is transparently reconstructed from visible primary evidence, not a transcription of that copyrighted appendix.
• Risso-Gill's 2011 and 2014 books are authorized, edited works about Cundill. Journal composition, editorial selection, and publication are separate stages; the raw journal run is private.
• The 1976 investor letter and “Notes from a Battle-Scarred Warrior” are verified archival leads whose clean full text remains unrecovered (McGill-Queen's University Press, 2014).
• The C&W devil's-advocate remedy comes from an editor-selected book page, not the official lecture captions. The tactical-flexibility maxim has conflicting origin attributions across direct and edited sources.
• Japanese consumer finance belongs to a named co-managed U.S. product; 2008 is transition-team evidence; current Mackenzie outcomes are successor evidence.
• The 1991 tax finding concerned corporate control and non-arm's-length dealing, not securities fraud; the 2006 OSC item records surrender-related registration suspension, not an adverse enforcement sanction (tax decision; OSC Bulletin). Bounded public searches are not a universal clean-record finding.
• No verified universal sizing formula, stop loss, catalyst deadline, country cap, hedge ratio, or averaging-down schedule was found.

As of 2026-07-18. Peter Cundill died on January 24, 2011. Current Mackenzie holdings, managers, process refinements, fees, and returns belong to a successor franchise and are not evidence of his present activity or personal record. (McGill Reporter) (current Mackenzie fund profile)

Executive Brief

Peter Cundill's durable achievement was not inventing deep value. It was turning Benjamin Graham's balance-sheet discipline into a global search process for public equities, distressed claims, and special situations. The chronology matters: the predecessor fund began in 1967; Cundill and Vanan assumed control in 1975; the vehicle became the Cundill Value Fund in April 1977; and Peter Cundill & Associates was formed in 1977. (Mackenzie historical chart) (Peter Cundill Foundation)

His edge was a sequence, not a screen. Statistical filters identified net-nets, low-book-value stocks, distressed securities, and markets under concentrated aversion. The team then reconstructed the balance sheet, marked assets conservatively, deducted senior and contingent claims, estimated cash burn, and compared adjusted net asset value with sum-of-the-parts and defensible earning power. Company visits, competitors, local law, control, incentives, and a plausible path by which value could reach shareholders decided whether an accounting bargain was investable. The Peter-led 2002 U.S. Ivy process often sought at least a 50% discount to estimated net asset value and contemplated three-to-five-year or longer holding periods; those were product-specific aspirations, not universal career laws. (Ivy semiannual report, 2002)

The sponsor-reported record is exceptional but not audit-clean for Canon purposes. Prem Watsa's foreword to the authorized investment book reports 15.2% annual compounding over 33 years through 2007 and growth of C$10,000 to more than C$1 million; the affiliated Foundation reports 13.2% over 35 years. Mackenzie's chart shows a 25.1% loss in 2008 and 13.0% annualized from December 1974 through December 2014, but the last six years include successors and the chart's historical “Series A” label is not reconciled to the current legal series' October 1998 inception. Original audited predecessor statements and a series bridge remain missing. (authorized 2011 preview) (Mackenzie historical chart) (current Mackenzie MRFP)

Cleveland-Cliffs, Pan-Ocean, Sibir, Fairfax, Ecuadorian debt, and earlier hidden-asset holdings are consistent with skill in asset appraisal, field research, cross-asset flexibility, and patient realization. Yet most exact security-level economics remain within one authorized-biography and derivative source cluster. Results embed teams and vehicles: 2005-2007 U.S. commentary is attributable jointly to Cundill and Hiok Hhu Ng, while 2008 names four managers. (Ivy annual report, 2007) (Ivy annual report, 2008)

The failures make the method more transferable. Cable & Wireless showed that gross cash is not distributable value when operations and obligations have deeply negative value. Japanese consumer finance showed that a changed legal regime can destroy book value across apparently diversified names; this was a Cundill/Ng product-level error, with Takefuji research particularly associated with Ng. Westar and LTV illustrate opposite patience failures: capitulating under external pressure versus allowing time and legal complexity to consume the apparent margin of safety. (Cable & Wireless Form 20-F) (Japan FSA reform summary) (LTV pension opinion)

The transferable core is demanding but compact: search where aversion is concentrated; reconstruct value rather than copy accounts; test survival and access to assets; attach a clock and kill test to patience; require fresh evidence before adding; aggregate shared failure factors; and let current prospective return govern sales. What cannot be copied cleanly is Cundill's global team, access, special-situation vehicles, governance structure, or unusually patient capital.

10 Transferable Lessons, Ranked

1. Rebuild the balance sheet as a survival test. Start with cash and realizable assets, then deduct restricted cash, doubtful receivables, debt, leases, pensions, tax claims, restructuring costs, and expected cash burn. C&W is the decisive falsifier: a cash-rich appearance concealed a value-consuming operating network and large obligations. Cheapness is useful only if the equity can survive to realization. (Cable & Wireless Form 20-F)

2. Require both asset access and a catalyst clock. Identify who controls the value, what minority owners can legally receive, the mechanism of realization, its probability, and its expected time. LTV shows that an asset appraisal can be directionally right while bankruptcy duration, pension priority, and opportunity cost make the investment poor. Cundill disclosed no universal catalyst deadline, so the clock must fit the claim and vehicle. (LTV pension opinion)

3. Re-underwrite from today's facts before adding. A lower price does not increase the margin of safety when intrinsic value has fallen faster. Recompute value, runway, liabilities, legal regime, control, catalyst, and contrary evidence before every addition. The Cundill/Ng product's continued Japanese consumer-finance exposure after acknowledging regulatory impairment shows that the organization did not always meet this ideal. (Ivy annual report, 2007) (Bank of Japan Financial System Report, 2008)

4. Use concentrated aversion as an idea funnel, not a buy signal. New lows, neglected industries, and badly performing countries create a search field. Travel, company work, local institutions, people, and accounting determine whether pessimism is excessive. This separates global contrarian research from automatically buying whatever has fallen most. (Ivey lecture, 2005)

5. Triangulate valuation and reject false precision. Move from adjusted asset value to sum-of-the-parts, then use future asset value or earning power only where assumptions are defensible. Cundill criticized precise discounted-cash-flow outputs built on unstable inputs; he did not reject modeling. The 50%-of-NAV aspiration was a 2002 U.S. product snapshot, not a universal threshold. (Ivey lecture, valuation ladder) (Ivey lecture, DCF discussion)

6. Preserve the framework while changing tactics. Margin of safety, evidence, and skepticism can remain stable while markets, instruments, and implementation change. Net-nets, asset breakups, distressed debt, country bargains, index protection, and later earnings-aware work were different expressions of the same downside-first discipline. The “change a winning game” maxim has conflicting attribution and should not be presented as Cundill's invention. (Ivey lecture, 2005)

7. Patience requires a clock, kill tests, and compatible capital. Precommit separate responses to price-only adversity and thesis impairment; after delay, recalculate cash burn, catalyst probability, and annualized prospective return. Westar and C&W show why “be patient” alone cannot distinguish premature sale from stubborn escalation. Client behavior, redemptions, and vehicle terms are part of the edge, not administrative details. (mental-model audit)

8. Concentrate research while diversifying failure mechanisms. Aggregate country, sector, legal-regime, funding, currency, liquidity, and counterparty exposures rather than counting tickers. The lecture's roughly 75% in about 25 holdings was a point-in-time snapshot, not a permanent rule, and multiple Japanese lenders were still one legal-regime bet. (Ivey lecture, portfolio snapshot) (Japan FSA ministerial briefing, 2011)

9. Institutionalize dissent before large additions. Require an independent downside memorandum that can change size or stop the trade. The C&W devil's-advocate remedy is preserved on controlled Google Books page 158; it is not present in the official Ivey caption streams. It is an editor-mediated reflection rather than proof that the original process contained this control. (Google Books, p. 158)

10. Govern exits prospectively, not from cost basis. Sell when value is realized, the thesis breaks, or another opportunity offers a better risk-adjusted return. Treat “sell half after a double” as a behavioral review mechanism, not proof that the remaining shares are free or that every winner should be cut mechanically. (Ivey lecture, sale discipline)

Style Taxonomy Tags

• Graham deep value and margin of safety
• Global contrarian and worst-market search
• Net-net, adjusted-NAV, and sum-of-the-parts investing
• Distressed debt, restructurings, and special situations
• Forensic balance-sheet and liability reconstruction
• Field research, travel, and local-law verification
• Asset-access and catalyst-aware patience
• Concentrated research with factor-level diversification
• Behavioral sale rules and fresh-page re-underwriting
• Bottom-up public mutual-fund and team-based process
• Flexible cash, currency hedging, and occasional index protection
• Vehicle-, source-, series-, and manager-attribution-constrained record

Cundill should not be tagged as the inventor of net-net or “Magic Sixes,” a formulaic liquidation investor, a pure earnings-value manager, a forecast-led macro trader, or a current Mackenzie portfolio manager. He used inherited screens as beginnings; judgment about liabilities, people, law, asset access, and time decided whether the apparent value was real.

Regime Dependence

Closest And Most-Opposite Investors Already In The Repo

Closest overall: John Templeton. Both searched worldwide where pessimism was concentrated, bought before consensus improved, tolerated uncomfortable holding periods, and used diversified public-fund vehicles. Templeton more often expressed country and industry dislocation through probabilistic baskets of fundamentally sound bargains. Cundill placed more weight on net-nets, reconstructed asset value, company visits, and a security-specific realization route.

