Part IThe Story
A Hangover and a Paperback
One evening in December 1973, a thirty-five-year-old Vancouver money manager named Peter Cundill got on a plane in Toronto to fly home for Christmas. By the account his biographer later assembled from his diaries, he was badly hungover and in a sour mood. He had spent a decade in the investment business. He had made some excellent calls, but he could not explain them in a way he could repeat. He had tried forecasting models, market timing and a home-made formula full of weighted variables, and none of it had held together.
A colleague had pressed a book into his hands for the flight: Supermoney, by George Goodman, the financial writer who published as "Adam Smith." Its third chapter was about Benjamin Graham and Warren Buffett, and about Graham's idea of a margin of safety. That night Cundill wrote in the journal he had kept daily since 1963 that he now knew what he wanted to do for the rest of his life. He wrote it in capital letters.
— Peter Cundill, journal entry, December 1973, quoted in There's Always Something to Do, 2011It struck me like a thunderbolt – there before me in plain terms was the method, the solid theoretical back-up to selecting investments based on the principle of realizable underlying value.
What follows is the story of how he put that method to work. For the next thirty-five years he ran a small, failing Canadian mutual fund on Graham's principles. He went looking for bargains in Stockholm, Tokyo, Paris, Panama City and Quito. He compounded his investors' money at a rate few fund managers anywhere matched over so long a period. He wrote nearly all of it down as it happened. The journals, more than 200 handwritten volumes by the time he stopped, were the basis of two books by Christopher Risso-Gill, a former director of the Cundill Value Fund: There's Always Something to Do (2011) and the full biography Routines and Orgies (2014). They are the main reason so much is known about how he thought.
By the Numbers
The Cundill Value Fund
~$7MAssets in the fund when Cundill took over its management at the end of 1974
~$20BAssets managed by Peter Cundill & Associates by the time of the Mackenzie buyout
15.2%Annual compound return over roughly 33 years to 2007, per Prem Watsa
13.2%Annual compound return over the fund's 35 years, including 2008, per the Globe and Mail
1990The fund's first losing year, after fifteen straight gains
45Years he kept a daily journal, 1963 to 2007
US$75,000Top award of the Cundill history prize he founded at McGill in 2008
The Camphor King's Grandson
Francis Peter Cundill was born in Montreal on 29 October 1938, the son of Frank Cundill and Ruth Grierson. The family had once been very rich. His great-grandfather had built a successful import business in Montreal. His grandfather moved to New York, made a fortune in camphor and lost it when synthetic camphor undercut the trade, going bankrupt in 1927. Frank, raised to expect a life of leisure, went west to work as a cowboy and rodeo rider before he came back to Montreal and a job as a trader. In 1937 he bought a seat on the Montreal Stock Exchange.
Frank left for the war when Peter was a toddler and did not come home for good until 1946. In the meantime mother and son moved between relatives' houses. Even after the war the family had noticeably less money than the people around them. Risso-Gill presents this as one source of Cundill's lifelong carefulness with capital. His father had lived through the 1929 crash and never forgot it, and his grandfather had shown what happens when someone keeps playing a losing game.
At Lower Canada College, a Montreal private school, Cundill played football and basketball. He took a Bachelor of Commerce at McGill University in 1960. He wanted to go into investments, probably following the example of his uncle Pete Scott, a partner and later chairman of the brokerage Wood Gundy. His father told him the business was a form of gambling, and that a gambler ought to have a profession to fall back on. Cundill took the advice. He served his articles at Price Waterhouse and qualified as a chartered accountant in 1963. He never practised, but accounting became his main working tool. Much of what later made him unusual came from his ability to read a balance sheet more carefully than the people selling to him. He started the journal the same year.
Greenshields, Yorkshire and a Copper Mine
His first job was in the special projects group at Greenshields, a Montreal investment dealer, working for its young general manager, Frank Trebell. When Trebell moved to the Yorkshire Trust group in Vancouver, Cundill went with him. At Yorkshire he became one of the early holders of the new Chartered Financial Analyst designation, qualifying in 1968.
The journal from these years shows him feeling his way toward a method. He noted that investors tended to follow trends and fashion rather than look for value, and that this should create openings for a professional. He decided that forecasting a company's earnings was useful for spotting obstacles but should never be the main basis for buying a stock. He also concluded that most research-department models were close to worthless and that calling the direction of the market was a mug's game.
