This is episode 005 of "Reading with AI." The Intelligent Investor (4th edition, 1973, with Jason Zweig's commentary), Chapter 5: The Defensive Investor and Common Stocks. The previous chapter settled on the recipe—50/50 stocks and bonds, rebalanced every six months—and covered the bond half in depth. This chapter answers the remaining question: for the stock half, what exactly should you buy? Graham's answer is disappointingly plain—four rules, not one of which may be relaxed. And it's the fourth rule (the P/E cap) that sweeps every most-adored growth stock of the day right out of the portfolio; he even uses IBM and Texas Instruments specifically to explain why. Zweig's commentary, meanwhile, takes aim at a slogan that had been popular for thirty years: "buy what you know"—he calls it the greatest misreading of Peter Lynch, and the price of that misreading was Enron employees' entire retirement savings.

AI 带我读书·投资经典
The entire wisdom of defensive stock selection: four hard constraints that shut out the most tempting stocks. What's left looks boring, but it won't take you twenty-five years to break even

The Core Question of This Chapter

By what criteria should an ordinary person unwilling to do deep research pick common stocks—so as to stay hassle-free without stepping on a landmine? And why is the most worth-buying company often precisely one that "doesn't look dazzling"?

Graham opens by going back to the scene of the first edition in 1949. Back then he had to write an entire passage just to convince readers that "the portfolio must contain stocks," because "common stocks were generally thought to be highly speculative and therefore unsafe." The deep slump after the 1946 high had destroyed everyone's confidence: prices were more reasonable, yet people were scarred by "the adverse effect of the decline"—cheapness didn't attract anyone; the fall itself scared them off.

By late 1971, with the Dow at 900, his position had fully flipped:

The reader evidently knows that we have no great enthusiasm for common stocks at these levels.

Yet he still forbids you from selling out of stocks entirely:

The defensive investor cannot afford to be without a proportion of common stocks in his portfolio, if only as the lesser of two evils, since the risk of an all-bond portfolio is greater.

Note the phrasing: "the lesser of two evils." Graham never says stocks are a good thing; he says "all bonds would be worse." That is a completely different grammar from today's "holding stocks long-term always wins" slogans.

Graham's Main Text

Key Point 1: The Two Advantages of Common Stocks, and the Condition That Makes Them "Evaporate"

Graham sums up the 1949 conclusion in two points:

Advantage	Content
1. Inflation protection	"Stocks in large measure protected the investor against the loss of purchasing power caused by inflation, whereas bonds offered no such protection."
2. Higher average return over many years	From two sources: ① average dividend levels higher than bond interest; ② long-term appreciation in market value from reinvested undistributed profits

Then the warning comes immediately:

While these two advantages were extremely important... we would continue to urge the following caution: if the investor pays too high a price for his stocks, these advantages will evaporate.

The word "evaporate" is the foundation of the entire chapter. He offers two pieces of evidence:

1929: clearly a case in point—"it took 25 years for the market to recover." Zweig nails the number down in a footnote: the Dow stood at 381.17 on December 3, 1929, and did not return to 382.74 until October 23, 1954. Then he asks a chilling question:

When you try to hold stocks "for the long term," do you realize how long that long term can be? Did it occur to you that some investors who bought stocks in 1929 were no longer alive by 1954?

1957: common stocks "again lost their traditional advantage of a dividend yield exceeding bond yields."

On this 1957 turning point, Zweig adds a highly valuable piece of history: Wall Street's authorities at the time widely believed this state of affairs was "unsustainable"—stocks are riskier than bonds, so if even their yield fell below bonds, who would buy them? They predicted the relationship would last only "a few months." The result:

More than forty years have now passed, and the relationship has never returned to "normal." To this day, stock yields remain below bond yields.

"What the experts unanimously declared unsustainable lasted forty years"—this book has enough cases like this to fill a separate volume.

And this episode also surfaces a more important fact. Zweig cites research by three London Business School professors (Dimson / Marsh / Staunton):

$1 invested in stocks in 1900	Result in 2000
All dividends spent	$198
All dividends reinvested	$16,797

Clearly, dividends were the greatest attraction of stock investment.

An 84-fold difference. Even the grim 1929–1954 stretch—that "quarter century without breaking even"—produced a positive return for the patient investor who kept reinvesting dividends, because the average dividend yield then was 5.6%. What brought you back to even wasn't the price; it was the dividends.

