Jason Jiang's  Instablog

Jason Jiang
Send Message
  • Buffett's Investment Style In His Early Years 2 comments
    Apr 19, 2013 2:07 PM

    Like those who gaze into the sky and confuse galaxies with faint stars, readers of Warren Buffet often stare at his later years, his brightest years. Consequently, what they learn either leads to the realization that Buffett's success is not repeatable or the frustration the methods don't work as desired. It is necessarily so.

    Handicaps and Advantages of Berkshire in these days:

    Both handicaps and advantages come with increase in size, and those in turn shape one's investment style.

    Buffet's handicaps are straight forward: /1/Berkshire has to report its open market operations to SEC on regular basis. /2/ it takes typically months to acquire or unload enough stocks of even multi-billion dollar businesses. These two reduce the universe to effectively a couple of hundreds of public companies. /3/ all the capital for market securities come from insurance floats, which requires extra scrutiny from Warren over liquidity, volatility and risk. Those with great businesses but uncertain future are avoided (IT stocks for instance).

    The advantages were cultivated, somewhat unique to Berkshire: /1/ Insurance floats are effectively free thanks to the eccentric way Buffett organized his insurance subsidiaries. (rigid pricing standards, catastrophic insurance, etc) When we are looking for 25% compound growth, Buffett could achieve it by investing in projects that earns 12%. /2/ More than half of the company consists of privately owned subsidiaries of all kinds. They were usually purchased at a price cheaper than PE companies could find, thanks to the promises he has kept throughout the years. (by the way, it is a handicap too. Washington Post has stagnated years ago for example, but he kept it even after K. Graham's death) /3/ His reputation brings up good deals. Rating agency gave him AAA, GS issued 9% bonds and attractive warrant features, and investment in Solomon Brothers can be saved as long as he is willing to be interim CEO. /4/ business connection, access to proprietary information, I could go on and on but they are common among large investing houses.

    Comparing investment results:


    Equity weighted (%)

    S&P (%)































    Above is a table of investment return if Berkshire's market securities were marked to market, minor holdings were excluded from calculation due to lack of information. The table was produced by the author of <Warren Buffett's portfolio> in his book. In the decade between 1988 and 1997, Buffett averaged 28% while S&P returned 18%. That was remarkable compared with other similar sized investment companies, but it was not nearly as good as his results in BPL years. We will see it later.

    I want to take a moment to look at the individual stocks Berkshire held over the period. To save space, only 1988 and 1997's portfolio are shown below. What is immediately clear is that Buffett didn't sell any of the stocks held in 1988. Capital Cities was acquired by Walter Disney Company, Freddie Mac preferred was converted into equity stake, and GEICO became wholly owned by Berkshire.

    From what I know, Buffett had either some involvement with these companies or have intimate knowledge about the management's quality. Let's go over each of them. Washington Post Company was acquired when the stock declined due to general lack of confidence in the ability of K.Graham, the widow who inherited the CEO position. Buffett essentially assumed the role of personal tutor for K. If he hadn't done so, things would still work out OK because newspaper business is less competitive and K was intrinsically competent. We do not know because it didn't happen. With GEICO, the history goes way back to the 50s. Buffett had 30 years to get to know the business. He intervened on multiple occasions, e.g. he helped to appoint Lou Simpson as CIO, who became indispensable. With Coke, Buffett was initially a fan of Pepsi, and was later converted to drink Cherry Coke after Don, the CEO of Coca-Cola Company, presented the new product to him. We know Buffett is very close to the management from this alone. But that is not it. In 1999, Herbert Allen and Buffett "forced" the then CEO Ivester to retire, at the age of 52, that is something I want to try once just for the hell of it. With the Cap City ABC, we know it was Buffett who provided the capital for the merger with ABC. I don't know any involvement Buffett had with Freddie Mac.

    After learning these, do retail investors still think they can copy the buy and hold strategy that is so advocated in most of the books and articles about Buffett? Those who do may be flattering themselves.