His first big stock purchase already had the shape of his later style. Bethlehem Copper, a British Columbia miner, had fallen sharply with no bad news to explain it. When Cundill worked through the accounts, he found that the share price roughly equalled the cash on the balance sheet. The company had no debt and owned a profitable mine with long-term sales contracts. He built a position for Yorkshire clients at about $4.50 a share. Within four months the stock had doubled, and within six he was selling at $13. He had found what Graham called a net-net, a stock priced below its working capital after all liabilities, before he knew the term.
In March 1969 he made his first trip to Japan. A friend from Nippon Steel showed him Tokyo, and the journal records his conclusion that the country was a major economic power in the making. His later investments there started with that trip.
Credit Foncier and a Lesson in Character
The next discovery came from reading a shareholder list. Credit Foncier Franco-Canadien, a mortgage lender founded by the French bank Paribas, had bought shares in the Yorkshire. Cundill read its annual report closely. Over decades of prairie lending it had foreclosed on farms in the Depression, kept many of the farmers on as tenants and kept the mineral rights whenever it sold land. It also held commercial property in eastern Canadian cities. Its accounting was so conservative that the real estate sat on the books at cost, and the mineral rights were barely valued at all.
He estimated its liquidation value at more than $150 a share. The stock traded at about $43. He began buying and brought in a group of institutional investors, and together they soon held more than 2% of the company. Around then he wrote in his journal that most people get one opportunity that tests their knowledge, nerve and judgement, and that Credit Foncier might be his.
His relationship with Trebell was falling apart at the same time. Cundill thought Yorkshire's expansion was badly planned and suspected he was not being told everything. In September 1971 his friend Warren Goldring persuaded him to open a Vancouver office for AGF Management. The move was well timed. Trebell was later investigated by securities regulators in British Columbia and Ontario, including over back-dated pricing in a group of mutual funds, and was eventually jailed. Cundill helped the investigators and testified at the trial. In his journal he wrote that one character flaw, a capacity to wipe inconvenient facts from view, had destroyed an otherwise admirable man. He drew a lesson about housekeeping, detail and self-delusion that he repeated for the rest of his career.
He also kept a small private portfolio, which he ran on his developing value rules. In 1973 it rose 35.2%, by his own later account to unitholders, against falls of 26% in the Dow and 20% in the Toronto industrial index on his figures. The portfolio never held more than ten stocks, and at one point half of it was in two of them: Bethlehem Copper and Credit Foncier.
Buying a Broken Fund
The Trebell scandal created Cundill's opening. The All Canadian group of funds, damaged by the back-pricing affair, was being hit by redemptions, and its management contracts were for sale cheaply. One of them was the All Canadian Venture Fund. It had launched in 1967 at $4.30 a unit and climbed to $6 in the late-1960s boom on a diet of technology stocks. The fund then switched into natural resources just in time for that sector to sour as well. By the end of 1974 the units were worth just over $2 and the fund held just over $7 million.
Cundill had sold his Yorkshire shares for $100,000, a quarter of what they had been valued at a year earlier. With Gowan Guest, a Vancouver lawyer he knew from Progressive Conservative politics, he used the money to buy the fund's management contract. It was, in effect, a value purchase of a business nobody wanted.
He wrote to the unitholders proposing yet another change of strategy, a return to value on the model of Graham and Buffett. The letter set out a list of buying tests. The share price should be below book value and preferably below net working capital less long-term debt. It should be less than half its former high. The price-earnings ratio should be under ten. The company should be profitable and pay a dividend, and its debt should leave room to borrow more. He wrote that nobody should become a slave to the criteria, but that they were the place to start. No unitholders replied. He appointed a new board of friends and allies, renamed the fund the Cundill Value Fund and wrote the Graham-based criteria into the minutes of its first meeting.
His first big success came from the sector the market most disliked. By early 1976 advertising agencies were in a slump. J. Walter Thompson had gone public in 1972 at over $20 a share and now traded at $4, with no institutional friends and an awkward Watergate connection through H. R. Haldeman, who had once run its Los Angeles office. Cundill found hard book value of $18 a share, before counting freehold buildings in Paris and Tokyo and a long lease in London's Berkeley Square. He bought up to his limit of 10% of the fund at an average of just over $8. When the company's president asked what he knew, Cundill said only that the stock was cheap. A few months later Sam Belzberg's First City Financial disclosed a stake, and Cundill eventually sold at more than $20.