Key Point 2: The Four Rules for Selecting Common Stocks (the Chapter's Skeleton)

Graham calls the matter "relatively simple," then gives four rules:

1. There should be adequate but not excessive diversification, meaning a minimum of ten and a maximum of thirty different stocks.

2. Each company selected should be large, prominent, and conservatively financed.

3. Each company should have a long record of continuous dividend payments. (He suggests starting from at least 1950—i.e., more than 20 consecutive years at the time.)

4. The investor should limit the price he pays for stocks to a range of P/E ratios: relative to average earnings per share over the past seven years, the P/E should be no more than 25; if based on earnings over the past twelve months, it should be no more than 20.

Put into a checkable table:

#	Rule	Specific threshold	What it guards against
1	Diversification	10–30 stocks, across industries	Any single stock blowing up to zero
2	Quality	Large, prominent, conservatively financed	The company itself disappearing
3	Dividend history	Continuous dividend payments (≥20 years)	"Story companies"—is the profit even real
4	Valuation	P/E ≤ 25 on 7-year average earnings, or ≤ 20 on trailing 12 months	Overpaying; "a great company that traps you for a decade"

On rule 3, Graham said that in 1971 all Dow components qualified. Zweig updated it to 2003: requiring 10 consecutive years of dividends would exclude Microsoft from the Dow, yet 317 stocks in the S&P 500 would still qualify; even requiring 20 consecutive years, 255 stocks in the S&P 500 would qualify. This filter is not nearly as stringent as it sounds.

As for rule 4, Graham himself admits its destructive power:

But this restriction would keep out of our portfolio nearly all the most powerful and most popular stocks, and in effect nearly all "growth stocks," which had for many years been the favorites of both speculators and investors in the stock market. We must therefore give our reasons for so thoroughgoing an exclusion.

Notice the gesture: he knows his rule will cut out the hottest entire sector of the market, so he stops and devotes a section to justifying it. That is the next key point.

Key Point 3: Why Growth Stocks Are Excluded "Outright"—IBM and Texas Instruments

First, the definition. A growth stock is one whose per-share earnings have grown significantly above average in the past and are expected to keep doing so. Graham cites the industry convention of the day: a true growth stock should at least double its per-share earnings within ten years—compound annual growth of 7.1%.

(Zweig, in a footnote, passes along a mental-math tool—the "Rule of 72": 72 ÷ growth rate = years to double. 6% growth → doubles in 12 years; 7.1% → 10.1 years. Worth memorizing on its own.)

Then comes Graham's turn:

Obviously such stocks are worth buying and owning, provided the price is not too high. There is the difficulty, of course: relative to current earnings, growth stocks have always been expensive; relative to earnings over some past period, their P/E is higher still. This introduces a large speculative element into growth-stock investing, making successful operation of it very difficult.

He doesn't deny that growth stocks are good companies; what he denies is that buying them at that price still counts as investing. Then come the two cases:

Case 1: IBM—even "the best common stock" can be cut in half twice

IBM had "long been the leader among growth stocks" and did reward those who bought and held years earlier. But:

Time	What happened
1961–1962 (within six months)	The stock lost half its value
1969–1970	It fell by nearly the same amount again

Case 2: Texas Instruments—price rose 5× as fast as earnings, then fell four-fifths

Phase	Stock price	Earnings per share
6-year rise	$5 → $256 (paying no dividend throughout)	$0.40 → $3.91
The following 2 years	Fell to $49 (down four-fifths)	Down nearly 50%

Graham makes the crucial point in parentheses:

Note that the advance in the stock price was equivalent to five times the gain in earnings—a general characteristic of this type of hot issue.

This is the most valuable technical observation in the chapter: when a hot stock rises, multiple expansion contributes several times what the business results contribute. And on the way down, this leverage works in reverse—earnings fall 50%, the stock falls 80%. A double blow: not only is the market declining, the company's profits are falling too.

Zweig proves the law hadn't expired with a comparison set 30 years later:

Graham's version (1970s)	Zweig's version (2000–2002)	Result
IBM	Microsoft	Stock down 55.7%
Texas Instruments	Cisco	Stock down 76% (after a 50-fold rise over the prior six years); earnings down only 39.29%

As always, they rose faster and fell harder.

