When the tide went out: Buffett's 2007 guide to great, good, and gruesome businesses

Published: February 28, 2008 — Berkshire Hathaway 2007 Chairman's Letter

On February 28, 2008, Warren Buffett sent his annual letter to Berkshire Hathaway shareholders. The document is dated, by convention, to the prior fiscal year — 2007 — but it landed in readers' hands as the first large American financial institutions were disclosing losses that would reshape the industry. Bear Stearns had five weeks left before its collapse. Lehman Brothers had seven months.

Buffett had been watching the credit machinery seize up for over a year. His diagnosis arrived in a sentence that has since become one of the most quoted lines in the language of finance:

"You only learn who has been swimming naked when the tide goes out — and what we are witnessing at some of our largest financial institutions is an ugly sight." 1

The exposure, he argued, was a direct consequence of a belief system that had run unchecked for years. Lenders had come to assume that house prices would rise indefinitely, which made borrower income and cash equity irrelevant as underwriting criteria — collateral appreciation alone would make every loan good. 1 He quoted Wells Fargo chief executive John Stumpf with visible exasperation: "It is interesting that the industry has invented new ways to lose money when the old ways seemed to work just fine." 1

But the 2007 letter is not primarily a crisis diary. The "swimming naked" passage occupies a few pages at the opening; the bulk of the letter does something more enduring. In the same document that recorded the financial system's worst excess, Buffett set out a taxonomy of business quality that continues to frame how serious analysts think about capital allocation today.

www.berkshirehathaway.com

Berkshire Hathaway 2007 Chairman's Letter

Warren Buffett's 21-page letter to shareholders, published February 28, 2008. Written as the subprime crisis unfolded, it covers the Great/Good/Gruesome business taxonomy, pension math criticism, the Dexter Shoe disaster, trade deficit analysis, and Berkshire's operating results for 2007.

リンクプレビューを読み込んでいます…

Great, good, and gruesome

Buffett opens the taxonomy with a provocation: "It's better to have a part interest in the Hope Diamond than to own all of a rhinestone." 1 He then makes the categories concrete with three businesses Berkshire actually owned.

See's Candies is his exhibit for a Great business. Berkshire acquired it in 1972 for $25 million, at a time when it generated $30 million in sales and under $5 million in pre-tax earnings. 1 By the end of 2007, the same business was producing $383 million in sales and $82 million in pre-tax profits — and had done so while requiring the reinvestment of only $32 million in additional capital over the entire 35-year period. 1 Total cumulative pre-tax earnings since purchase: $1.35 billion. The business paid for itself more than fifty times over while barely consuming any capital to do it.

FlightSafety, the commercial aviation training company Berkshire acquired in 1996, illustrates a Good business. Pre-tax earnings grew from $111 million in 1996 to $270 million by 2007 — a $159 million improvement. 1 The catch: producing that $159 million earnings gain required $509 million of incremental capital, mostly in flight simulators, which become obsolete and require replacement on a regular cycle. Growth here is real, but capital-intensive. You keep reinvesting to maintain your position.

Airlines represent the Gruesome category. Buffett had direct experience here: he bought U.S. Air preferred stock in 1989, watched it deteriorate, then sold out in 1998 — fortunately, at a profit. 1 The company then went bankrupt twice in the following decade. The structural problem with airlines is that they consume enormous capital continuously — planes, maintenance, gates, fuel infrastructure — while competing in a commodity market where no carrier reliably earns its cost of capital over a full cycle.

Buffett translates the distinction into a savings-account analogy that makes the underlying arithmetic plain. A Great business is an account paying an extraordinarily high interest rate that rises over time. A Good business pays an attractive rate on deposits you make, but requires you to keep making them. A Gruesome business pays an inadequate rate and demands you keep depositing money at that inadequate rate — perpetually.

What makes a moat real

The deeper argument behind the taxonomy is about durability. A business can earn high returns on invested capital for a period without those returns being defensible. Buffett's definition of a genuinely Great business requires an enduring moat — protection against the competitive assault that capitalism guarantees will come:

"A truly great business must have an enduring 'moat' that protects excellent returns on invested capital. The dynamics of capitalism guarantee that competitors will repeatedly assault any business 'castle' that is earning high returns." 1

Two types of moats pass that test in his framework: being the durable low-cost producer (GEICO, Costco), or owning a powerful worldwide brand that creates consumer preference rather than price sensitivity (Coca-Cola, Gillette, American Express). 1

Two types explicitly do not. First, businesses whose continued success depends on a superstar individual: "A medical partnership led by your area's premier brain surgeon may enjoy outsized earnings, but the partnership's moat will go when the surgeon goes." 1 Second, businesses in industries prone to rapid and continuous change — industries where the moat must be perpetually rebuilt:

"Though capitalism's 'creative destruction' is highly beneficial for society, it precludes investment certainty. A moat that must be continuously rebuilt will eventually be no moat at all." 1

The most memorable illustration of this comes in a line about the airline industry's structural economics. Buffett writes that if a farsighted capitalist had been present at Kitty Hawk in 1903 when Orville Wright made his first flight, "he would have done his successors a huge favor by shooting Orville down." 1 The aviation industry that emerged has consumed more capital than it has ever returned to investors in aggregate — a textbook Gruesome outcome, vast in scale and persistent in duration.

www.berkshirehathaway.com

Berkshire Hathaway 2007 Chairman's Letter — Great, Good, and Gruesome taxonomy

Pages 5–8 of the letter contain Buffett's full three-tier business taxonomy, including the See's Candy financials, the FlightSafety capital arithmetic, the U.S. Air anecdote, and the savings-account analogy that closes the section.

