The Intelligent Investor

Graham opens by saying who the book is not for: traders who buy because a price has risen and sell because it has fallen. He calls that rule the opposite of sound business practice anywhere else, and says that in fifty years he had not met one person who made money from it for long.

The introduction carries the warning most useful to anyone watching a new industry grow. Graham recalls that the funds of the 1960s were right that air travel would boom and wrong about what the airline shares would earn: the industry lost money for its shareholders in 1970 despite record traffic. Computers told a milder version of the same story. He draws two morals, and the first is the one to keep: “Obvious prospects for physical growth in a business do not translate into obvious profits for investors.”

The second moral is that the experts have no reliable way to pick the winning companies inside the winning industries. Graham adds that he never ran money that way himself, and offers little help to those who want to try.

What the book offers instead is a habit of mind. Graham says the investor’s “chief problem—and even his worst enemy—is likely to be himself,” and that he has seen ordinary people with the right temperament keep more money than experts without it. The other habit he wants to teach is asking what a thing costs: he once told readers to buy stocks “as they bought their groceries, not as they bought their perfume.”

The defensive investor splits his money between high-grade bonds and leading stocks, never holding less than a quarter or more than three-quarters in stocks. The default is half and half, put back in balance when the market pushes it out by a few points.

Graham’s reason for so mechanical a rule is candid. The textbook advice to buy more stocks in a slump and fewer in a boom is, in his phrase, “always easy to enunciate and always difficult to follow,” because it runs against the same human nature that produces the booms and slumps. A fixed ratio forces the investor to sell some stock after a rise and buy some after a fall without having to predict either.

The chapter gives four rules for the stock half of a defensive portfolio: hold somewhere between ten and about thirty issues; pick companies that are large, prominent and soundly financed; require a long unbroken dividend record; and cap the price at 25 times average earnings of the past seven years and 20 times the latest year’s.

Graham notes the cost of the last rule: it rules out most growth stocks. He defines a growth stock as one whose per-share earnings have risen well above the general rate and are expected to keep doing so, and says the trouble is never the company but the multiple paid for it.

His example could have been written about any favourite of the last few years. Texas Instruments rose from 5 to 256 in six years while its per-share earnings went from 40 cents to $3.91. Graham’s comment: “Note that the price advanced five times as fast as the profits; this is characteristic of popular common stocks.” Two years later earnings had fallen by nearly half and the price by four-fifths. Even IBM, the leading growth stock of the day, had lost half its price in six months of 1961–62 and nearly as much in 1969–70.

He concludes that growth stocks as a group are too uncertain for the defensive investor, and points him instead to large companies that are out of favour and so priced at modest multiples.

The enterprising investor’s first task is to avoid what looks like opportunity and is not. Chief among these, for Graham, are new stock issues sold during a bull market. In a passage carried over from earlier editions he traces the pattern: early offerings are fairly priced, later ones come from poorer companies at higher prices, and near the top small, unknown firms are floated at prices above those of established mid-sized companies.

The collapse that follows, he says, in many cases takes new issues down by three-quarters or more. His verdict on joining in, even when an issue passes every test: “For every dollar you make in this way you will be lucky if you end up by losing only two.”

Graham defines a bargain as an issue that analysis shows to be worth at least half as much again as its price. He gives two ways to find one: an appraisal of future earnings, and an estimate of what the business would be worth to a private owner, with weight on its working capital.

The plainest bargain is a stock priced below its net working capital alone, after all prior claims, so that the buyer pays nothing for plant, machinery or goodwill. Graham reports a test: one share each of 85 such companies bought at the end of 1957 and held two years gained 75%, against 50% for Standard & Poor’s 425 industrials. None showed a significant loss; seven ended about even and 78 rose.

He is equally specific about why such stocks recover. The reasons he lists are ordinary: a high dividend, retained earnings large relative to the price, bull markets that lift cheap issues most, the slow pull of price back toward value, and the chance that whatever hurt earnings gets fixed. None of them requires the market to admit it was wrong.

Graham accepts well-run mutual funds as a sound choice for the defensive investor, and reports that the ten largest funds of 1961–1970 did about as well as the S&P composite.

His scepticism is reserved for the “performance” funds of the late 1960s. Of nineteen young managers profiled in a 1969 book, he tracks the twelve whose fund results were published: strong in 1966, brilliant in 1967, and all showing losses in 1969, with only one doing slightly better than the index. A record built in one rising market, he warns, may show only that a manager took speculative risks and got away with them for a while.

The chapter opens with two pieces of advice that Graham admits pull against each other: “Don’t take a single year’s earnings seriously,” and, for those who will anyway, watch for traps in the per-share figures.

His example is ALCOA’s 1970 report, which carried four different per-share earnings figures for the year, from $5.20 down to $4.19, depending on special charges and dilution. Taking the December quarter’s headline figure, the stock at 62 looked like less than ten times earnings; taking the quarter after special charges, it looked like 22 times.

He names the usual sources of distortion: special charges kept out of the headline number, tax credits from past losses, dilution from convertibles and warrants, and changes in depreciation method. His remedy is the older practice of averaging earnings over seven to ten years with the special charges left in, since those losses are part of the company’s real history.

Graham takes four companies that sat next to each other on the exchange list at the end of 1970 — ELTRA, Emerson Electric, Emery Air Freight and Emhart — and compares them. “The most striking fact about the four companies is that the current price/earnings ratios vary much more widely than their operating performance or financial condition.” Two sold at about 10 and 12 times average 1968–70 earnings; the other two at 33 and 45.