Hence Graham's conclusion:

For the defensive investor, growth stocks are too uncertain and too risky. To be sure, miracles can happen if the right stock is picked at the right price and sold after a huge rise but before the possible decline. For the ordinary investor, however, such things are a matter of luck, not design.

Note that his reason for excluding growth stocks is not "growth stocks are bad" but "growth stocks demand that you get three things right in a row": pick the right company + buy at a suitable price + sell after the big rise but before the fall. Get any one of the three wrong, and the correctness of the others counts for nothing. For someone not doing deep research, those odds aren't worth betting on.

The alternative:

We believe that less glamorous large companies, with less popular and hence more reasonable earnings multipliers, are actually a more suitable choice for most investors, unattractive though they may appear.

Key Point 4: Portfolio Review—Once a Year Is Enough

Graham's attitude toward "rebalancing" is understated to the point of indifference:

Our defensive investor should likewise seek such advice on improving his portfolio—at least once a year—just as he sought advice when making his initial investment.

But with three provisos, each more important than the last:

Lacking the expertise to judge advisers, he should go only to the most reputable institutions; otherwise "he is likely to be taken in by some 'second-raters'";
He must make it clear to every adviser that he intends to adhere to the selection principles set forth at the beginning of this chapter;
The final sentence is the key one:

If the stock list is well chosen initially, there should be no need for frequent or radical changes.

Zweig updated the tools (various online "portfolio trackers" can build automated monitoring systems) but added a warning:

Graham warned us not to rely exclusively on such systems. You must use your own judgment to supplement what the software lacks.

Proviso 2 deserves to be singled out. The act of "declaring your principles to your adviser" turns "you pick for me" into "you pick for me within my framework"—the position of control is entirely different. Most people buying financial products today hear the recommendation first and ask about principles afterward; Graham demands the reverse.

Key Point 5: Dollar-Cost Averaging—Tested Across 23 Ten-Year Periods

That is, investing a fixed amount at fixed intervals. The NYSE then promoted a "monthly purchase plan"—a fixed sum each month into one or more stocks. Graham says that during the great bull run since 1949, the results were "quite satisfactory," above all in effectively preventing investors from buying heavily at the wrong times.

He cites Lucile Tomlinson's complete study, using Dow Jones Industrial Average stocks, covering 23 ten-year periods (the first ending in 1929, the last ending in 1952):

The average profit at the end of the 23rd purchase period came to 21.5%—dividends not included.

He also concedes that "in some of these periods the market value of the investor's stocks would show a marked decline." And Tomlinson's conclusion is, unusually, without any hedging:

"No matter how security prices fluctuate, this investment method makes possible final success with confidence; to date, no investment method comparable to dollar-cost averaging has been devised."

Graham raises a reader's objection himself, and answers it himself:

Objection: Who can put the same sum into stocks every month for twenty years? "Very few indeed."
Answer: The objection "has lost much of its force in recent years." Common stocks have been widely accepted as a necessary component of savings-investment plans, so "just as with the continuous purchase of U.S. Savings Bonds and life insurance, systematic and consistent purchases of stocks should cause the investor little psychological or financial distress." The monthly sums are small, but over twenty years "the total will be quite considerable."

A methodological point hides here: the real difficulty of dollar-cost averaging isn't the math; it's whether it can become something that requires no decision each time. Graham chose his analogy well—nobody suffers over paying their insurance premium each month, because it has been automated.

Key Point 6: Three Case Studies—the Widow, the doctor, and the young person

This is the most lifelike section of the chapter. Graham gives us three people:

#	Situation	Graham's judgment
1	A widow with $200,000 who must support herself and her children from it	Funds divided roughly equally between U.S. government bonds and high-grade common stocks (stocks could go up to 75%, but only if she is psychologically prepared and confident the prices paid are not high—"obviously, the stock market in early 1972 did not meet this requirement")
2	A successful physician in mid-career with $100,000 in savings, adding $10,000 a year	"His investment choices would be essentially the same"; new savings deployed in the same proportions
3	A young person earning $200 a week and saving $1,000 a year	Part into Series E bonds; the capital is too small to justify the rigorous training of an enterprising investor—the defensive standard approach "is unquestionably the simplest and most sensible strategy"

(Zweig's note: multiply these figures by 5 for a rough equivalent to the early 21st century.)