リンクプレビューを読み込んでいます…

The helpers problem

The 2007 letter also contains Buffett's sharpest treatment of what he calls the "helpers" — the financial intermediaries who sit between investors and returns. The context is pension fund accounting, but the argument applies broadly.

For the 363 S&P 500 companies that carried defined-benefit pension plans in 2006, the average assumed investment return was 8%. 1 Buffett walks through the arithmetic. Pension funds at that time held roughly 28% of their assets in bonds and cash, which could return at most 5%. That meant the remaining 72% — in equities and alternatives — would need to earn approximately 9.2% after all fees to hit the 8% blended target. 1

For context: over the entire twentieth century, the Dow Jones Industrial Average advanced from 66 to 11,497 — a compounded annual gain of 5.3%. 1 For the Dow to deliver an equivalent 5.3% gain over the twenty-first century, it would need to close December 31, 2099 at approximately 2,000,000. An adviser promising investors 10% annual equity returns — 2% dividends plus 8% price appreciation — is implicitly forecasting a Dow near 24,000,000 by 2100. 1

The helpers who encourage these projections, Buffett observes, have a structural incentive to do so: higher assumed returns allow corporate plan sponsors to report higher current earnings, which benefits the executives who select the assumptions. The chickens, as he notes, will not come home to roost until well after those executives have retired. The investors who bear the ultimate cost are the pension beneficiaries, and the helpful advisers will be long gone.

His prescription is blunt: "Beware the glib helper who fills your head with fantasies while he fills his pockets with fees." 1 The mathematically correct alternative — a portfolio of low-cost index funds, which he calls the "know-nothing" investor strategy — will, as a group, beat the professionally managed alternative after costs. This is not a controversial claim in academic finance; it is simply not in the helpers' commercial interest to say so.

www.berkshirehathaway.com

Berkshire Hathaway 2007 Chairman's Letter — pension assumptions and the helpers critique

Pages 17–19 of the letter contain Buffett's full analysis of pension return assumptions, the Dow century math, and the argument that fee-extracting intermediaries must, as a group, deliver below-average net results.

リンクプレビューを読み込んでいます…

What this means for investors reading the letter today

The "swimming naked" metaphor and the Great/Good/Gruesome taxonomy are not separate ideas. They describe the same phenomenon from opposite ends: one shows what happens to Gruesome businesses when the environment turns hostile, the other provides the prior diagnostic that would have flagged them.

A business that must continuously reinvest to defend its position — one whose moat requires perpetual rebuilding — has a hidden leverage embedded in its capital structure. In good times, when credit is cheap and prices are rising, that leverage is invisible. The returns look adequate. The losses from reinvestment are buried in capital expenditures rather than disclosed as operating losses. When the tide recedes, those hidden obligations surface at once, which is precisely what Buffett was watching happen to financial institutions in early 2008.

Three questions, derived directly from the taxonomy, are worth applying to any business under consideration:

How much incremental capital has the business consumed over the past ten years to sustain its earnings growth? See's Candy required $32 million over 35 years to grow from $5 million to $82 million in pre-tax earnings. FlightSafety required $509 million to grow earnings by $159 million. The ratio matters more than the absolute number.
Does the competitive position require continuous spending to maintain, or is it self-reinforcing? A brand that grows stronger as more people use it (American Express, Coca-Cola) is qualitatively different from a technology or infrastructure position that depreciates and requires replacement. The former compounds without reinvestment; the latter demands it.
What happens to earnings if the business cannot borrow or invest for two years? A Great business — one whose moat is genuine and whose earnings are largely free cash flow — can absorb that scenario without structural impairment. A Gruesome business cannot. The tide test is simply this: what is left when the water pulls back?

Buffett acknowledged in the same letter that he is not immune to getting the answer wrong. Dexter Shoe, which he bought in 1993 for $433 million paid entirely in Berkshire stock — roughly 25,203 Class A shares — was, by his own assessment, worthless within a few years, as its competitive position dissolved under import competition. 1 Because he paid with stock rather than cash, the damage was compounded: those shares, at Berkshire's 2007 valuation, were worth approximately $3.5 billion. 1 He called it "the worst deal that I've made" and noted that no consultant, board member, or investment banker pushed him into it — it was, in his phrase, an unforced error. 1

That admission belongs to the same intellectual tradition as the taxonomy itself. The taxonomy tells you what to look for. The Dexter Shoe story tells you how expensive it is to find out you were wrong — and why the diagnosis is worth running before the purchase, not after the tide has already retreated.

Cover image: AI-generated illustration

参考ソース

1Berkshire Hathaway 2007 Chairman's Letter