Closest operating-process analogue: Anthony Bolton. Both treated unpopularity as a funnel, investigated recoveries and hidden assets, used institutional research teams, staged conviction, and asked what would change perception. Bolton admitted a wider class of unrecognized growth, used a written counter-thesis and charts as warnings, and worked mainly in the UK and Europe before China. Cundill was more consistently global, asset-based, distressed, and explicit about the access-to-assets problem. Benjamin Graham is the primary intellectual ancestor, but his documented process is less centered on worst-country search, extensive fieldwork, and a global public-fund platform.

Most opposite: William J. O'Neil. O'Neil favors accelerating earnings, relative-strength leaders, price-volume breakouts, broad-market confirmation, pyramiding after price advances, and rapid loss limits. Cundill begins with a discount to conservatively appraised value, often enters before price confirmation, can add after declines only if value survives fresh underwriting, and allows years for realization. Both are disciplined, empirical, and rule-aware; their preferred evidence, entry timing, and tolerance for adverse price action are structural opposites.

Skill, Luck, Team, And Attribution

The skill case rests on duration, a coherent process, and adaptation across countries and instruments. Cundill combined screens with reconstruction, fieldwork, catalysts, and postmortems rather than relying on one historical factor. The authorized 15.2% figure through 2007 covers 33 years, not one favorable cycle, and the trade library spans equities, distressed debt, liquidations, hidden assets, and hedges. The process also changed after errors: C&W sharpened negative-value and dissent tests, while later materials distinguish adjusted NAV, SOTP, and more speculative future value. (authorized 2011 preview) (Ivey lecture, 2005)

The causal limits are substantial. Country recovery, commodity cycles, takeovers, courts, and counterparties contributed to outcomes. Most security-level returns come from one authorized-book family; the public record does not provide a complete holdings, cash-flow, tax-lot, drawdown, or factor series. Analysts and co-managers supplied ideas and diligence. Different funds had different mandates, currencies, hedging, fees, and holdings, and the 2008 loss sits in a four-manager transition. Neither sponsor longevity nor a few reconstructed winners isolates Peter-only, factor-independent alpha.

Legal records require equally careful boundaries. A Federal Court of Appeal decision made an adverse corporate non-arm's-length and control finding in a tax dispute; it was not a securities-fraud finding. The Ontario Securities Commission consented to suspension following surrender of registration during the Mackenzie transition; the bulletin does not establish an enforcement sanction. A Delaware fiduciary allegation involving Cundill ended in judgment for the defendants. These records neither justify a blanket “clean record” nor a misconduct narrative. (tax decision) (OSC bulletin) (Delaware Chancery decision)

What An Individual Can Copy

An individual can run a conservative net-current-asset and low-price-to-book screen; reconstruct cash, liabilities, and burn from original filings; use SOTP only with explicit haircuts; identify control, asset access, catalyst, and time; read local law; seek disconfirming evidence; aggregate shared factors; maintain a fresh-price review; and write postmortems. Smaller capital can exploit neglected securities that cannot matter to a multi-billion strategy.

An individual should not imitate Cundill's breadth, information access, cross-border legal workload, distressed instruments, derivative overlays, or long holding periods without matching competence and liquidity. The safer adaptation is fewer positions, no borrowed leverage, more severe governance and jurisdiction haircuts, an explicit cash-runway test, and a precommitted point at which delay turns the apparent bargain into an inadequate annualized return.

Unresolved Questions

1. Recover original audited predecessor statements and a formal redesignation or series bridge connecting the 1967 vehicle, December 1974 chart start, 1975 control change, April 1977 rename, current Series A, and every manager transition.
2. Reconstruct monthly NAV, flows, fees, cash, currency, hedges, holdings, turnover, benchmark, and drawdowns through Cundill's last decision date; then separate Peter, analysts, co-managers, and successor contributions.
3. Obtain a complete transaction and position ledger for Cleveland-Cliffs, Pan-Ocean, Sibir, Fairfax, Ecuador, TBS, C&W, Westar, LTV, Japan consumer finance, and the other named cases; replace book-derived arithmetic with fund P&L and opportunity cost.
4. Recover the raw journals and version history behind Risso-Gill's edited books, preserving composition date, editorial selection, and later publication date; locate the clean full texts of the 1976 letter and “Notes from a Battle-Scarred Warrior.”
5. Obtain the original 1975 net-net worksheet and document its actual fields without reconstructing inaccessible content; establish the provenance and inventor, if any, of “Magic Sixes.”
6. Reconstruct C&W contemporaneous downside memoranda, cash-location analysis, additions, dissent, and exit authority; date the later devil's-advocate remedy and test whether it became an enforceable control.
7. Separate Peter's, Ng's, other managers', analysts', and product committees' decisions in Japanese consumer finance; reconstruct security-level exposure, additions, refunds, exits, and currency effects across every mandate.
8. Determine the currencies, double-counting boundaries, client mandates, and actual capacity constraints behind the 2006 transaction's more-than-12.5-billion Mackenzie and more-than-3-billion external figures.
9. Test the full portfolio for hidden factor concentrations, value traps, premature exits, and selection bias rather than inferring a hit rate from the surviving case library.
10. Compare the Cundill process out of sample with Graham, Templeton, Schloss, Bolton, and later successors using common definitions for value, catalyst, concentration, patient capital, and realized return.

Annotated source map started during T0470 A-profile research. Ranked roughly by usefulness for future tasks.

Task A - Profile (T0470)

As of 2026-07-18. This map supports the profile, vehicle chronology, track-record caveats, legal-status check, and later B-H research.

Guiding questions

1. What is the cleanest biography and current deceased/legal-status evidence for Francis Peter Cundill?
2. When did the Cundill Value Fund actually begin: predecessor fund, Cundill control, rename, or current Mackenzie legal series?
3. Which vehicles belonged to Cundill personally, to Peter Cundill & Associates / Cundill Investment Research, to Mackenzie, or to U.S. subadvised products?
4. Can the 13.2% over 35 years and 15.2% over 33 years return claims be reconciled?
5. What AUM figure is best supported at the 2006 Mackenzie transaction, and how does it relate to the nearly $20 billion affiliated claim?
6. What documented mistakes, criticism, legal matters, or successor-product issues should later files not miss?
7. Which sources are primary enough for later philosophy, trades, quotes, and writings tasks?

Annotated source map

1. Peter Cundill Foundation official biography - Best official memorial biography for education, 1975 acquisition/control shorthand, Graham style, 13.2%/35-year return claim, $7 million to just under $20 billion AUM claim, illness, philanthropy, and books. Treat as affiliated/commemorative.
2. McGill Reporter obituary / Senate statement, 2011 - Strong memorial source for death in London on January 24, 2011, McGill background, career summary, relationships, History Prize, illness, and book timing.
3. McGill Alumni memorial-service page, 2011 - Institutional confirmation of death date/location, chairman emeritus status, McGill BCom, career-achievement award, and 2008 history-prize gift.
4. Companies House, Francis Peter Cundill appointments - Official UK record for full name, October 1938 birth, Canadian nationality, UK residence, London address, and directorship chronology.
5. Ivey Ben Graham Centre former advisory-board biography - Best concise chronology for Greenshields/Yorkshire/AGF, the 1967 predecessor, 1975 Vanan takeover, April 1977 rename, 1977 firm formation, Mackenzie relationship, and awards.
6. McGill Reporter profile, 2008 - Contemporaneous institutional source resolving London residence from 1984 and documenting the new History Prize and McGill fellowships.
7. Mackenzie acquisition release, 2006 - Official announcement for the separate Cundill Investment Research asset purchase, eight-year partnership, more than $12.5 billion of Mackenzie mandates, more than $3 billion external AUM, team continuity, and Cundill's CIO role.
8. Power Corporation annual information form, 2009 - Primary SEDAR+ source dating the 2006 closing to September 22 and identifying $3.3 billion of incremental acquired AUM.
9. SEC Ivy Funds SAI, 2006 - Primary adviser and attribution source for Cundill Investment Research, the Peter Cundill & Associates former name, Ivy fee breakpoints, Peter's five registered vehicles/$5.5 billion, and the Mackenzie transaction.
10. SEC/IAPD Form ADV for Cundill Investment Research Ltd., 2006 - Primary structure source for 75%-or-more trust ownership, Cundill's chairman/director roles, exact private-fund assets/minimums, Bermuda manager, and absence of adviser DRP pages.
11. SEC Ivy Funds semiannual report, June 30, 2002 - Best primary process source: NAV discounts, sum-of-parts work, visits, catalysts, long holding periods, and Cable & Wireless as a real-time value-trap warning.
12. SEC Ivy Funds annual report, 2002 - Primary Ivy product evidence: inception dates, -12.17% in 2002 versus weaker comparators, holdings, and audited product reporting.
13. McGill-Queen's University Press page for There's Always Something to Do, 2011 - Authoritative book metadata, contents, author biography, and provenance statement that the authorized book drew on interviews, speeches, and Cundill's 45-year journals.
14. Authorized preview of There's Always Something to Do, 2011 - Prem Watsa foreword for 15.2%/33 years, C$10,000 to more than C$1 million, nearly $20 billion scale, plus early biography and method; affiliated source family.
15. JSTOR record for Routines and Orgies, 2014 - Controlled biography record and access lead for journal-based personal/professional material beyond the investment-approach book.
16. SEC Ivy Funds annual report, 2007 - Primary adverse-path evidence for the co-managed U.S. product's Japanese consumer-finance error, regulatory-risk miss, concentration, and continued averaging down.
17. Peter Cundill Foundation home page - Official source for the Bermuda foundation's January 11, 2012 establishment after Cundill's death and its youth-focused mission.
18. Mackenzie/Morningstar historical chart, 2015 - Official sponsor chart with every 1975-2014 annual return, -25.1% in 2008, 13.0% since December 1974 through 2014, and the historical Series A label.
19. Mackenzie Cundill Value Fund Series A profile, May 2026 - Current successor source for Richard Wong, October 1998 Series A inception, benchmark, and roughly $1.1 billion total fund assets.
20. Mackenzie Cundill Value Fund annual MRFP, March 31, 2026 - Current legal-series source showing fund formation in 1998, current Series A in 1998, closed Series AG in 1967, fees, and successor operations.
21. Mackenzie mutual-fund prospectus, September 29, 2025 - Current mandate and legal/product context; useful for avoiding predecessor/current-series confusion.
22. Mackenzie fund-merger release, April 10, 2026 - Says Fund II was expected to merge into the continuing Cundill Value Fund around July 10, 2026; no primary completion notice was found by July 18.
23. SEC proxy, 2002 - Primary-adjacent corporate evidence that Mackenzie's chief executive had chaired Cundill Funds Inc. since September 1998, supporting the retail-manager/adviser distinction.
24. Peter Cundill & Associates Ltd. v. The Queen, 1991 - Adverse corporate tax appeal dismissed with costs; includes Cundill-specific control and non-arm's-length findings, but is not personal securities enforcement.
25. Bond Buyer, Mississippi 1841 bond lawsuit, 1995 - Cundill as plaintiff seeking payment on defaulted bonds; relevant to distressed-debt interests, not a controversy against him; final disposition was not recovered.