The fund rose 32% in 1975, 32% in 1976 and 21% in 1977, a year when the Dow fell sharply. Redemptions from weary old investors roughly cancelled out the gains, so the fund barely grew. The fix came from Yorkshire's founding shareholders, the Norton family of Huddersfield. Their private holding company had no way for family members to sell without triggering a large tax bill. Cundill swapped Value Fund units for their shares, sold the underlying assets and more than doubled the size of the fund in one transaction.
Tiffany and the Problem of Selling
The late 1970s were full of bargains, and the fund's records show how Cundill's method developed. He bought 200,000 shares of the American Investment Company, a personal-loan firm trading at $3 against book value above $12. He built a stake of more than 5%, joined its board after some hesitation, and was eventually bought out when Leucadia acquired the company at $13 a share, two years after his first purchase. In his journal he noted that very few people did their homework properly, so he always checked for himself.
The Tiffany & Co. trade taught him the hardest lesson of all. The Fifth Avenue jeweller traded below book value of $10.50. Its famous yellow diamond was carried at $1, and its flagship building was on the books at its 1940 value. Cundill bought 3% at about $8 a share and called on Walter Hoving, the controlling shareholder, who told him he would never sell. Cundill sold his stake within a year at $19. Six months later Avon made an all-share offer worth about $50 a share, and Hoving accepted.
The board spent some time debating what the right selling rule should be, and settled on a compromise. When a position doubled, the fund would sell half, which in effect brought the cost of the remainder down to zero, and the manager would decide what to do with the rest. The rule stayed in place for decades.
By the end of 1979 nearly 70% of the fund was in American stocks, because that was where the numbers pointed. The American holdings were concentrated in about twenty companies. After five years the fund was one of only eight Canadian funds, in a universe of nearly 800 North American funds, with a compound growth rate above 30%.
The Worst Market of the Year
Cundill had noted as early as 1964 that Sweden punched above its weight industrially. In November 1977 he went to see for himself. He toured Volvo's plant in Gothenburg, joined the Swedish Shareholders' Association and had long conversations with demoralised local brokers. A month later he had satisfied himself that Volvo was a genuine net-net, and he started buying. He later added SKF, and both roughly doubled.
Sweden became a habit. For years, he later told a conference audience, he would travel each November to whichever stock market had done worst over the previous eleven months. His view of foreign accounting was unusual too. When a set of accounts was hard to read, he treated the difficulty as a barrier that kept other investors out. He then spent years building networks of local professionals who understood Graham's approach and could spot a stock that fitted his criteria.
In 1983 he and his wife, Joanie, tried three months in Paris. He decided it was a financial backwater compared with London, and in 1984 they moved there. London suited him as a base. Concorde had turned New York into a day trip, and Cundill travelled constantly, well over 100,000 miles a year by Risso-Gill's account. The fund's 1984 annual report adopted the slogan "Buying a Dollar for Forty Cents." It also reported a compound return of 26% a year over its first ten years.
He was also becoming a serious endurance athlete. He ran his first marathon in 1976, Buffalo to Niagara Falls, on six weeks of training, and finished in three hours thirteen minutes. In 1983, shortly before his forty-fifth birthday, he broke three hours in Portland, Oregon, with a time of 2:59. He ran more than twenty marathons in all. He believed the stamina and the investing were connected.
Cash Before the Crash
Cleveland-Cliffs, North America's biggest producer of iron ore pellets, was exactly the kind of dull company the mid-1980s market ignored. Cundill's friend Alan Kahn, son of the veteran Graham disciple Irving Kahn, pointed out that it owned a power plant in Michigan carried on the books at almost nothing. Cundill started buying at $15. The stock kept falling, and at one point he and a fellow value investor were splitting blocks of stock bought at $6. His average cost ended up around $9.75. By July 1987 the stock had reached $20, the doubling rule kicked in, and he began to sell. After the October crash he bought all of it back and more at about $10. The overall compound return on the holding was above 30% a year.
The crash itself found the fund with more than 40% of its assets in short-term money-market instruments. At the board meeting the following month, Cundill was careful about how he described this. He had expected a crash but could not have known when. The cash had built up because holdings kept reaching prices well above book value and were sold under the fund's rules. He was in Hong Kong on Black Monday, where the rout began, and spent the following days buying.