Each of the three carries a different lesson:

The widow's lesson: she must never speculate for "extra income."

There is one thing she cannot do: undertake speculative operations to earn some "extra income"... To maintain herself, she would be better advised to take $2,000 a year from her principal than to risk half of it in unreliable and hence speculative ventures.

"Drawing on principal" sounds bad; "speculating for extra income" sounds enterprising—and Graham ranks them exactly the other way around.

The doctor's lesson: intelligence and confidence are handicaps, not advantages.

He has one great limitation: he lacks the time to acquire an investment education and manage his own investments. In truth, the ineptitude of the medical profession in handling securities is proverbial. The reason is that physicians typically have great confidence in their own intelligence and an eagerness to profit, without recognizing that investment success requires a great deal of energy and professional judgment of security values.

"Great confidence in their own intelligence" is explicitly listed as the cause of failure. That sentence applies far beyond doctors—programmers, lawyers, engineers: anyone whose intelligence has been rewarded in another field should read it three times.

The young person's lesson: you may try, but with small money.

It is advantageous for a young investor to begin his investment education and practice early. If he operates as an enterprising investor, he will surely make some mistakes and suffer some losses. Youth can absorb these failures and benefit from them. Our advice to beginners is this: do not waste your energy and money trying to beat the market. Study the value of securities and test your judgments about price and value with the smallest sums of money you can.

"Test your judgment with the smallest possible sums"—that is Graham's entire advice to the young, essentially "keep the tuition within what you can afford."

The three cases converge on a standard of judgment that runs through the whole book:

What securities the investor should buy, and what return he should seek, depends not on how much money he has but on his financial capability—comprising knowledge, experience, and temperament.

What determines what you should buy is not how much money you have; it is how much knowledge, experience, and temperament you have.

Key Point 7: A Note on "Risk"—Fluctuation Is Not Risk

This section is short, but it foreshadows Chapter 8 on market fluctuation and is one of value investing's defining passages.

Graham begins by noting that "risk" and "safety" have two different meanings in the securities field, causing confusion. He lists "genuine risk":

A bond's failure to pay interest and principal at maturity → unsafe;
A cut or cancellation of preferred or common dividends → unsafe;
The holder's real possibility of having to sell at a price far below what was paid → involves risk.

Then the crucial exclusion:

Yet people often extend the concept of risk to cover the possibility that a security's price may decline, even if the decline is only cyclical and temporary and the holder is under no necessity of selling at that point.... In any practical sense, we do not consider this genuine risk.

He uses two analogies, both tough:

Analogy	Logic
A home mortgage	Being forced to sell the house at a bad time would mean a huge loss, yet no one judges the loan's safety on that basis—"the sole criterion of its safety is whether payments of interest and principal are made on time"
Running a business	Risk should be judged by "the chance of losing money," not by "what happens when the owner is forced to sell out"

The conclusion:

So far as the true investor is concerned, a decline in the market price alone does not make him poorer; hence the mere fact that the market may fall does not mean he faces an actual risk of loss.

Graham gives "risk" a deliberately narrowed definition—applying it only to "a loss of value"—in three specific situations:

The security is actually sold;
The company's position seriously deteriorates;
Most commonly—a loss occurs because the purchase price was too high relative to intrinsic value.

And situation 3 is the entire rationale for this chapter's rule 4 (the P/E cap):

If the price paid relative to the stock's intrinsic value is too high, this risk presents itself—even if a badly falling market recovers its lost ground after many years.

"Even if the market recovers its lost ground years later, overpaying was in itself already a loss"—this sentence shuts down the faith that "long-term holding fixes everything."

Key Point 8: Turning "Large, Prominent, Conservatively Financed" into Actionable Numbers

Graham knew perfectly well that adjectives can't be relied upon:

Any criterion based on adjectives is always vague.

So he offers quantitative definitions:

Adjective	Actionable standard
Conservatively financed	Industrial company: book value of common stock ≥ half of total capital (including all bank debt); railroad or public utility: ≥ 30%
Large	At the time, assets of at least $50 million, or revenue of at least $50 million
Prominent	Size in the top quarter or third of its industry (regarded in the trade as "major" rather than "minor")

(Zweig's update: today "large" should mean a market capitalization of at least $1 billion; in 2003 about 300 U.S. stocks qualified.)