Evidence limitations

• Five research lanes covered biography/timeline, vehicles and track record, primary and near-primary writings, controversies/legal/successor-product issues, and independent source audit. The lanes completed broad web, SEC/ADV, Mackenzie, McGill, Foundation, book, and legal searches. Each lane returned read-only findings; no subagent edited repo files.
• The sponsor annual-return chart arithmetically reconciles Watsa's 15.2% through 2007 with the later 13.2%/35-year claim once the -25.1% 2008 return is included. Original audited predecessor statements, exact fee/tax conventions, and a series-redesignation bridge remain missing.
• Historical "Series A" and current Series A are not the same label: the 2015 chart begins in December 1974, while the 2026 MRFP dates current Series A to October 1998 and closed Series AG to January 1967. Current returns, AUM, managers, and mergers are successor-brand evidence.
• Mackenzie's release supports more than $15.5 billion of team-managed transaction scale; Power identifies $3.3 billion as incremental acquired external AUM. The release does not label currency. Nearly $20 billion remains an affiliated peak/rounded claim, not the transaction's added AUM.
• The Cundill Value Fund origin date should be split into four events: predecessor fund formed in 1967, Cundill/Vanan control in 1975, April 1977 rename, and Peter Cundill & Associates formation in 1977. Memorial sources often compress this into 1974/1975 shorthand.
• The 2006 ADV's missing DRP pages are adviser-level evidence, not a complete personal clearance. The broader Canada/Bermuda/UK/U.S. legal search is bounded.
• Cable & Wireless is the clearest Cundill-era error found in primary open sources, but exact Canadian-fund dollar loss and position size still need original Cundill Value Fund records or full biography page verification.
• The 2007 Japanese consumer-finance episode is a second primary error, but its figures and attribution belong to the co-managed U.S. Ivy product, not automatically to every Cundill mandate.
• Several colorful labels and quotes are provenance-sensitive: Buffett CIO remark, "Indiana Jones," "bravest value investor," and "sharp pencil/calculator/patience" should not be repeated as exact quotes unless sourced to an original venue or a verified book page.
• The Globe and Mail last interview, Mackenzie News obituary, Outstanding Investor Digest interviews, and full Ivey lecture/transcript were located as leads but not fully recovered in this run.

Task B - Investment Philosophy (T0471)

As of 2026-07-18. This map supports the doctrine, idea-to-sale process, regime analysis, and stated-versus-actual audit. Sources are ranked by their usefulness for Task B, not by fame.

Guiding questions

1. Which claims come directly from Cundill, and which belong to a co-managed vehicle, colleague, authorized biographer, or successor team?
2. How did a Graham net-net starting point evolve into a global and more eclectic value process?
3. What numerical discount, holding-period, concentration, and cash evidence is genuinely primary and vehicle-specific?
4. How did field research, control, governance, and catalysts determine whether accounting cheapness was realizable?
5. Where did the risk controls fail, and what did Cundill change as a result?
6. Which regimes rewarded or punished the method?

Annotated source map

1. Official Ivey guest-speaker index - Institutional index identifying Cundill and linking the canonical 2005 class recording. It establishes provenance but contains little doctrine itself.
2. Peter Cundill, Ivey Value Investing class, 2005 - Strongest direct source. A 74-minute first-person lecture covering origin, screens, travel, valuation, portfolio construction, cash, hedging, selling, flexibility, regime dependence, and Cable & Wireless. Automatic captions are navigational only; the write-up uses timestamped paraphrase.
3. Ivy Cundill Global Value Fund semiannual report, June 2002 - Best primary Peter-led team-process source: 50%-or-greater NAV discount aspiration, three-to-five-year-or-longer typical holding period, bottom-up construction, fieldwork, catalysts, and sum-of-parts/liquidation work. It captures Cable & Wireless as the worst performer, a continuing core holding, and an addition; it does not explain the liabilities later described by Cundill. Applies to one U.S. product.
4. Ivy Funds annual report, December 2002 - Direct manager interview and audited product evidence for full currency hedging, hedge drag, Japanese weakness, fair-value sales, relative-value switches, and buying during weakness.
5. Ivy Funds prospectus, 2003 - Contemporaneous regulatory description of intrinsic-value analysis, significant discounts, bottom-up selection, no market timing, potential catalysts, and unlimited defensive cash.
6. Ivy Funds prospectus, 2007 - Primary process and co-manager source naming Peter Cundill and Hhu Ng and adding governance, accounting, financial capacity, competition, and industry analysis. Team/vehicle evidence, not solely Peter's voice.
7. Ivy Funds annual report, 2007 - Strongest primary adverse case for the co-managed U.S. vehicle's Japanese consumer-finance concentration, regulatory-risk miss, and continued additions. Its exact weights and performance must not be generalized to the Canadian flagship.
8. Ivy Funds annual report, 2005 - Audited holdings and contemporaneous construction evidence: about 30 investments, substantial cash, and concentrated performance contribution. Contribution is not portfolio weight.
9. Ivy Funds annual report, 2006 - Team commentary and holdings showing a top-ten concentration around 40%, willingness to retain cash, and an actual portfolio discount less demanding than earlier ideals.
10. Ivy Funds annual report, 2008 - Later Peter-era/team evidence that a position such as Kirin could remain after an earlier target when fair value continued to compound; also records a more selective currency-hedging formulation.
11. Authorized preview of There's Always Something to Do, 2011 - Twenty-five-page publisher preview with Prem Watsa's foreword, the 1973 liquidation-value conversion, journal-derived principles, forecasting limits, Bethlehem Copper, and Credit Foncier. It is an edited, affiliated source family.
12. McGill-Queen's University Press book page, 2011 - Authoritative metadata and provenance: Christopher Risso-Gill used interviews, speeches, and exclusive access to Cundill's 45-year journals.
13. JSTOR controlled book record - Chapter metadata and short controlled previews supporting career evolution, adversity, calculated-risk language, and the Graham/Buffett/Cundill blend. Full institutional text was unavailable; snippets are not treated as verified quotations.
14. Tim McElvaine on Cundill's sell process, 2021 - Former colleague's retrospective explanation of the sell-half-after-a-double convention and its behavioral purpose. Useful corroboration, but not contemporaneous Peter-authored evidence.
15. Tim McElvaine, Stock Market Superstars interview - Manager-hosted copy of a published retrospective with important evidence that the flagship portfolio moved from hundreds of securities toward greater concentration. Colleague memory, not an audited portfolio series.
16. Hiok Hhu Ng interview, Washington Post, 2005 - Contemporaneous co-manager account of scandal-driven ideas, a below-two-thirds screen, MCI sizing, and high cash. It attributes the U.S. Ivy vehicle's initial Takefuji decision and later addition to Ng, preventing Peter-only attribution.
17. Rick Rule interview, Mining.com - Former mentee's description of cash-flow-positive net-nets and hidden or redundant assets as recurring Cundill categories. Secondary recollection.
18. Mackenzie Cundill-team profile, 2020 - Official successor source for later quality-and-value scoring, milestones, concentration, and time horizon. It presents a codified later-team process and must not be back-projected.
19. Ivy Funds annual report, 2009 - Useful boundary showing Peter no longer in day-to-day portfolio management. Later formal rules belong to successors unless earlier evidence independently confirms them.
20. Canadian Senate Debates memorial, 2011 - Official memorial preserving attributed last-interview advice about manager selection, homework, patience, and long-term compounding. Senatorial quotation of an interview, not the original interview.
21. Forbes profile, 2004 - Contemporaneous account of Cundill's early Japanese purchases and directional Tokyo short, including the admission that the short began years too early and remained until 1995. Strong tension evidence, but reported narrative rather than a trading ledger.
22. Japan Financial Services Agency press conference, 2011 - Official subsequent context for Takefuji's court-led overpayment claims and restructuring. It confirms the regulatory regime's severity after the 2007 thesis, not what was fully knowable at the earlier decision date.