    Even with the intimate knowledge and the occasional intervention, Buffett's 88-97 result was not sufficiently good. If he was still managing investment company, the return limited partners get would have been 21%, slightly better than the general market's 18%. With that track record he would not have been able to attract many investors working from Omaha. But let's not fool ourselves into thinking Buffett is no good. He had free leverage! With minimum risk taken, he was making a killing with 28% return on capital.

    Major holdings in 1988

    Capital Cities/ABC, Inc.

    The Coca-Cola Company

    Freddie Mac Preferred

    GEICO Corporation (privatized later by Berkshire)

    The Washington Post Company


    Major holdings in 1997

    American Express Company

    The Coca-Cola Company

    The Walt Disney Company ( ABC turned)

    Freddie Mac

    The Gillette Company

    The Washington Post Company

    Wells Fargo & Company

    Buffett partnership

    Before fees

    Dow 30








































    The table above was extracted from annual letters to partners during Buffett's investment partnership years. Over the 13 year period, he averaged 28.8% annual compound rate while the Dow 30 returned 8.3% after taking account of dividend. Limited Partners gained 24% per annum.

    What is more impressive we do not have minutely data: from fall 1950 to spring 1956, Buffett's net worth increased from $12.6k to $174k, with an annual growth of 61%. During the time, he studied under Graham for a year and went on working as a stock broker at his father's back in Omaha. The job was paid on commission and Warren wasn't good at it. We could assume the earning he made in those years were barely enough for living expense, noting he was already married and having three children. In August 1954 he was finally hired by Graham. That job lasted for only a year and a half, before Graham decided to retire. I do not know how much he got paid, presumably more than $7k a year. (Research analyst working for Value Line get paid that sum) Anyway, allowing some error in net salary estimate, he was averaging more than 50% a year.

    Why was his result better in early years, when he had neither experience nor business connection?

    Investment style in early years:

    When Buffett was once asked about his investment in South Korea, he replied he got a handbook of Korea companies and leaped through a couple of thousands pages one Sunday afternoon, bought some stocks just based on that. He said it was like reliving his youth. He added If he was to work with smaller sums again, he is definitely more inclined to the classical Graham style, low PE and perhaps trading under working capital type of investment. Maybe that is all you need to know, read no further.

    Reconstructing Buffett's investment styles in early years involves a fair amount of guess work. In the process, I used BPL Letters to partners and Alice's <The Snowball> extensively, knowing that she sometimes get the basics wrong. (For example, on page 80, when mentioning Graham's control on GEICO, she wrote Graham's idea was to buy stocks trading for less than the value of their assets. Well, Graham only oversold that idea in 1934 version of <security analysis> for a good reason, but reversed to normal in the 1940 and 1951 editions. <The Intelligent Investors> also has little mention of working capital plays)

    From very early on, Buffett already had diverse investment tactics. There were low PE plays, balance sheet bets, special-situations (work outs), and he even invested in growth stocks.

    The followings are some pre-partnership examples.

    PE play: In 1950, he invested in Marshall-Wells. The company was earning $62 while trading at $200 a share, or a little over 3 times earnings. Another one, Western Insurance was more ridiculous. The company was earning $29 a share but was selling for $3. I don't think it worked as expected.

    Balance Sheet bets: He bought Cleveland Worsted Mills in 1951. The company was selling for less of its current asset. He bought a gas station in that year too. Those two investments turned out badly. In 1955, he invested in Union Street Railway, which later paid a special dividend out of its over-abundant cash reserve.

    Special situation: Reading Coal & Iron Company was initially bought in 1951 as a balance sheet bargain, and then he learnt in 1954 that Graham controlled the company and was using the extra cash to buy other companies. He loaded up. Another example is the Rockwood cocoa bean deal. By liquidating its cocoa bean inventory via share exchange and switching to profitable line of business, the company changed investor's expectation. The stock price quadrupled.

    Growth Company: Warren put three quarters of his 1951 net worth in GEICO, which was trading at 8 times PE. At this price, he was making a call on the company's future.

    More examples in partnership years: Commonwealth Trust Co, trading at 5 times earnings, with prospect to merge with a larger bank. Buffett later sold it outright at 8 times earnings and bought into Sanborn Map Co. The second company was sold at $45 per share, while sitting on $65 worth of marketable securities. The earning per share was less than $1. A third investment is in Dempster Mill Manufacturing Company, which was sold at significant discount to net working capital.