He grants that the favoured pair had grown faster, but notes the others’ growth was respectable too. The point is the size of the gap, not its direction: the market was paying several times more for a growth record than the records differed.

Here the defensive rules become seven tests: adequate size; a strong financial position; dividends paid for at least the past twenty years; no loss in any of the past ten; growth of at least a third in per-share earnings over ten years; a price no more than 15 times average earnings of the past three years; and a price no more than one and a half times book value.

The last two can trade off against each other. Graham’s rule of thumb is that the earnings multiple times the price-to-book ratio should not exceed 22.5. He expects the tests to exclude most stocks, and says the financial-strength test was the hardest to pass at the time, because many large companies had let their debt run up.

The enterprising investor gets looser screens: current assets at least one and a half times current liabilities, modest debt, no loss in five years, some dividend, rising earnings, and a price under 120% of net tangible assets.

Graham also returns to the sub-working-capital stocks, which the 1970 decline had brought back, some of them well-known brands. He finds it telling that the market could value such names at almost nothing while paying billions above book value for companies with faster growth: “But now, if the ‘market doesn’t like a company,’ not only renowned trade names but land, buildings, machinery, and what you will, can all count for nothing in its scales.”

Graham’s language on warrants is the bluntest in the book: “We consider the recent development of stock-option warrants as a near fraud, an existing menace, and a potential disaster. They have created huge aggregate dollar ‘values’ out of thin air.”

His precedent is American & Foreign Power, whose warrants had a market value above a billion dollars in 1929 while appearing only in a footnote to the balance sheet. By 1932 they were worth $8 million, and in 1952 they were cancelled in a recapitalization even though the company had stayed solvent. The broader lesson is about dilution: claims on future shares are a cost to existing owners even when the accounts barely show them.

Each case shows an extreme. Penn Central shows warning signs ignored by everyone responsible for its securities, with the stock falling from 86½ in 1968 to 5½ in 1970. Ling-Temco-Vought shows fast, unsound empire-building, helped by careless bank lending. NVF shows a small company swallowing one seven times its size on borrowed money and with startling accounting. AAA Enterprises shows a small company floated on the word “franchising” and little else, bankrupt within two years.

Graham’s point on Penn Central is that the simplest tests of security analysis would have exposed its weakness in 1968, when the shares stood at their highest since 1929. The failure was not a lack of information but a lack of attention to it.

The AAA case ends with the book’s most quoted passage on speculative crowds: “The speculative public is incorrigible. In financial terms it cannot count beyond 3. It will buy anything, at any price, if there seems to be some ‘action’ in progress.” Graham then turns the question on the brokerage houses that sponsored such issues.

The first pair is the chapter’s whole argument in small. Real Estate Investment Trust, a New England trust nearly a century old with dividends since 1889, grew steadily and borrowed little. Realty Equities, its neighbour on the exchange list, expanded recklessly. In 1968 Wall Street responded “by paying as little attention as possible to the Trust and a lot to Realty Equities”: Equities ran from 10 to 37¾, while its asset value at the next balance sheet was $3.41 a share.

Graham asks shareholders to judge management on results. He sets out three grounds for asking hard questions: results poor in themselves, poor against similar companies, and a share price that has stayed low for a long time.

On dividends he records a shift. Retaining earnings used to be the mark of a weak company; more recently strong, growing companies held dividends down with investors’ blessing. He has no quarrel with that when the record justifies it. But when a mediocre business keeps its earnings to expand under the same management, he calls the policy illogical on its face and says owners should demand a full defence before accepting it.

This lecture by Graham, reprinted at the back of the edition, argues that the speculative element in stocks had moved. It once lay in weak companies. It had come to lie in the prices of the strongest ones, through valuations built on long-range expectations.

His example is Minnesota Mining and Manufacturing, valued by the market at $1.7 billion with $200 million of net assets behind it, the rest being the market’s estimate of goodwill; within months the market cut that estimate by some $450 million. The conclusion he draws is general: “It follows as a kind of mathematical law that the more important the good will or future earning-power factor the more uncertain becomes the true value of the enterprise, and therefore the more speculative inherently the common stock.”

This appendix is Warren Buffett’s, not Graham’s: a talk answering the claim that no one beats an efficient market except by luck. Buffett presents nine investment records of people he had picked out years earlier for how they decided, not for how they later did.

What they share, in Buffett’s words, is that “they search for discrepancies between the value of a business and the price of small pieces of that business in the market.” He stresses that they took less risk than average, and that their methods otherwise differed widely, which is why he rejects luck as the explanation.

The introduction’s moral about airlines — that a booming industry need not reward the people who own it — is the question this site keeps asking of the AI build. Graham’s two tests travel without adjustment: would the holding make sense if the quote disappeared, and how much of the price rests on earning power nobody can yet measure. The AI capex page applies the cash-flow side of that discipline to the deals now being signed.

Appendix 4 is the passage to reread when a market values a company mostly on what it may earn later. Graham does not call such prices wrong; he says they are speculative by construction, and that the speculation now sits in the best companies rather than the worst.

The lens built on this book is at Benjamin Graham. It pairs with Security Analysis, the textbook where the definitions in Chapter 1 began, and with Irrational Exuberance, which tracks how the story behind a high price spreads.