Evidence limitations

• Five independent lanes covered direct doctrine/chronology, process and numerical mechanics, interview/journal provenance, adversarial losses/legal/regime evidence, and final citation/source QA. Research exceeded 100 meaningful searches and primary-document retrievals before the final audit. All lanes were read-only; only the main agent edited files.
• The Ivey video is the anchor for Peter's personal doctrine. Its automatic captions are noisy, so the philosophy file uses timestamped paraphrase and no caption-dependent long quotation.
• The SEC corpus is unusually strong but mostly describes the U.S. Ivy Cundill Global Value Fund. Commentary may be directly attributed to Peter, jointly attributed to co-managers, or written as a team; holdings, weights, performance, and cash are product-specific.
• Original Peter-era Canadian flagship reports were not recovered. Retrospective counts of hundreds of securities and later concentration come from colleague testimony, not a complete audited series.
• Numerical entry rules varied: the 2002 team often sought at least 50% below NAV, a 2005 co-manager described a below-two-thirds screen, and later actual portfolio discounts could be narrower. They are date/vehicle/opportunity-set evidence, not one timeless formula.
• The 100% currency-hedge statement applied to the relevant U.S.-dollar accounts and was later described less absolutely. Hedge drag in 2002 is one product outcome, not a universal Cundill-fund return effect.
• The authorized book combines journal material, interviews, speeches, and editorial narrative. The publisher preview is only 25 pages; JSTOR chapter snippets are controlled metadata, not full-text access.
• Cable & Wireless is direct self-criticism plus contemporaneous fund evidence. The 2007 consumer-finance episode is a co-managed U.S. product case. Neither supplies a verified Canadian-flagship dollar loss.
• The adviser's June 29, 2006 Form ADV contains no filed criminal, regulatory-action, civil-judicial, judgment/lien, or arbitration disclosure pages. Bounded SEC, Canadian, UK, and general searches located ownership filings, registration records, and exemptive matters but no public securities-enforcement order against Peter; the 1991 corporate tax appeal in Task A is not securities misconduct. This is an adviser/date/jurisdiction-bounded search result, not a personal or global legal clearance.
• Mackenzie's scorecard, milestone rules, and later constraints are successor codifications. They illustrate inheritance and evolution but are not Peter-era doctrine without independent support.

Task C - Greatest Trades (T0472)

As of 2026-07-18. This map supports greatest-trades.md, with source quality ranked roughly from most probative to most derivative. The research used five read-only lanes: primary/fund filings, book/interview sources, trade-candidate reconstruction, adversarial exclusions, and citation/source audit.

Guiding questions

1. Which trades are Cundill-identified or biography-identified as unusually important?
2. Which trade cases have dates, thesis, size, drawdown, exit, and P&L evidence?
3. Which numbers are primary-filed, and which repeat the same authorized-biography source family?
4. Which candidates are really mistakes, successor-team activity, allocator exposure, or unresolved legal curiosities?
5. What should later D-H tasks reuse or avoid?

Annotated source map

1. ValueBuddies repost of David Berman / Report on Business final interview, 2011 - Best recovered support that Cundill himself treated Cleveland-Cliffs as one of his best investments; also excludes Cable & Wireless as worst. Forum repost, not original publisher page.
2. Morningstar India, Peter Cundill: The calculated risk taker, 2016 - Compact secondary source for Bethlehem, JWT, Tiffany, Cleveland-Cliffs, and hidden-asset case details. Mostly derivative; do not treat as independent of Risso-Gill/GuruFocus.
3. GuruFocus, case studies of the Peter Cundill approach, 2013 - Broadest case-study summary for early winners, Cleveland-Cliffs, Sibir, TBS/Japan, Anglo American, and related examples; derivative source family.
4. Authorized preview of There's Always Something to Do, 2011 - Most authoritative opened book preview for early Cundill method, Watsa foreword, Bethlehem/Credit Foncier framing, and track-record context.
5. Online copy of There's Always Something to Do, 2011 - Useful for non-quoted verification of Pan-Ocean, Sibir, Ecuador, Fairfax, Brascan, and other book cases; copyrighted/third-party copy, paraphrase only.
6. McGill-Queen's University Press book page, 2011 - Publisher provenance for the authorized biography and Risso-Gill's access to interviews, speeches, and journals.
7. Ivy Funds semiannual report, June 30, 2002 - Primary U.S. product source for Peter Cundill & Associates process, Japan exposure, and Cable & Wireless as real-time worst performer.
8. Ivy Funds annual report, December 31, 2002 - Primary source for Ivy Cundill Global Value Fund return, Japan 35% country weight, and top-ten holdings including TBS at 3.9%.
9. Ivy Funds annual report, 2009 - Primary source for Fairfax holding evidence in the U.S. Ivy product during the crisis aftermath.
10. SEC Fairfax 13G, 2005 - Confirms Cundill-linked passive beneficial ownership; useful for Fairfax sizing/attribution caveats.
11. Fairfax Q1 2008 release - Primary company source for Fairfax CDS gains; supports Fairfax-as-proxy discussion, not Cundill's direct P&L.
12. Fairfax annual report, 2010 - Primary company source for Fairfax book-value/share-price history and early-2000s drawdown path.
13. Risso-Gill interview, GuruFocus, 2010 - Supports Cundill/Fairfax CDS proxy narrative and book provenance; author/director perspective.
14. Rigzone/Addax release on Pan-Ocean subsidiaries, 2006 - Transaction support for Addax acquisition of Pan-Ocean operating subsidiaries for C$1.605 billion plus debt.
15. Lexpert transaction note on Addax/Pan-Ocean, 2006 - Legal/transaction corroboration for the Pan-Ocean realization event and deal scale.
16. RFE/RL on Gazprom Neft and Sibir, 2009 - Public support for Sibir control-event context and political/transaction backdrop.
17. Jones Day Sibir/Gazprom Neft transaction note, 2009 - Transaction support for Sibir sale context; does not prove Cundill's exit price.
18. Justia, Dalton v. American Investment Co., 1985 - Primary court source confirming Cundill as AIC director defendant and Leucadia's $13 common-share cash-out.
19. TIME archive, Avon/Tiffany, 1978 - Contemporaneous support that Avon's offer was $45/share; use to flag secondary $50 versions as disputed.
20. GuruFocus, staying skeptical of Mr. Market's scorecard, 2013 - Lead for Nikkei put / Japan-short episode; insufficient alone for top-ten inclusion.
21. SEC BRL 13D/A, 1998 - Primary holding/sale lead for BRL; no cost basis or P&L found.
22. SEC Control Data 13D/A, 1995 - Primary sale/remaining-position lead for Control Data; no cost basis or P&L found.
23. Guardian, Cable & Wireless warning, September 2002 - Adversarial context for why Cable & Wireless was a value trap rather than a winner.
24. Guardian, Cable & Wireless losses/liability context, December 2002 - Additional contemporaneous context for C&W collapse and write-down/liability risk.
25. Bond Buyer, Mississippi 1841 bond lawsuit, 1995 - Documents Cundill's distressed-bond litigation claim; excluded as no successful public-market trade/recovery was found.
26. Peter Cundill & Associates Ltd. v. The Queen, 1991 - Historical corporate tax appeal context; not a trade, but useful legal/status caveat.

Evidence limitations

• Five research lanes returned consistent results: Cleveland-Cliffs is the best Cundill-identified candidate; Pan-Ocean and Sibir may have higher multiples but weaker Cundill-specific ledgers.
• Cleveland-Cliffs, Bethlehem, Credit Foncier, JWT, Tiffany, Ecuador, Sibir, and Pan-Ocean exact trade math largely remains in the authorized-biography / derivative source family. Mark precise figures [single-source cluster] until original Cundill Value Fund reports, Canadian filings, or contemporaneous manager letters are recovered.
• SEC/Ivy filings corroborate product holdings and period weights, but the U.S. Ivy product is not the same vehicle as the Canadian Cundill Value Fund and does not prove Peter alone made every decision.
• Fairfax's credit-crisis gains belong to Fairfax. Cundill's trade was Fairfax equity/proxy exposure, with no complete Cundill-level exit ledger found.
• Cable & Wireless, Mississippi bonds, current Mackenzie Cundill holdings, Ashmore allocator exposure, and post-Peter Mackenzie team activity should not be ranked as Peter Cundill greatest trades.
• Saturation searches mostly recycled Risso-Gill, GuruFocus, Morningstar India, Rational Walk, ValueBuddies, and book-preview material; no better primary Canadian fund reports surfaced in this run.

Task D - Mistakes and Losses (T0473)

As of 2026-07-18. This map supports mistakes-and-losses.md. Research used five lanes: concrete loss cases, drawdown/return reconciliation, criticism and legal checks, cross-corpus synthesis, and an independent adversarial audit. Each research lane used fixed guiding questions; the three delegated evidence lanes each completed more than 30 meaningful searches/retrievals and ended with three searches that produced no material new facts.

Guiding questions

1. Which losses can be quantified, and which are self-identified mistakes or opportunity costs?
2. What did Cundill himself say went wrong?
3. When did patience become stubbornness, and when did pressure cause premature capitulation?
4. Which errors came from leverage, cash burn, accounting, regulation, governance, timing, or legal complexity?
5. What documented process changes followed?
6. Which results belong to Peter, a co-managed team, or successors?
7. Where do vehicles, series, periods, currencies, or sources conflict?
8. What does the incomplete and affiliated record prevent us from concluding?