But Graham immediately adds a self-deprecating line whose methodological value exceeds the standards themselves:

It would be foolish, however, to insist rigidly on such arbitrary criteria. They are offered only as guides. Any standards an investor sets for himself are acceptable, provided they do not violate the accepted meanings of "large" and "prominent."

And he believes the vagueness is a good thing:

Because of the inherent vagueness of such definitions, some companies suitable for the defensive investor will be chosen by one buyer and rejected by another. This divergence of opinion and difference of choice is not harmful. Indeed, it is beneficial to stock-market trading, since it makes the distinction between leading and secondary stocks more subtle and more nuanced.

Graham never pretends his numbers are precise. He wants "constraints of the right order of magnitude," not "the optimum solution to two decimal places."

Zweig's Commentary (Fourth Edition)

It opens by quoting Benjamin Franklin: "Human felicity is produced not so much by great pieces of good fortune that seldom happen, as by little advantages that occur every day."

Commentary 1: Once Burned by Hot Milk, You Blow on Yogurt Too

After the crash of 2000–2002, why should the defensive investor still own stocks? Zweig begins with a Turkish proverb:

"After you've been burned by hot milk, you blow on your yogurt."

Then the key reversal:

Because the crash of 2000–2002 was so terrifying, many investors now feel stocks will burn them. What they don't realize is that it is precisely this decline that has released most of the market's risk. Before, it was indeed a cup of hot milk; now it has cooled to room temperature.

Whether you should own stocks now has nothing to do with how much money your stocks lost you a few years ago. If stock prices have become reasonable enough for your wealth to grow from here, you should buy them—regardless of whether they recently lost you money.

"Should I buy?" is a question about price, not about how much you lost in the past. This is one of the hardest hurdles in behavioral finance—separating the decision from the memory of pain.

He also delivers a withering verdict on all-bond portfolios: in early 2003 stocks were only slightly above their historical average valuation while bond yields were very low, so "people who buy bonds out of a desire for safety are like smokers who believe that low-tar cigarettes will protect them from lung cancer."

Zweig also, right at the start, corrects the very framing of "whether you dare to own stocks":

(Whether you hold stocks) does not depend on your tolerance for risk, but on how much time and effort you are willing to devote to your portfolio. If your approach is sound, investing in stocks can be every bit as easy as holding bonds and cash.

Commentary 2: "Buy What You Know"—the Greatest Misreading of Peter Lynch

This is the main battleground of the chapter's commentary. Zweig first sets up the target: through the 1980s and early 1990s, "buying what you know" was the most popular investing slogan, and its foremost champion was Peter Lynch, who ran Fidelity's Magellan Fund from 1977 to 1990.

Lynch's original meaning was that amateur investors have the advantage of "the power of common knowledge"—you discover a wonderful new restaurant, car, toothpaste, or jeans; you see the parking lot of a nearby store always packed; and so you form a first-hand sense of a stock that even professional analysts lack. "Above all, you know the news before Wall Street does."

Zweig doesn't deny the rule: "It's not ridiculous, and over the years thousands of investors have profited from it." But—

Lynch's rule works only if you follow his conclusion: "Finding a promising company is only the first step. The next step is researching it." What he really meant was that you must never buy a stock until you have studied its financial statements and weighed the value of its business—no matter how wonderful its product looks or how many customers' cars are in its parking lot.

Unfortunately, most stock buyers ignore that part.

"Familiarity" is only the entrance to the candidate pool, not a reason to buy. And it's exactly the second half that got cut off.

Commentary 3: Three Case Studies in Misreading—Streisand, Landis, and the Enron employees

Zweig gives three cases of escalating severity:

Case 1: Barbra Streisand's Starbucks (the personal version)

In 1999 she declared: "We go to Starbucks every day, so I buy Starbucks stock."

But the stage and screen star forgot that no matter how much you love the lattes, you still have to analyze the financial statements to make sure the stock is not more overpriced than the coffee.

Zweig adds that many people bought Amazon because they loved the website, and invested in E*TRADE because it was their online broker—"they are making exactly the same mistake Barbra Streisand did."

Case 2: Kevin Landis's confidence (the expert version)

The Firsthand Technology Value Fund returned an average of 58.20% per year from 1995 to 1999. On a CNN interview someone asked, "Kevin, how do you do it? Why can't I?" He answered excitedly:

"Oh, you can do it too. All you have to do is watch the things you're familiar with, stay in close touch with the industry, and talk to industry insiders regularly."