The fund ended 1987 up 13%, an unusually good result for that year. Even so, its capital fell by $15 million, because investors cashed out to cover losses elsewhere. The next year a BC Business profile called him a vulture, which he enjoyed. It quoted his advice to individual investors: be patient and don't be greedy.
After the crash he found his next big opening in Germany. German companies routinely overstated their liabilities, so blue chips such as Siemens, BASF, Volkswagen and Deutsche Bank turned out to be net-nets on his measures. At Deutsche Bank, the proprietary share portfolio alone was worth about as much as the whole bank's equity.
Defaulted Nations
Distressed debt fitted Graham's framework because the analysis starts with the balance sheet, as his stock analysis did. The question is whether the assets, sold in an orderly way, would be worth more than the price of the debt. After Drexel Burnham collapsed, Cundill turned to Deltec, a Nassau-based firm that specialised in Latin America. The introduction came through his friend and client John Templeton, whom he regarded as a mentor.
Panama was the clearest case. After the 1989 American invasion removed Manuel Noriega, Panamanian government debt stayed in default. Cundill calculated that the Panamanian assets frozen in American banks roughly equalled the defaulted obligations, so a buyer was getting cash at a discount and the country for free. He held off at just over 50 cents on the dollar because selling was still heavy. In December 1992 the bonds dropped below 35 cents, and he bought at an average of 39. He sold in May 1995 for a gain of 121%.
Ecuador followed the same logic. Costa Rica had just restructured its bank loans into Brady bonds and the debt rallied, and Cundill thought Ecuador, an oil exporter with a budget surplus before debt service, could make a similar deal. He started buying in November 1992 at 20 cents. According to Risso-Gill, repeated trades in Panamanian and Ecuadorean debt between 1992 and 2000 made more than $20 million.
His curiosity also led him into a lost cause. In 1990 he bought a lot of 1838 State of Mississippi bonds at auction for just under £7,000, mostly because they were handsome and historically interesting. Mississippi had repudiated them in the 1840s. In 1992 he joined other bondholders in suing the state. A Hinds County judge ruled that the 1875 constitutional amendment barring repayment was itself unconstitutional, but held that the claim was blocked by a statute of limitations. The appeals went nowhere.
Bruises
In 1990 the fund lost 9%, its first losing year after fifteen years of gains. Weak results of 5% in 1991 and 7% in 1992 followed. Part of the drag was a bet against Japan. From March 1987 until early 1991 the fund held put options on a Japanese index, which Cundill kept renewing at a loss while the Nikkei climbed. He thought the market's valuations were beyond any statistical common sense. He was eventually proved right, but his own president doubted the position and the renewals cost money every quarter. Other holdings had gone wrong too. In a New Year review he admitted that with one company he had been so taken with the cash flow that he ignored the debt, breaking his own rules.
— Peter Cundill, journal, quoted in There's Always Something to Do, 2011Being out on a limb, alone and appearing to be wrong is just part of the territory of value investment.
The journal from these years is unusually candid about self-doubt. It includes a night when he woke in a cold sweat about his ability as a manager, and a copied passage from the opening of Dante's Inferno. His response was to cut the number of holdings by about a third and make each position bigger. He cited his father, a good shot, who warned against firing into a flock without picking one bird. He also began asking harder questions about earnings and management quality alongside asset value. At the end of 1991, he told investors, the fund's stocks traded at about 64% of book value, against 270% for the S&P 500.
In 1993 the fund rose 43%. European holdings, by then over a quarter of the portfolio, nearly doubled in value. Among them was the Telegraph, the London newspaper group, whose stakes in John Fairfax and Southam covered much of its share price. When riots and strikes hit France in 1995, he bought Paribas, the parent of his first great stock, Credit Foncier. By the end of 1996 France was the fund's biggest market.
Mackenzie, Japan and the Worst Investment
By the mid-1990s Peter Cundill & Associates had a problem common to mid-sized fund companies. Fixed costs were high and competing on distribution was expensive. James Morton, a manager Cundill had recruited after an interview for the Financial Times Global Guide to Investing, persuaded him to find a larger partner. In 1998 he formed a strategic alliance with Mackenzie Financial, then Canada's second-largest fund company with about $25 billion in assets. Cundill's firm ranked forty-first, with under half a billion. Mackenzie took over marketing and administration and gave him a fifteen-year contract that left his investment approach alone.