Annotated source map

1. Ivey Peter Cundill lecture, 2005 - Best direct retrospective source for C&W as the worst investment, the negative value of a cash-burning operation, and early buying/selling. Relevant C&W discussion begins around 1:08:34. Both official caption encodings were checked in Task G and do not contain the later devil's-advocate response; that reflection appears in the edited 2011 book instead.
2. Ivy semiannual report, June 2002 - Primary contemporaneous U.S.-product evidence that C&W was the worst performer, remained a core holding, and was added to while tax losses were realized; not the Canadian flagship.
3. Ivy annual report, December 2002 - Primary U.S.-product return, benchmark, currency-hedge, Japan, and broader attribution context.
4. Ivy annual report, March 2007 - Primary evidence for Cundill/Ng co-management, admitted Japanese regulatory miss, three worst performers, portfolio weights, continued additions, and relative underperformance.
5. Ivy annual report, March 2008 - Primary evidence for another loss year, Aiful/Takefuji attribution, holdings, concentration, and the expanding transition-team roster.
6. Ivy annual report, March 2009 - Primary evidence for the U.S. product's -33.87% return and -42.58% benchmark. The discussion is presented under Massie/Thompson, but the filing says Massie managed during the fiscal year and Thompson became co-manager after period-end; not evidence for the Canadian flagship.
7. Cable & Wireless Form 20-F - Primary company evidence for large operating losses, impairments, restructuring charges, and why gross cash overstated liquidation protection.
8. Mackenzie/Morningstar historical chart - Official sponsor exhibit for calendar returns, including 1990 -9.4%, 1998 -10.7%, 2002 -13.8%, and 2008 -25.1%; predecessor/current Series A bridge remains unexplained.
9. Current Mackenzie MRFP - Current legal-series inception evidence used to qualify the sponsor's long historical Series A presentation.
10. Authorized biography, online copy - Main source for C&W Canadian trade arithmetic, Westar, LTV, the 1989 mistake list, 1991 process reset, Japan timing, and Tiffany; copyrighted third-party copy, paraphrase only, and exact trade figures remain one source family.
11. McGill-Queen's University Press book page - Publisher provenance for the authorized biography and its use of interviews, speeches, and journals.
12. Guardian, C&W warning, September 2002 - Contemporaneous reporting on the fourth profit warning in less than two years and deteriorating operating expectations.
13. Guardian, C&W restructuring, November 2002 - Contemporaneous reporting on the £4.43 billion loss, job cuts, and restructuring.
14. Guardian, C&W liability, December 2002 - Contemporaneous evidence for the possible £1.5 billion tax liability and the collapse of the cash-margin thesis.
15. Washington Post interview with Hhu Ng, 2005 - Contemporaneous attribution of the original Takefuji work and additions particularly to co-manager Ng.
16. Japan FSA reform summary, 2006 - Primary regulatory description of interest-rate, underwriting, and total-lending restrictions.
17. Bank of Japan Financial System Report, 2008 - Primary evidence that the five largest consumer lenders recorded fiscal-2006 net losses due to refunds and loan-loss provisions.
18. Japan FSA ministerial briefing, February 2011 - Primary aggregate reimbursement figures for three major consumer lenders.
19. Japan FSA ministerial briefing, March 2011 - Primary context for Takefuji reorganization and the scale of creditor claims.
20. TIME, Avon/Tiffany, 1978 - Contemporaneous $45 Avon offer evidence; conflicts with the authorized book's approximate $50 figure.
21. Los Angeles Times on LTV pensions, 1989 - Contemporaneous external evidence for the scale of LTV's pension complexity.
22. Second Circuit LTV pension litigation - Legal context for the union, benefit, bankruptcy, and pension issues omitted from a simple asset appraisal.
23. Advisor.ca on the Cundill team, 2010 - Contemporary boundary evidence that Massie had portfolio-management responsibility for the Value Fund since 2004 and later co-led the team.
24. Peter Cundill & Associates Ltd. v. The Queen - Adverse corporate tax ruling; not a personal securities-enforcement or fraud finding.
25. Dalton v. American Investment Co. - Fiduciary allegation against a board including Cundill that ended in judgment for defendants; useful allegation/finding boundary.
26. SEC/IAPD Form ADV, June 2006 - Adviser/date-bounded absence of filed disciplinary-response pages, not a lifetime or global clearance.
27. Japan FSA, full implementation of amended Money Lending Business Act, 2010 - Primary evidence that the reform passed in December 2006 and later came fully into force; complements the 2006 bill-design summary.
28. Aiful business-revitalization plan - Primary company evidence for Aiful's alternative dispute-resolution restructuring and continuing reimbursement pressure.
29. Ontario Securities Commission Bulletin, September 2006 - Primary administrative record of consent to suspension under the registration-surrender rule; not an adverse enforcement finding.
30. Ivey advisory-board biography - Institutional context for the 2006 Mackenzie acquisition and Cundill's continuing team role, used to interpret the registration surrender without implying a sanction.

Evidence limitations

• C&W's just-under-$100-million cost, roughly 6% weight, nearly $59 million loss, and approximately -11% Canadian-fund year are one authorized-biography source cluster. The source does not identify the dollar currency. SEC Ivy filings corroborate behavior in a different U.S. product, not those Canadian figures.
• The biography's approximately -11% Canadian 2002 return conflicts with the sponsor chart's -13.8%; its 1998/1999 pair of about -9%/+16% conflicts with -10.7%/+33.4%. Treat these as unreconciled series, period, or convention differences.
• Canadian calendar returns and U.S. Ivy fiscal returns have different vehicles, currencies, fees, holdings, periods, and manager rosters. They must not be spliced. Annual figures are point-to-point observations, not maximum drawdowns.
• Peter/Ng co-managed the 2007 consumer-finance record; Burton/Massie joined the 2008 transition. The 2009 discussion is presented under Massie/Thompson, but the filing assigns fiscal-period management to Massie and says Thompson joined after period-end; later recovery belongs to the successor team. Initial Takefuji work is particularly associated with Ng.
• Cundill-specific realized losses for Aiful, Takefuji, Acom, LTV, BC Capital, and the other 1989 mistakes were not recovered. Later Takefuji/Aiful outcomes demonstrate severity, not perfect ex-ante foreseeability.
• The biography is authorized but retrospective and affiliated; sponsor histories foreground a surviving flagship; derivative articles often recycle the same source family. No complete trade ledger, audited predecessor-series bridge, monthly drawdown history, or all-vehicle comparison was found.
• The legal search distinguishes allegation, adverse tax finding, defense judgment, administrative status, and bounded absence. It cannot support a claim of a universally clean record.
• Saturation checks returned the same biography family, SEC filings, sponsor chart, and derivative summaries. No independent Canadian trade ledger, realized-loss schedule, or better return-series reconciliation surfaced.

Task E - In His Own Words (T0474)

As of 2026-07-18. This map supports in-their-own-words.md. Research used five lanes: direct recordings, print interviews and institutional statements, authorized journals/books, SEC authorship, and an independent provenance/copyright audit. The first three evidence lanes completed more than 30 meaningful searches or retrievals; direct-recording saturation found no second authentic Peter Cundill recording.

Guiding questions

1. Which surviving words are Peter-only, joint team commentary, institutional copy, reported recollection, or editor-selected journal material?
2. What is the underlying speech, interview, report, or composition date, and how does it differ from filing, upload, or publication date?
3. Can every fragment be checked on an original publisher, official recording, filing, authorized preview, or controlled book record?
4. Do mirrors, reposts, alternate encodings, or book excerpts belong to the same underlying source family?
5. Which quotations expose mistakes, changing views, and team boundaries rather than merely repeating polished doctrine?
6. What primary materials exist only in archives, behind controls, or as bibliographic leads?
7. Which popular formulations are misattributed or laundered through quote sites?

Annotated source map

1. Official Ivey guest-speaker index - Institutional provenance for Cundill's 2005 class appearance and the canonical recording.
2. Ivey Peter Cundill lecture, 2005 - Sole verified full direct recording. Four short caption-exact fragments were cross-checked against an alternate Ivey encoding; the two uploads remain one source family.
3. Ivy semiannual report, June 2002 - Peter-only fund interview, speaking in we for the team. The separate “Cundill Story” feature is institutional Mackenzie copy except where it explicitly tags quotations to him.
4. Ivy annual report, December 2002 - Peter-only interview for currency hedging, Japan, security switching, and 2003 strategy; hosted as an annual-report exhibit within an N-14.
5. Ivy annual report, March 2005 - Unsegmented joint interview with Cundill and Hhu Ng; no answer is attributable to one of them alone.
6. Ivy annual report, March 2006 - Joint Cundill/Ng interview documenting idea scarcity, cash, concentration, and Japanese consumer-finance exposure.
7. Ivy annual report, March 2007 - Joint Cundill/Ng discussion admitting the Japanese regulatory-risk miss and documenting continued additions; amended N-CSR/A filed later than the report period.
8. Ivy annual report, March 2008 - Collective discussion by Cundill, Ng, Wade Burton, and Andrew Massie during the transition; not Peter-only speech.
9. The Wall Street Transcript interview, 2001 - Original publisher exposes a direct excerpt but controls the remainder.
10. Forbes, “Found in Translation,” 2004 - Contemporaneous profile with directly attributed wording about Japan and the premature market short.
11. Mackenzie/Fund Library feature, 2002 mirror - Reproduces Cundill's Career Achievement Award poem and a 1996 internal memo excerpt; used as a mediated bridge because the originals were not recovered.
12. Advisor.ca fund-closing report, 2006 - Trade-press report attributing to Cundill the remark that global bargain scarcity motivated restricting new purchases.
13. Mackenzie acquisition news release, 2006 - Original one-page institutional release with Cundill's direct rationale for formalizing the relationship.
14. McGill Reporter interview, 2008 - Direct source for the analogy between historical and investment research and the new history prize.
15. The Globe and Mail final interview, 2011 - Original edited Q&A with Cundill's last published ordinary-investor advice. Publication followed his death; precise interview date undisclosed.
16. Canadian Senate Debates memorial, 2011 - Official memorial preserving the 1985 “Don't send me your money” AGM recollection and the final-interview answer; not the original 1985 transcript.
17. Peter Cundill Foundation biography - Affiliated source for an undated criteria adage, book provenance, and the distinction between the 2011 investment book and 2014 full biography.
18. McGill-Queen's University Press page for There's Always Something to Do - Canonical 2011 metadata and provenance: Risso-Gill used interviews, speeches, and exclusive access to the 45-year daily journal.
19. Authorized preview of There's Always Something to Do - Publisher preview with title/copyright pages, Prem Watsa foreword, early journal material, and facsimile evidence; one 2011 book family.
20. Google Books record for There's Always Something to Do - Limited searchable record used only where exact returned text and printed pages were visible: December 1973 entry, p.4 and curiosity fragment, p.185.
21. JSTOR record for There's Always Something to Do - Controlled e-book metadata and chapter map; inaccessible content was not reconstructed.
22. UTP distribution page for Routines and Orgies - Canonical 2014 title, ISBN, format, page count, and publication metadata.
23. Authorized preview of Routines and Orgies - Publisher preview establishing Foundation copyright, editor/source method, and 44-year journal description.
24. Google Books record for Routines and Orgies - Limited searchable record for the exact 1991 retrospective fragment and printed p.173.
25. JSTOR record for Routines and Orgies - Controlled metadata only. Its appended “Philanderer” text conflicts with the actual title page and other catalogues and is treated as a metadata error.
26. McMaster Peter C. Newman fonds catalogue - Records an undigitized October 23, 1980 Cundill interview, 22-page transcript, and audio cassette 32; indexed without quotation.
27. Risso-Gill interview, GuruFocus, 2011 - First-hand provenance that Cundill invited Risso-Gill to write the book as illness prevented an autobiography. It is editor testimony, not Cundill speech.
28. Investment Executive author archive, 2009 page - Bibliographic evidence for the chairman-emeritus announcement. The article itself currently returns unavailable, so its indexed quotation was excluded from the anthology.
29. Deseret News syndicated investor notebook, 1989 - Repeats a short liquidation-analysis remark that it attributes to an unidentified recent Forbes interview. The original Forbes item was not recovered, so the wording remains an index lead rather than an anthology fragment.