The ending sits in a footnote: from the end of 1999 to the end of 2002, Landis's fund (invested mostly in technology companies he claimed first-hand knowledge of) lost 73.2% of its value—more than the average loss of high-tech funds over the same period.

"All you really need to do" turned out not to be merely "watching the things you're familiar with"—he supplied the counterexample himself.

Case 3: Your own company's stock in the 401(k) (the institutional version)

The biggest misuse of Lynch's rule came in corporate retirement plans. If you should "buy companies you're familiar with," surely the best investment of all is to put the retirement money in your 401(k) into your own employer's stock. After all, you work there—don't you know that company better than any company outside?

Unfortunately, the employees of Enron, Global Crossing, and WorldCom (many of whom had put virtually their entire retirement savings into their own company's stock and were wiped out) discovered that what insiders possess is often a pile of illusions, not the facts.

The devastating thing about this case: those employees were the most "familiar" people on earth with those companies. Familiarity maxed out; the retirement savings went to zero.

Commentary 4: Home Bias—Why Familiarity Is Actually Dangerous

Zweig supplies the mechanism-level explanation, citing research by psychologists including Fischhoff at Carnegie Mellon:

Learning more about something does not appreciably reduce people's tendency to exaggerate how much they actually know.

That is the danger in "investing in stocks you know about." The more information you hold, the laxer your policing of that stock's weaknesses may become.

This overconfidence has a name—"home bias", the habit of becoming obsessed with what is already familiar. Three data points:

Group observed	Manifestation of home bias
Retail investors	They hold 3 times as much stock in their local phone company as in all out-of-town phone companies combined
Mutual funds	The companies the funds hold are, on average, 115 miles closer to fund headquarters than the average U.S. company is to its own headquarters
401(k) investors	They put 25%–30% of their retirement assets into their own employer

Then comes a brilliantly apt analogy:

On television news, we always see the criminal's neighbors, close friends, or parents lament in shock, "He always seemed like such a nice person!" That's because whenever we are too close to certain people and things, we take them for granted in a spirit of acceptance rather than inquiry—not with the skepticism we bring to things that are more remote.

By the same token, the more familiar you are with a stock, the lazier it may make you as a defensive investor, making you feel you need not bother. Don't let that happen to you.

"Familiarity breeds complacency"—this is the central thesis of the chapter's commentary. Closeness weakens skepticism, and investing is an activity that must be sustained by skepticism.

Commentary 5: Three Implementation Paths for the Defensive Investor

Zweig breaks "how to actually do it" into three paths, each with clear boundaries:

Path 1: Do it yourself

Buy regularly through a low-commission automated investing brokerage site (commissions then ran about $4 per purchase with no account minimum); you can invest weekly or monthly, reinvest dividends, and fund the account automatically from your bank. He pours on three buckets of cold water:

Selling commissions run higher than buying commissions—"beware: quick selling is the great enemy of stock investing";
If you can save only $50 a month, a $4 commission amounts to 8%—but for the small investor, these micro-investing sites "are probably your only path to a diversified stock portfolio";
Small annual purchases create mountains of tax paperwork: "If you're too lazy to keep detailed records of your purchases, you shouldn't invest in stocks."

And the most important diversification bottom line:

Next, do not buy just one stock—or even a handful. If you won't spread your bets, don't bet at all. Graham's requirement that investors diversify across 10 to 30 stocks remains the fundamental law of stock investing, and you must not let those stocks become too concentrated in any one industry.

And one self-check red line, which I think is the single most practical sentence in the whole commentary:

After you've set up this autopilot web portfolio, if you find yourself trading more than twice a year, or spending more than an hour or two per month on your stocks, something is wrong. Do not let the ease and immediacy of the Internet turn you into a speculator; defensive investors win by keeping their cool and their patience.

"No more than 2 trades a year, no more than 1–2 hours a month"—a quantitative standard you can copy straight into your own investment discipline.

Path 2: Get help

Through a discount broker, a financial planner, or a full-service broker. But:

There is one responsibility you can never hand off to anyone else: thoroughly investigating your adviser's trustworthiness and the fairness of the fees (before entrusting your money).