    From the examples presented above, we can see that Buffett was very Graham like in his approach to stock selection. Unlike Graham, he visited management and took control when necessary. To make these efforts worthwhile, his bets had to be concentrated, apposite to Graham's over diversification. More important, his investees were tiny.

    Seize Control:

    A certain proportion of investment, regardless of your style, is destined to fail. Over the years, Warren had his fair share of mistakes and near misses. For example, in 1989, he held 4 preferred positions. Of which, USAir and Salomon Brothers, he highly praised, turned out to be troublesome for about a year or two, although he did end up with profit exceeding 15% annually on each of these. His investment in Net jet showed consistent loss for almost a decade. General Re lost money for some years after his purchase. The same happened with Buffalo News. His other mistakes were Dexter Shoes, Dempster Windmill, Berkshire Textile business (which took him 20 years to make the decision to close it) and Hostile Con. He bought stamps (very early on) and silver reserve and booked hardly any profit holding them for a few years. He sold Cap City stock at around 4 dollar and bought them back few years later at 17. He bought ConocoPhillips in 2008 and sold substantial portion a year later at half of the cost. Cleveland Worsted Mills was a disaster, and the gas station bought that year was a losing game. To name but a few.

    Those mistakes have little cost to him as he bought them with margin of safety in mind. USAir for example, recorded 2.4 billion dollar write down in 1994, erasing the equity. Warren marked his preferred stock down by 75%, and tried to sell it at 50% of book value. No one took it. A year later, the company resumed profitability and the market price went back to full face value. Another example: when he began to purchase Berkshire, it was sold at $12 per share(the price kept going down). The company has $19 dollar net current asset. By depriving the company's cap expense; I estimate the annual free cash flow was about 5-10% of equity and about 20-40% of his initial investment. As he always says, you want to be in a position that you can make a lot of mistakes and still come out fine. For retail investors, sticking with the principle of Margin of Safety grant you that position. But there is another route, seizing control.

    I think the failure with Cleveland Worsted Mills showed him the route. Before his purchase, the company has paid consistent dividend, approximating 7% of his purchase price per year. Then the management decided to cut it off. Warren got so angry that he flew to attend the annual meeting but changed nothing. Being 5 minutes late, the meeting had been adjourned. Although the return on this stock is no longer mentioned, experienced investors know with that kind of management and sudden change in dividend policy, the stock price could only go down. (My experience has been about a half of the net working capital type of investments don't work or take a long time to work due to management incompetence or downright evilness. Value investors don't usually have the temperament to get into proxy fight. Graham, after his Pacific Railway fight, had chosen to become more diversified and ignored the ones that don't work. Buffett, after the Dempster Windmill case, valued management competence and integrity more dearly.)

    In his annual letter to partners, Buffett explained why he gets into control situations: They started as normal investment in undervalued issues. If the stock prices do not rise, he simply keeps buying until he is elected to the board. Then he will persuade the board to distribute excess working capitals. If he fails at that, chances (when he was young) are he would try to take control of the company. That was the case with Sanborn Map, Dempster Windmill, as well as Berkshire Hathaway. In other words, all of the great battles we hear about started as failures. If he had less capital, or he invested in larger companies or the size of his partnership did not grow exponentially, some of the deals would not have happened.

    That type of hostile takeover deals reduced over time, but he had another mean to profit from taking control. By buying companies at low multiple, and squeezing capital expenditure, even lousy companies could produce some free cash flow, which is invested in more companies of that kind( Much like Russian dolls). Berkshire is the best example. The idea works with good company too; blue chip stamps and See's Candy are early examples.

    Avoid Competition:

    The best way to get good deal is to avoid competition, and the best way to avoid competition is to buy into obscure names. In <The Snowball>, there are phrases like "daily phone call to brokers". Why did he need to call brokers so often when managing $300k? Look at these numbers: GEICO had a market value of $2m; Commonwealth Trust can be estimated at $2-4m and averaged 2 trades per month; Sanborn map was trading at $4.8m level; Dempster Windmill's average total market value was only $1.7m and the market was effective non-existent, it took Warren 5 years to acquire 90% of it. (To comprehend these numbers in today's currency, a Value Line Research Analyst got paid about 7k a year in 1952, a similar job pays a salary of 65k now.)