Evidence limitations

• The 31-fragment ledger spans 18 underlying source families. Each fragment is at most 25 words and each family's combined quotation is at most 25 words; mirrors, alternate uploads, and snippets do not reset the allowance.
• Only the two 2002 Ivy Q&As are formally Peter-only. The 2005–06 text is joint Cundill/Ng interview copy, 2007 is a joint Cundill/Ng discussion, and 2008 is a four-manager discussion. Standard risk, performance, holdings, style-box, and prospectus language is issuer copy.
• The Ivey lecture is the sole verified direct recording found after direct-video saturation. Its automatic captions are navigational rather than a human transcript; only short wording shared by two Ivey encodings is used and labeled.
• The raw journals are private. Cundill's composition date, Risso-Gill's selection, and the 2011/2014 publication dates are separate facts. The books' 44-year, 45-year, and 1963–2007 descriptions are not forced into a false audited inventory.
• The 1980 Newman interview, 1985 AGM, 1996 memo, 2004 Canadian Business item, Barron's profiles, The Province material, and Outstanding Investor Digest appearances remain controlled, undigitized, partial, or bibliographic leads.
• No authentic Peter Cundill podcast was located. Later recordings feature protégés, biographers, posthumous narrators, or investors discussing him.
• Quote aggregators and derivative summaries were discovery aids only. Well-known Buffett/Templeton sayings were excluded, and the authorized 2011 book credits the title phrase “there's always something to do” to Irving Kahn's advice rather than to Cundill's authorship.
• After the 1989 Forbes-via-syndication lead was indexed, the final three saturation searches returned that same lead, quote aggregators, and book-derived summaries; no new quote-ready primary family or direct recording surfaced.
• The independent final audit returned NO-GO on one overstrong attribution label for the 2006 fund-closing remark. It was repaired from direct institutional statement to press-attributed remark in both files; the exact repaired draft then received GO with no remaining material defect.

Task F - Key Writings (T0475)

As of: 2026-07-18T04:34:48Z

Guiding questions

• What did Peter Cundill actually write or say in public, versus what was later edited by Christopher Risso-Gill or successor institutions?
• Which works best communicate the Cundill investment process: checklist, margin of safety, net-net analysis, global search, patience, fieldwork, sell discipline, and stress behavior?
• Which books and articles about him are worth using, ranked by source quality and analytical usefulness?
• What sources should be avoided or downgraded because of quote laundering, copyright problems, weak provenance, or successor-team attribution?

Annotated source map

1. McGill-Queen's University Press, There's Always Something to Do: The Peter Cundill Investment Approach - https://www.mqup.ca/there-s-always-something-to-do-products-9780773538634.php - Official publisher page; confirms author Christopher Risso-Gill, February 2011 publication, 248 pages, book framing, journal/interview/speech provenance, and table of contents including the net-net worksheet and holdings appendices.
2. JSTOR, There's Always Something to Do - https://www.jstor.org/stable/j.ctt7zrfw - Bibliographic and controlled-access record; confirms EISBN, finance subject, table of contents, visible chapter snippets, and institutional-access limitation.
3. Google Books, There's Always Something to Do - https://books.google.com/books/about/There_s_Always_Something_to_Do.html?id=RgkIIBSF1PkC - Bibliographic confirmation of Risso-Gill, McGill-Queen's Press/MQUP, 2011, 237-page Google pagination, and subject categories.
4. Internet Archive, There's Always Something to Do - https://archive.org/details/theresalwayssome0000riss - Bibliographic controlled-lending record; confirms publisher, language, index, page count, and access-restricted/print-disabled status.
5. Pageplace preview PDF, There's Always Something to Do - https://api.pageplace.de/preview/DT0400.9780773581098_A23656995/preview-9780773581098_A23656995.pdf - Authorized preview with copyright page, table of contents, Prem Watsa foreword, and first-chapter text; useful for provenance and careful page-location checks, not long quotation.
6. McGill-Queen's University Press, Routines and Orgies - https://www.mqup.ca/routines-and-orgies-products-9780773544727.php - Official publisher page; confirms author, biography framing, four decades of Cundill journals, close-friend/confidant framing, publication metadata, full table of contents, Appendix A net-net worksheet, and Appendix B "Notes from a Battle-Scarred Warrior."
7. JSTOR, Routines and Orgies - https://www.jstor.org/stable/j.ctt1287h4s - Bibliographic and chapter-map source; visible chapter 27 snippet identifies "Notes from a Battle-Scarred Warrior" as a defining Cundill investment-philosophy statement since the 1976 letter.
8. Google Books, Routines and Orgies - https://books.google.com/books/about/Routines_and_Orgies.html?id=sPJsBgAAQBAJ - Bibliographic confirmation of title, author, McGill-Queen's University Press, 2014, ISBNs, and 422-page Google pagination.
9. Ivey Ben Graham Centre video library - https://www.ivey.uwo.ca/bengrahaminvesting/resources/video-library/ - Official institutional listing for the March 28, 2005 Peter Cundill Ivey Value Investing Classes video.
10. Ivey Ben Graham Centre guest-speaker page - https://www.ivey.uwo.ca/bengrahaminvesting/events/guest-speakers/ - Confirms Cundill as Principal of The Cundill Group, founder of Peter Cundill & Associates, and links the 2005 presentation.
11. YouTube, 2005 Ivey Value Investing Classes Guest Speaker: Peter Cundill - https://www.youtube.com/watch?v=JUP_hyqYpdU - Direct video lead; no official transcript recovered, so use only with timestamp verification.
12. SEC, Ivy Funds semiannual report, June 30, 2002 - https://www.sec.gov/Archives/edgar/data/52858/000095014402009301/g77245nv30d.txt - Best compact official process source; covers "Zigging While Others Are Zagging," margin of safety, NAV discount, bottom-up process, sum-of-the-parts work, management and competitor checks, catalysts, and holding periods.
13. SEC, Ivy Cundill Global Value Fund annual report, Dec. 31, 2002 - https://www.sec.gov/Archives/edgar/data/883622/000110560703000032/exh17giv_ivyar123102.htm - Direct Q&A with Peter Cundill as management team leader; useful for down-year framing, hedging, Japan drawdown, portfolio upgrading, and 2003 strategy.
14. SEC, Ivy Funds SAI, 2006 - https://www.sec.gov/Archives/edgar/data/52858/000110560706000232/ivytsais7-31_06main.htm - Entity/provenance source for Cundill Investment Research, Mackenzie connection, subadviser role, portfolio manager disclosures, compensation, and Peter Cundill and Associates proxy voting policy.
15. SEC/IAPD Form ADV, Cundill Investment Research Ltd. CRD 111887 - https://reports.adviserinfo.sec.gov/reports/ADV/111887/PDF/111887.pdf - Public adviser filing; shows CRD number and no criminal/regulatory/civil/judicial/bond/judgment/lien/arbitration DRP information filed as of 2006 amendment.
16. ValueBuddies repost, "Peter Cundill's last interview" - https://www.valuebuddies.com/thread-735-nextoldest.html - Best accessible text of David Berman/CTV/Report on Business final Q&A; useful but derivative and encoding-damaged. Original Globe/CTV page not recovered.
17. Senate of Canada Hansard, Feb. 2, 2011 - https://sencanada.ca/en/content/sen/chamber/403/debates/081db_2011-02-02-e - Official memorial corroborating Cundill's death date, career/philanthropy context, and final-interview investor-advice excerpt.
18. Peter Cundill Foundation biography - https://www.thepetercundillfoundation.com/peter-cundill.html - Affiliated/institutional biography; useful for education, fund history, 13.2%/35-year track-record framing, assets under management, athletics, curiosity, and foundation context.
19. Cundill Prize press archive - https://www.cundillprize.com/press-archive/p7 - Source-discovery page for obituaries, Globe final interview, McGill/Mackenzie/Aspen items, and Cundill Prize-related context.
20. Rational Walk review of There's Always Something to Do - https://rationalwalk.com/book-review-theres-always-something-to-do/ - Strong independent practitioner review; highlights journal access, Risso-Gill's fund-director provenance, sell discipline, cash through valuation sales, and case-study usefulness. Use as secondary, not primary quote source.
21. Business Insider / Daily Reckoning, Chris Mayer review - https://www.businessinsider.com/theres-always-something-to-do-2011-5 - Derivative review/excerpt; confirms Risso-Gill's Cundill Value Fund directorship and summarizes book themes. Useful for reception and leads, not exact quote provenance.
22. Net Net Hunter, "Peter Cundill Investing: Your Essential Guide" - https://www.netnethunter.com/essential-guide-peter-cundill-investing/ - Useful practical summary and secondary reproduction of the 1976 All-Canadian Venture Fund checklist; original letter still unrecovered.
23. Net Net Hunter book review - https://www.netnethunter.com/peter-cundill-theres-always-something-to-do/ - Secondary review; useful for reception and limitations of the book as biography versus detailed manual.
24. Tax Interpretations, Peter Cundill & Associates Ltd. v. Her Majesty the Queen - https://taxinterpretations.com/content/615283 - Canadian tax/arm's-length appeal involving fees paid to Bermuda entity; relevant legal context but not securities misconduct.
25. Bond Buyer, "The Debt That Won't Die" - https://www.bondbuyer.com/news/the-debt-that-wont-die-new-suit-hits-mississippiover-1841-bond - Contemporaneous trade-press source on Cundill as plaintiff in Mississippi defaulted-bond litigation; useful for distressed-debt context.
26. Pageplace preview PDF, Routines and Orgies - https://api.pageplace.de/preview/DT0400.9780773596870_A23658206/preview-9780773596870_A23658206.pdf - Search/opened as a public preview source; useful for title/copyright/TOC leads, but retrieval was partial and MQUP/JSTOR were cleaner anchors.
27. Outstanding Investor Digest, Mar. 13, 1998 issue lead - https://theoraclesclassroom.com/wp-content/uploads/2023/12/1998-OID-13-Mar-XIII-12.pdf - Located in search as potentially valuable Cundill media/interview material; clean readable retrieval failed in this run, so not used for substantive claims.
28. Outstanding Investor Digest, Dec. 29, 1997 issue lead - https://theoraclesclassroom.com/wp-content/uploads/2023/12/1997-OID-29-Dec-XII-3.pdf - Located as a possible Cundill/Japan context lead; not used substantively without clean retrieval.
29. Forbes "Buy Japan Now" lead - searched as a contemporaneous 2004 media source; page did not render cleanly enough in this run, so it remains a lead rather than citation support.
30. McGill History Department Cundill Prize page - https://www.mcgill.ca/history/news-events/cundill-history-prize - Institutional source confirming F. Peter Cundill (1938-2011), McGill affiliation, and Cundill Prize/fellowship context.