The timing looked bad at first. By the end of 1997 he had put more than 40% of the fund into Japan, where many companies traded below the cash and securities on their balance sheets. He expected the move toward share buybacks and shareholder returns to act as a catalyst. The Nikkei fell further in 1998, and the fund lost 9% in the alliance's first year. The fund then gained 16% in 1999 and 20% in 2000, as the growth-stock boom collapsed and cheap assets came back into favour. Within a Mackenzie multi-manager fund, where new money followed performance, his allocation grew from $200,000 in 1998 to about $1.5 billion by March 2001. By then he had about $75 million of his own money in Cundill products.
He never owned Nortel, the telecom stock Canadian investors loved, and made money instead in the conglomerate Brascan, now Brookfield Asset Management. His worst investment was also in telecoms. In 2000 he began buying Cable & Wireless below its net asset value, attracted by more than £3 billion of cash and securities. Management then spent heavily on loss-making internet businesses. After a restructuring and a £4.4 billion loss in November 2002, he sold the whole position. The fund lost nearly $59 million, and 2002 was an 11% down year. On Christmas Eve he wrote to his research team.
— Peter Cundill, memo to the Cundill Investment Research team, 2002, quoted in There's Always Something to Do, 2011Over the holiday we should all muse as to why we screwed up so badly on Cable & Wireless. There was really no dissenting voice.
The fix was procedural. When the team agreed unanimously on a stock, someone would be assigned to argue against it. The same period produced one of his best-known holdings. In 2000 he bought $122 million of Fairfax Financial, Prem Watsa's insurer, below book value. He halved the position after a rebound, held on through the losses that followed and more than doubled it again when short sellers attacked the stock in 2005 and 2006. Fairfax's credit default swaps paid off heavily in the 2008 crisis.
Fragile X
Around 1998 Cundill noticed a slight tremor in his right arm. By the early 2000s it had spread to both arms and was affecting his balance. Doctors first treated it as Parkinson's disease, without effect. In 2006 he was diagnosed with Fragile X-associated tremor/ataxia syndrome, a degenerative neurological disorder with no effective treatment. That year he sold his remaining 5% of the business to Mackenzie. The purchase of Cundill Investment Research, announced in August 2006, made the team Mackenzie employees and set out a plan for his succession.
The honours came as his health declined. In 2001 Investment Executive named him fund manager of the year, and the Canadian Investment Awards gave him its Analysts' Choice Career Achievement Award. At the ceremony he was introduced as the Indiana Jones of Canadian money managers, a description he enjoyed. He won fund manager of the year again at the 2003 Canadian Investment Awards and became a fellow of the Institute of Chartered Accountants of British Columbia in 2004.
He gave up day-to-day management in April 2009, handing the funds to longtime colleagues Andy Massie and Lawrence Chin, and became chairman emeritus of Mackenzie Cundill. In 2008 he had founded the Cundill International Prize in History at McGill, with a top award of US$75,000. It was billed as the largest history-book prize in the world, and its first winner was Yale historian Stuart Schwartz. He said he saw an analogy between history and finance, because both study the past to understand the present. His personal money, meanwhile, went into value funds run by managers he had trained or admired.
He died in London on 24 January 2011, aged 72. The Globe and Mail obituary put the fund's compound return over three decades at 13.2% a year, against 10.8% for the MSCI world index. The fund had only five losing years, none before 1990. The book of his investment approach came out a few weeks after his death. Its title comes from a line by his friend Irving Kahn: there's always something to do.
Part IIThe Playbook
Cundill left an unusually complete record of his method because he wrote down both his reasoning and his mistakes as they happened. The principles below come from that record. They include his written buying criteria, the fund's sell rules, his notes on travel and networks, and the post-mortems he held on his failures. Most of them are about discipline rather than cleverness, which is how he saw the job.
The principles
- 1.Value the company as if it were being wound up.
- 2.Write the buying tests down before you buy.
- 3.Sell half when a position doubles.
- 4.Let cash accumulate from discipline, not predictions.
- 5.Visit the year's worst-performing market.
- 6.Treat confusing accounts as a moat for you.
- 7.Cultivate local analysts who speak your language.
- 8.Check every figure yourself.
- 9.Expect the painful dip before the payoff.