    Nearly all the names mentioned here were associated with control situations. They are necessarily small. So we cannot conclude from the above numbers that Buffett bought into obscure names. But in the letters he expressed interest in getting into control situations and did explain control situations start as general. We may speculate, with one control situation, there were several other general investments as potential candidates, which are by default small and illiquid.

    There is further evidence. On page 111 in <The Snowball>, I quote "one of his favorite sources was the Pink Sheet". "Another was the National Quotation book, which came out only every six months and described stocks of companies so minuscule that they never even made it into the Pink Sheet".

    Another is in his choice to invest his own money separate from BPL in early years. He teamed up with his college roommate to buy stock certificates from farmers, who were victim of a fraud many years ago but failed to realize the fraudulent had turned into a legitimate company. We could only speculate the reason he wanted to separate the money was that he would take on greater risks in even smaller issues that no trade existed at all.

    Now when I think about it, while Wall Street likes mega deals, Buffett was always buying companies that are much smaller than his size. He avoids competition whenever possible.

    Reliance on numbers:

    On page 111, Alice wrote "Visiting management was part of Warren's way of doing business". I find it increasingly true as his fund gets bigger, or in other words- not so true in early days. During his days as a broker, there is only one mention of visiting management. That was the Cleveland fiasco. I quote "Warren was so mad that he decided to spend some money to find out what was wrong… flew all the way to Cleveland". When Graham retired in 1956, Buffett packed up and headed back, he "did this zigzag across the country,… just thought it was a great time to hit these companies,…..visited Greif Bros. Cooperage…a company he first discovered in 1951 by flipping through the Moody's… He and his farther had each bought two hundred shares." Later in 1964, he showed up at American Express's president's door unannounced a few months after he started investing in the company to lobby him. In all three cases, he bought stocks first and visited later. American Express, which accounted for 1/3 of his portfolio at the time, surely warrants a visit before pouring in money. He didn't. It appears Warren only visit management when they are close and the visit carries low opportunity cost. It was not part of his way of doing business. Rather, it looks like a way to entertain.

    Buffett's hate for travel was worldly known. His first jet, although bought at huge discount, was named "Indefensible". In an interview aired on Fox TV, he showed a Macdonald's membership card from his wallet. He said Bill Gates has one and could eat for free all around the world, while his card is only accepted in Omaha, which is effectively the same as Bill's. In his advertisement in annual shareholder letters, he invites potential investees to fax him financials and he could usually give a preliminary reply within five minutes. Meeting the management still comes to the second after seeing the right numbers.


    Through reading BPL's letters to partners, I believe by then Buffett had already possessed the key qualities he so often stress in later years, although he had not yet found folk-song-like parable to express them. These qualities (or practices) include concentration in portfolio management, endurance for occasional downtimes, extreme modesty, and the most important ones --focus, rationality, staying in circle of competence, and Margin of Safety. I shall stop here.


    No conclusion is better than Buffett's reply to the South Korea question. He leaped through manuals of small companies; some of them had names that he could not pronounce, and bought some that were just ridiculously cheap. That is pretty much his youth. I told you not to read any futher.

    Disclosure: I have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

Back To Jason Jiang's Instablog HomePage »

Instablogs are blogs which are instantly set up and networked within the Seeking Alpha community. Instablog posts are not selected, edited or screened by Seeking Alpha editors, in contrast to contributors' articles.

Comments (2)
Track new comments
  • Talgat Akhmetov
    , contributor
    Comments (34) | Send Message
    Informative analysis. Thanks!
    21 Oct 2014, 01:29 AM Reply Like
  • Jason Jiang
    , contributor
    Comments (7) | Send Message
    Author’s reply » You are welcome.
    22 Oct 2014, 08:20 AM Reply Like
Full index of posts »
Latest Followers
Most Commented
Instablogs are Seeking Alpha's free blogging platform customized for finance, with instant set up and exposure to millions of readers interested in the financial markets. Publish your own instablog in minutes.