Adversarial and provenance notes

• The two Risso-Gill books are about Cundill, not authored by Cundill. They are unusually valuable because of journal/interview/speech access, but they carry insider and foundation-adjacent bias risk.
• Exact Cundill quotes should be page-checked in MQUP/JSTOR/library/authorized copies or timestamped against the Ivey video. Do not quote from Goodreads, A-Z Quotes, podcast transcripts, AI transcripts, Scribd, Dokumen, or other large-copy mirrors.
• The 1976 letter and "Notes from a Battle-Scarred Warrior" are the highest-priority archival targets for future improvement.
• The final interview is useful through ValueBuddies and partly corroborated by Senate Hansard, but the original Globe/CTV text was not recovered.
• Bounded searches for Peter Cundill/Cundill Investment Research plus SEC/OSC/CSA/enforcement/lawsuit as of 2026-07-18 found tax litigation and plaintiff-side distressed-bond litigation, but no opened primary SEC/OSC/CSA enforcement action against Peter Cundill personally. This is public-source negative evidence, not legal clearance.
• Current Mackenzie Cundill materials and post-Cundill fund documents may describe lineage or successor process only. Do not attribute post-2009 successor performance or commentary to Peter Cundill personally.

Search log

Targeted searches included: Peter Cundill key writings, Peter Cundill 1976 All-Canadian Venture Fund letter, Peter Cundill net-net worksheet, Peter Cundill Notes from a Battle-Scarred Warrior, Peter Cundill Ivey 2005 lecture/video/transcript, Peter Cundill final interview David Berman Globe CTV, Peter Cundill Outstanding Investor Digest 1997/1998, Peter Cundill Forbes Japan 2004, Peter Cundill lawsuit/SEC/OSC/CSA/enforcement, Peter Cundill tax case, Peter Cundill Mississippi bonds, Christopher Risso-Gill There's Always Something to Do, Christopher Risso-Gill Routines and Orgies, and current Mackenzie Cundill lineage queries.

Task G - Mental Models (T0476)

As of 2026-07-18. This map supports mental-models.md. Research used exactly five lanes: primary model evidence, operational cases, interviews and edited journals, adverse/falsifying evidence, and independent audit. Together they completed more than 250 meaningful searches or retrievals and reached saturation. Labels distinguish Cundill's direct statements, inherited screens he used, reconstructions, team products, and successor practices.

Guiding questions

1. Which decision rules did Cundill state or demonstrably use, and which are later reconstructions?
2. Which rules were inherited from Graham, Templeton, Kahn, or colleagues rather than invented by Cundill?
3. How did balance-sheet valuation, catalysts, fieldwork, people, sizing, patience, and sale discipline interact?
4. What evidence would falsify each model or turn patience into stubbornness?
5. Do the rules survive contact with C&W, Japan consumer finance, Tiffany, Sibir, Pan-Ocean, and other cases?
6. Which observations belong to Peter alone, the Cundill/Ng team, the four-manager transition, or successors?
7. What can an individual investor reproduce without Cundill's access, organization, or capital base?

Annotated source map

1. Ivey Ben Graham Centre video library - Official institutional provenance for the March 28, 2005 class and its canonical recording.
2. Ivey Peter Cundill lecture, 2005 - Sole verified full direct recording; supports scarcity screens, travel, case learning, portfolio snapshots, C&W's cash-burn lesson, fresh review, selling half, and a source-conflicted maxim about changing a winning game.
3. Ivy semiannual report, June 2002 - Peter-only interview and institutional process feature covering margin of safety, sum-of-the-parts work, management/competitor checks, catalysts, a 50%-of-NAV aspiration, and a three-to-five-year horizon for that U.S. product.
4. Ivy annual report, December 2002 - Peter-only Q&A on hedging, Japan, replacing positions, and preserving cash for better bargains.
5. Ivy annual report, March 2005 - Joint Cundill/Ng interview; supports team-level balance-sheet work and portfolio construction, not Peter-only attribution.
6. Ivy annual report, March 2006 - Joint Cundill/Ng evidence for idea scarcity, cash as a residual, concentration, and Japanese consumer-finance exposure.
7. Ivy annual report, March 2007 - Joint Cundill/Ng admission that regulatory risk was missed; primary falsifier for cheapness without a complete liability map.
8. Ivy prospectus, April 2007 - Mandate and personnel boundary for the Cundill-managed U.S. product.
9. Ivy annual report, March 2008 - Four-manager transition evidence; supports team-level holdings and loss attribution rather than Peter-only rules.
10. Ivy annual report, March 2009 - Successor-period boundary and severe drawdown context; not evidence of Cundill's personal 2009 decisions.
11. McGill-Queen's University Press, There's Always Something to Do - Publisher provenance for Risso-Gill's edited 2011 investment book based on journals, interviews, and speeches.
12. Authorized 2011 preview - Copyright, contents, foreword, and early journal evidence; used for provenance and paraphrase, not bulk reproduction.
13. JSTOR controlled record, 2011 book - Confirms Appendix 3 is a net-net worksheet designed by Peter in 1975; inaccessible worksheet fields were not reconstructed.
14. Google Books, 2011 book, p.158 - Controlled page preserving the editor-mediated Cundill reflection that a formal dissenter might have interrupted the C&W error; this replaces an earlier mistaken attribution to the Ivey recording.
15. McGill-Queen's University Press, Routines and Orgies - Official provenance for the 2014 biography and journal-based account, including the net-net worksheet and “Battle-Scarred Warrior” appendices.
16. Authorized 2014 preview - Establishes editorial method and journal provenance; controlled content was not reverse-engineered.
17. JSTOR controlled record, 2014 book - Chapter-map and bibliographic support; metadata anomalies were not treated as substantive evidence.
18. The Globe and Mail final interview - Original edited Q&A for ordinary-investor discipline and the distinction between simple principles and difficult execution.
19. McGill Reporter interview, 2008 - Direct analogy between historical research and investing: evidence gathering, interpretation, and skepticism.
20. Peter Cundill Foundation biography - Affiliated overview for career and book provenance; useful context but not independent validation.
21. Forbes, “Found in Translation,” 2004 - Contemporaneous profile on Japan, contrarian patience, and an acknowledged premature market short.
22. Business Day, “Street Dogs Taking a Hit,” 2015 - Secondary recollection attributing “always change a winning game” to Peter Robertson; conflicts with Cundill's Ivey attribution to Templeton and Graham.
23. McElvaine Investment Management, sale discipline - Practitioner recollection of Cundill's sale framework; useful triangulation, not a direct Cundill record.
24. Washington Post interview with Hhu Ng, 2005 - Contemporaneous team-attribution evidence for Takefuji research and additions, transport-fragile and not used alone for an essential claim.
25. Fairfax 2008 annual report - Independent institutional context for Cundill's long association with Prem Watsa and value-investing networks.
26. Cable & Wireless Form 20-F - Company filing for cash burn, impairments, restructuring, and why gross cash could not be equated with realizable value.
27. Sibir Energy transaction record - External transaction anchor for eventual asset realization; later recovery is not automatically attributed to Peter.
28. Pan-Ocean transaction record - External deal anchor for testing asset-value/catalyst reasoning.
29. Japan FSA reform summary, 2006 - Primary regulatory evidence for rate, underwriting, and total-lending restrictions missed in the original consumer-finance thesis.
30. Bank of Japan Financial System Report, 2008 - Primary evidence that refunds and provisions drove losses at major consumer lenders.
31. Japan FSA ministerial briefing, February 2011 - Primary aggregate reimbursement data, demonstrating the liability scale behind the value-trap falsifier.
32. LTV pension opinion - Legal evidence that pension and bankruptcy claims complicate a simple asset-minus-debt appraisal.
33. TIME, Avon/Tiffany, 1978 - Contemporaneous offer-price evidence used to bound the Tiffany case rather than rely on retrospective memory alone.
34. Mackenzie historical mountain chart - Sponsor return exhibit; useful for stress periods but not a reconciled predecessor/current legal-series bridge.
35. Current Mackenzie Cundill team profile - Successor-process evidence only; current milestones and scorecards are not retroactively Peter's model.
36. Current Mackenzie Cundill Value Fund profile - Current product and lineage boundary; post-Cundill holdings, returns, and managers are excluded from personal attribution.
37. Ontario Securities Commission Bulletin, September 2006 - Administrative consent to suspension following registration surrender, not an enforcement sanction or investment-model endorsement.
38. Peter Cundill & Associates Ltd. v. The Queen - Adverse corporate tax decision; relevant governance context but not securities fraud or a personal disciplinary finding.
39. Authorized-biography online text - Third-party online copy of Risso-Gill's authorized 2011 book; used only to paraphrase the Pan-Ocean fieldwork and owner/control narrative. It remains one edited-book source family, not independent primary evidence or permission for extensive quotation.