- 10.Exit when the asset cushion is being spent.
- 11.Appoint a devil's advocate when everyone agrees.
- 12.Keep a daily record of your thinking.
Principle 1
Value the company as if it were being wound up.
Cundill's first question about any stock was what the business would fetch if it were sold off piece by piece. Earnings multiples, dividend yields and growth rates came second. At Bethlehem Copper the answer was the cash in the bank. At Credit Foncier it was farmland, mineral rights and property carried at old costs. At J. Walter Thompson it was book value plus buildings in three capital cities. In each case he paid well under that figure.
This approach set a floor under his downside. If he was right about the liquidation value and the price was far below it, the stock had room to fall a long way before he actually lost money. It also gave him a reason to hold when the market disagreed, because his estimate did not depend on anyone else's mood. His later refinement, in the early 1990s, was to add questions about earnings and management. He never dropped the balance sheet as his starting point.
Tactic: For every holding, write down a conservative break-up value per share, listing each asset at a realistic sale price. If you can't produce the number, you don't yet have a margin of safety.
Principle 2
Write the buying tests down before you buy.
When Cundill took over the All Canadian Venture Fund, he put his criteria in a letter to unitholders and then in the minutes of the first board meeting. The price had to be below book value and preferably below working capital net of debt. It also had to be less than half its old high, under ten times earnings, from a profitable dividend payer, with manageable debt. He warned against becoming a slave to the list, but it was the first filter every idea had to pass.
Writing the rules down in public made them hard to bend quietly. When he did break them, as with a heavily indebted holding around 1989, the journal records it as a failure of discipline, not an innovation. The written criteria turned his instinct into something a team could learn and continue after he left. His successors kept using the net-net worksheets he designed in the 1970s.
Tactic: Before your next purchase, write a one-page statement of the tests a stock must pass, and share it with someone who can hold you to it. Record every exception, with the reason for it.
Principle 3
Sell half when a position doubles.
Tiffany taught Cundill that selling was harder than buying. He sold at $19, pleased with a big gain, six months before Avon's offer worth about $50 a share. The fund's board eventually settled on a mechanical answer. When a holding doubled, half of it would be sold, and the manager kept discretion over the remainder.
The rule made two things simpler. Selling half returned the original investment, so the rest of the position was held at no cost, which made it easier to be patient with it. It also stopped him from having to guess how far a stock would overshoot its value, a judgement he admitted in his journal he had not solved. Cleveland-Cliffs shows how it worked: the rule sold stock near $20 before the 1987 crash, and after the crash he bought back at around $10.
Tactic: Choose a threshold now, such as a double, at which you automatically take out your original capital. Decide in advance, in writing, what would justify holding the rest.
Principle 4
Let cash accumulate from discipline, not predictions.
When the 1987 crash arrived, the fund had more than 40% of its assets in cash and short-term paper. Cundill told his board that this was not a market call. As prices rose, more holdings passed book value and were sold under the rules, and fewer new bargains qualified to replace them. The cash was what the rules left behind.
He thought that trying to time markets was a waste of effort, and he once told a conference audience as much. Value rules produce the same result without a forecast: they sell into rising markets and have little to buy near peaks. Cundill accepted that this usually meant selling too early and buying too early. In exchange he got a portfolio that suffered less in crashes and had cash to spend when they came.
Tactic: Track what share of your portfolio trades above your estimate of fair value. When that share rises, let the sale proceeds sit as cash instead of forcing them into second-rate ideas.
Principle 5
Visit the year's worst-performing market.
His November trips to the year's worst stock market started with Sweden in 1977. It was a small, orderly economy with English-language annual reports, and its local brokers were demoralised because no one outside the country cared. Volvo and SKF were the result. Over the next three decades the same habit took him to Japan, Germany, France, Panama, Ecuador and Russia.
Going to the place mattered to him as much as reading the numbers. He wanted to judge the local politics, the culture and the mood himself. If a place felt wrong he would pass, even when the figures were compelling, as he did for years with Russia, where he found no margin of safety. The trips kept him out of whatever consensus had formed at home.
Tactic: Once a year, pick the worst-performing market or sector you can find and study it seriously, including talking to people who work in it. Treat the trip as research, with no obligation to buy.
Principle 6
Treat confusing accounts as a moat for you.