Evidence limitations

• “Magic Sixes” is documented as a screen Cundill used, but no primary source found in this run establishes that he invented it.
• The surviving 1975 worksheet is identified by an authorized controlled record; its unavailable fields were not reconstructed from unauthorized copies.
• Cundill used net-net, NAV, sum-of-the-parts, future-NAV, earnings, and cash-flow reasoning. His skepticism of false precision is not a rejection of modeling.
• The 50%-of-NAV entry aspiration and three-to-five-year horizon come from the 2002 U.S. Ivy product discussion, not a timeless universal rule.
• The lecture's roughly 55 names/55,000 screened securities, 75% in 25 holdings, and sell-half-after-a-double observations are snapshots, not audited universal constraints.
• The 2005–07 reporting is joint Cundill/Ng; the 2008 discussion has four managers; later Mackenzie rules, recoveries, and performance belong to successors unless separately attributed.
• The devil's-advocate remedy is absent from both official Ivey caption streams and appears on controlled Google Books p.158. The Ivey lecture and later Business Day recollection conflict on who coined the changing-game maxim.
• The book title phrase is Irving Kahn's advice in the authorized account, not a Cundill-authored maxim.
• Cundill died in 2011. The tax case is not a securities-fraud finding, and the OSC registration-surrender suspension is not an enforcement sanction; bounded searches cannot establish a blanket clean record.
• No complete trade ledger, raw journal, audited holdings history, or reconciled predecessor-series bridge was recovered. Case arithmetic and reconstructed gates therefore remain explicitly labeled.

Search log

Targeted searches covered Cundill net-net worksheet/1975 design, Magic Sixes origin and use, liquidation/NAV/SOTP valuation, catalysts and holding periods, travel and management checks, cash policy, concentration, sell discipline, C&W and dissent, Japan consumer finance/regulation/refunds, Tiffany, Sibir, Pan-Ocean, LTV pensions, Westar, BC Capital, direct interviews/recordings, journal provenance, team attribution, current successor process, legal/adverse records, and source conflicts. Each lane ended after three consecutive searches or retrieval paths yielded no material new model or falsifier.

Task H - Synthesis (T0477)

As of 2026-07-18. This map supports synthesis.md. The synthesis inspected the complete Task A-G Canon corpus and its evidence ledger, then used exactly five independent lanes: corpus and ranked lessons; taxonomy and regime dependence; completed-Canon comparables; adverse gaps and unresolved questions; and independent audit. No new derivative search was needed because the preceding tasks had reached saturation and the synthesis could be grounded in their controlled, primary, official, and explicitly labeled affiliated sources.

Guiding questions

1. What is Cundill's durable edge after separating inherited screens from his operating contribution?
2. Which lessons remain useful after contact with C&W, Japan consumer finance, Westar, LTV, and the 2008 transition drawdown?
3. In which regimes should asset-backed contrarianism be strong, mixed, or weak, and why?
4. What belongs to Peter, a Peter-led team, the Cundill/Ng product, the four-manager transition, or successors?
5. Which completed Canon investors are closest and most opposite at the process level?
6. What evidence is still required before exact returns, trade economics, or personal alpha can be treated as reconciled?

Annotated synthesis evidence map

Source priority for this synthesis was: Tier 1 direct recordings, filings, regulators, and courts; Tier 2 controlled authorized-book pages, publisher records, institutional interviews, and transaction corroboration; Tier 3 sponsor or affiliated history; and Tier 4 prior Canon synthesis used only for comparison. The seven internal Cundill documents organize those families but do not create independent corroboration.

1. profile.md - chronology, status, fund and firm lineage, team scale, performance-claim boundaries, and legal-record distinctions.
2. investment-philosophy.md - screens, valuation hierarchy, fieldwork, catalysts, cash, portfolio construction, and product-specific thresholds.
3. greatest-trades.md - cross-case evidence for asset appraisal, catalysts, distressed claims, and the limits of single-book trade arithmetic.
4. mistakes-and-losses.md - C&W, Japan, Westar, LTV, early tactical rigidity, transition drawdown, and patience failure modes; the Task G correction supersedes its legacy Ivey attribution for the devil's-advocate remedy.
5. in-their-own-words.md - direct, attributed, joint, edited, and successor voice boundaries.
6. key-writings.md - authorized-book provenance, private-journal boundary, unrecovered writings, and the distinction between Cundill as source/subject and Risso-Gill as author/editor.
7. mental-models.md - operational rules, falsifiers, source conflicts, reproducibility limits, and the corrected controlled source for formal dissent.
8. Ivy semiannual report, June 2002 - Peter-only interview speaking for a team; supports the process sequence and product-specific 50%-of-NAV and three-to-five-year aspirations.
9. Ivy annual report, March 2006 - joint Cundill/Ng evidence for idea scarcity, cash as residual, concentration, and Japanese exposure.
10. Ivy annual report, March 2007 - joint acknowledgement of the Japanese regulatory miss and continued exposure; primary falsifier for cheapness without refreshed legal liabilities.
11. Ivy annual report, March 2008 - four-manager transition boundary; holdings and losses are not Peter-only evidence.
12. Ivey Peter Cundill lecture, 2005 - sole verified full Peter-only recording; supports the process, valuation, fieldwork, portfolio snapshots, fresh review, C&W postmortem, and sale discipline.
13. McGill-Queen's University Press, There's Always Something to Do - publisher provenance for Risso-Gill's authorized edited book based on journals, interviews, and speeches.
14. Authorized 2011 preview - Watsa foreword, early process, affiliated performance claim, and journal provenance; used by paraphrase.
15. Google Books, 2011 book, p. 158 - controlled page for the editor-mediated Cundill reflection on appointing a formal dissenter after C&W; it replaces the mistaken legacy Ivey attribution.
16. Cable & Wireless Form 20-F - primary evidence that gross cash could coexist with severe operating losses, impairments, restructuring, and negative business value.
17. Japan FSA reform summary, 2006 - primary legal-regime evidence for rate, underwriting, and lending restrictions.
18. Bank of Japan Financial System Report, 2008 - primary evidence for refund claims, provisions, and losses at consumer lenders.
19. LTV pension opinion - primary legal evidence that priority, pensions, and bankruptcy duration complicate simple asset-value arithmetic.
20. Mackenzie historical mountain chart - sponsor annual-return history and 2008 drawdown, without a complete predecessor/current-series bridge.
21. Current Mackenzie MRFP - legal-series and successor boundary; not Peter's current record.
22. Mackenzie acquisition release, 2006 - transaction structure and team-managed scale, with currency unstated.
23. Pan-Ocean transaction record and Sibir transaction record - independent asset-realization anchors, not fund-level entry, exit, or return ledgers.
24. Peter Cundill & Associates Ltd. v. The Queen, OSC Bulletin, and Dalton v. American Investment Co. - controlled legal boundaries: adverse corporate tax finding, surrender-related suspension, and defense judgment respectively.
25. John Templeton synthesis, Anthony Bolton synthesis, Benjamin Graham synthesis, and William J. O'Neil synthesis - completed Canon evidence for the closest, ancestor, and opposite process comparisons.

Synthesis limitations

• The headline returns remain sponsor- or affiliated-reported claims with different endpoints; the post-2008 segment belongs to successors, and no original audited predecessor-series bridge was recovered.
• Exact named-trade economics remain disproportionately concentrated in one authorized-book and derivative source family; transaction records validate realizations, not fund P&L.
• Regime classifications are evidence-informed synthesis, not a formal causal or factor-return study.
• Screens and snapshots were not promoted into universal formulas: “Magic Sixes” is not established as Cundill's invention, and the 50%-of-NAV, holding-period, concentration, and sale observations are dated/product-bounded.
• The synthesis preserves Peter, team, joint-product, transition, issuer, and successor attribution. Current Mackenzie practices and returns are excluded from Peter's personal record.
• All book-derived ideas were paraphrased. No private journal, inaccessible worksheet field, unrecovered letter, or long copyrighted passage was reconstructed.