Many foreign investors avoided Japanese, German or Swiss accounts because they were hard to read. Cundill took the opposite view. If a balance sheet was opaque, most competitors would stay away, and anyone willing to do the work had the field to themselves. His accounting training made this possible.
The approach paid off repeatedly. Japanese banks and insurers carried share portfolios at cost. German blue chips overstated their liabilities. Neue Zürcher Zeitung, with illiquid shares and restrictive voting rules, traded below 30% of his conservative liquidation value. In each case, understanding the local rules was what revealed the value.
Tactic: When a company's accounts put you off, spend an extra hour on the notes before you walk away. Rebuild the balance sheet under standards you trust, and see whether the discount survives.
Principle 7
Cultivate local analysts who speak your language.
Cundill spent years building a network of like-minded investors around the world, including Enskilda bankers in Stockholm and analysts at Sumitomo Trust, Nomura and Nikko in Tokyo who knew Graham's books. He also worked with Deltec in Latin America, the Kahns in New York and Don Kennedy in San Francisco. He compared the search to panning for gold: many calls went nowhere, but the kindred spirits he found were worth the trouble.
The network brought him ideas, such as Cleveland-Cliffs through Alan Kahn, and checked his own work, as when Kennedy gave a second opinion on the Michigan power plant. Because they all shared a method, these people understood what he was looking for and would notice it when they saw it.
Tactic: Name three people in markets or industries you don't know well who share your investment approach. Keep in regular contact, and trade ideas with them rather than just asking for tips.
Principle 8
Check every figure yourself.
Early in his career Cundill noted that very few people really did their homework, so he always checked for himself. He read annual reports, 10-Ks and the notes to the accounts himself, and he visited companies. The fund's 1984 annual report said of him that he did not act on tips.
He was also firm about inside information. His journal cites the banker Siegmund Warburg's remark that all you get from inside information is a whiff of bad breath. He added that it was worse than useless, because it clouded the detached judgement that decisions required. Doing his own analysis made him an independent investor.
Tactic: Before acting on any recommendation, rebuild the key numbers from the primary filings yourself. If your figures don't match the pitch, trust your own.
Principle 9
Expect the painful dip before the payoff.
After Cleveland-Cliffs, Cundill wrote that none of the great investments came easily and that there was almost always a major blip before the reward. The stock fell from $15 to $6 while he kept buying. Panama's debt fell from above 50 cents to below 35 before he bought his main position. The Nikkei rose against his put options for years before it collapsed.
He regarded this as a normal part of value investing rather than evidence that he was wrong. Stocks that are cheap because they are unfashionable usually get more unfashionable before they recover. The discipline was to reopen the analysis at each fall, and to keep buying only if the numbers still held.
Tactic: When a holding drops sharply, redo the valuation from scratch without looking at your purchase price. If the value is intact, consider adding; if it isn't, move to the next principle.
Principle 10
Exit when the asset cushion is being spent.
Cable & Wireless is the other side of Principle 9. When he bought it, the company had cash and securities worth more than £3 billion and traded below its net asset value. The cushion then shrank, because management spent the cash on loss-making internet businesses, profit warnings piled up and the accounting became hard to trust. A falling share price was one thing. What finally mattered was that the value itself was disappearing.
When he sold everything after the November 2002 restructuring, the fund took a loss of nearly $59 million. A week later the company disclosed a possible £1.5 billion tax liability, and the shares halved again. Knowing when to hold on and knowing when to leave both depended on asking whether the margin of safety was still there.
Tactic: Separate the price from the value. Sell when the value you calculated is being destroyed by management decisions or hidden liabilities, even at a loss, and don't wait for the price to confirm it.
Principle 11
Appoint a devil's advocate when everyone agrees.
Cundill's own diagnosis of the Cable & Wireless failure was that there had been no dissenting voice. The fix he proposed was that whenever the research group was unanimously in favour of an investment, someone would be assigned to argue against it. At Fairfax, a young analyst took the role willingly and kept pressing his case. Cundill halved the position, and the timing was lucky: soon afterwards the September 2001 attacks hit Fairfax's New York business hard.
He did not let the dissenter decide the matter. He kept Fairfax on the basis of the asset analysis and Watsa's management, and added heavily later. The point was to make sure the case against a stock had been heard in full before the team committed.
Tactic: For any position where your team or your own notes show no doubts, write the strongest bear case you can, or have someone else write it. File it with the investment memo and revisit it when the stock moves.
Principle 12
Keep a daily record of your thinking.
For forty-five years Cundill wrote in a journal every day. It recorded trades, meetings, doubts, marathons and dinners. Because he wrote it at the time, his own hindsight could not rewrite it. It shows what he was thinking before the 1987 crash, when he was agonising over the Nikkei puts, and when he broke his own rules. His New Year reviews of the previous year's winners and losers drew directly on it.
The record also made his method teachable. Risso-Gill could reconstruct Cundill's reasoning because it had been written down, and the Mackenzie team inherited worksheets and memos along with the funds.
Tactic: Keep a short dated entry for every investment decision: what you believe, why, and what would prove you wrong. Once a year, reread the whole year and score your reasoning, not just the results.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIMaxims
- Fashion creates the discount. Cundill's early journal noted that investors chase trends rather than value. Unpopular industries such as advertising in 1976, iron ore in 1985 and Japanese stocks in 1998 were where his biggest bargains came from.
- Forecasts illuminate; they don't decide. He thought earnings projections were useful for spotting obstacles and poor as a basis for buying stocks. He gave market calls even less credit.
- A tip is a whiff of bad breath. He refused inside information, both on principle and because it clouded the detached judgement investing requires.
- Pick one bird. He took his father's hunting advice as a case for concentration. A few high-conviction positions beat a portfolio spread too thin to understand.
- Saints have a past and sinners have a future. He quoted Oscar Wilde to argue that a company's history should never rule it out, because fallen stocks are where value investors do their detective work.
- Blame no one but yourself. He held that losses were never the market's or a broker's fault. Accepting responsibility for them was what allowed him to learn from them.
- Laugh at yourself before hubris sets in. He believed a sense of humour about oneself was the best protection against the idea that success had made one infallible.
- Train the body to steady the mind. He said running gave him the resilience to get through bad years, and he saw a link between the temperament of elite athletes and that of elite investors.
- Read far beyond finance. He called curiosity the engine of civilisation. He read widely outside finance and founded a history prize because he saw history and investing as related disciplines.
In Their Own Words
The best time to buy is when blood is running in the streets.— Peter Cundill
I will go anywhere in the world to find undervalued securities. Geography is not a constraint when seeking value.— Peter Cundill
The four most dangerous words in investing are 'this time it's different.'— Peter Cundill
The most important thing is not being wrong. If you're not wrong, you don't have to be spectacularly right to do well.— Peter Cundill
Time is the friend of the wonderful business and the enemy of the mediocre one.— Peter Cundill
The most important attribute for success in value investing is patience, patience, and more patience. The majority of investors do not possess this characteristic.— Peter Cundill
All we are doing is buying $1 for 40¢ and waiting for the market to recognize that it's worth $1.— Peter Cundill
The secret to successful investing is to figure out the value of something—and then pay a lot less.— Peter Cundill
Price is what you pay, but value is what you get. The two are often very different.— Peter Cundill
Risk comes from not knowing what you're doing. If you understand what you own, volatility becomes opportunity rather than risk.— Peter Cundill
Uncertainty is the friend of the buyer of long-term values.— Peter Cundill
The best opportunities often exist where others fear to tread.— Peter Cundill
Political uncertainty creates some of the best investment opportunities, provided you can separate temporary noise from permanent change.— Peter Cundill
I spend 80% of my time reading and thinking, and 20% acting. Most investors reverse this ratio.— Peter Cundill
You can't make good investment decisions without good information, and you can't get good information without doing the work.— Peter Cundill
The footnotes often tell you more about a company than the main financial statements.— Peter Cundill
The market is a voting machine in the short run, but a weighing machine in the long run.— Peter Cundill
Emotion is the enemy of rational decision-making. The best investors are those who can remain calm when others are panicking.— Peter Cundill
The crowd is usually wrong at extremes. When everyone is buying, you should be selling, and vice versa.— Peter Cundill
Successful investing requires the ability to do nothing for long periods of time.— Peter Cundill
The stock market is designed to transfer money from the impatient to the patient.— Peter Cundill
Discipline is the bridge between investment goals and investment accomplishment.— Peter Cundill
It's not how much you make when you're right that matters, it's how little you lose when you're wrong.— Peter Cundill
Further reading
- Routines and Orgies: The Life of Peter Cundill by Christopher Risso-Gill