Path 3: Borrow the hen (mutual funds / index funds)

For the defensive investor, mutual funds are an ideal way to capture the benefits of stock investing while sidestepping its hassles.

And the best form of all is the index fund:

Their best form, for both diversification and convenience, is the so-called index fund—one that requires no monitoring or adjustment whatsoever. An index fund is a pipe-dream investment, requiring no labor and bringing almost no surprises, even if you sleep for twenty years like Rip Van Winkle, the lazy farmer in Washington Irving's tale. They are where the defensive investor's dreams come true.

Commentary 6: Dollar-Cost Averaging Put to the Test—Starting at the Worst Point, September 1929

Graham laid out the principle; Zweig supplies the most extreme empirical test. The research is from Ibbotson Associates:

Approach	Starting point (early September 1929)	Ten years later (August 1939)
Lump-sum purchase	$12,000 into the S&P 500	Only $7,223 left
Dollar-cost averaging	Starting with $100, adding $100 per month	Grown to $15,571

Such is the power of buying on schedule—even in the face of the worst bear market in history, the Great Depression.

Same market, same starting point, same worst decade: one loses 40%, one gains 30%. The only difference is whether the money went in all at once or in installments.

He also gives a set of recent numbers: with the S&P 500 falling from the end of 1999 through the end of 2002, an index fund account opened with a $3,000 minimum and $190 added monthly—$6,600 in total investment—would have lost 30.2%, well below the market's 41.3% decline, and "your continued purchases at lower prices would earn you handsome profits when the market rebounded."

A concrete execution template (assuming $500 left over each month):

Allocation	Amount
U.S. stock market index fund	$300
Foreign stock index fund	$100
U.S. bond market index fund	$100

This way you can be confident you own nearly everything on this planet worth owning. Like clockwork, you buy more stocks every month; if the market declines, your fixed investment buys more shares than the month before; if it rises, the same dollars buy fewer shares than the month before.

And what it truly protects against is two things:

By handling your portfolio in this nearly automatic way and sticking with it, you avoid two scenarios: tossing money into the market when it seems most attractive (and is actually most dangerous), or refusing to buy more stocks when the market has crashed and prices are genuinely cheap (but seem more "risky").

Commentary 7: "I Don't Know, and I Don't Care"

The commentary closes with one of the most famous passages in the entire book. Once you have built a permanent, autopilot portfolio centered on index funds, you can give the defensive investor's most powerful answer to every market question:

Question	Answer	Reason
Will bonds outperform stocks?	"I don't know, and I don't care."	You automatically own both
Will health-care stocks outshine tech stocks?	"I don't know, and I don't care."	You are already a long-term owner of both
Who will be the next Microsoft?	"I don't know, and I don't care."	Once it's big enough, the index fund will buy it, and you'll be along for the ride
Will foreign stocks beat U.S. stocks next year?	"I don't know, and I don't care."	If they do, you profit; if not, you buy more at lower prices

The ability to say, with total conviction, "I don't know, and I don't care"—such a permanent, autopilot portfolio will set you free, liberating you from the exhausting effort of forecasting the market that others remain addicted to despite its futility. Admitting how little you know about the future, and being at peace with that ignorance, is the defensive investor's most powerful weapon.

Turning "ignorance" from a defect into a weapon—this is where the Chapter 5 commentary ends, and it is the book's most unmistakable statement of attitude.

Untangling the Chapter's Hard Points

Hard point 1: If growth stocks deliver higher long-term returns, why does Graham exclude them "outright"?

The key is that his reason is not "growth stocks are bad," but that "growth stocks demand that you get three things right in a row":

Pick the right company (IBM was right; Texas Instruments was, for a stretch, right too);
Buy at a suitable price (the hardest of the three, because growth stocks "relative to current earnings have always been expensive");
Sell after the huge rise but before the possible decline.

And the pricing mechanics of hot stocks make conditions 2 and 3 nearly impossible to satisfy at once: Texas Instruments' stock rose 5× as fast as its earnings, which means the overwhelming majority of what you paid for was multiple expansion, not business results. When the multiple reverses, earnings down 50%, stock down 80% is the inevitable result (Cisco replayed it 30 years later: earnings down 39%, stock down 76%).

For the defensive investor, the probability of "getting all three right" isn't worth betting on—this is not a verdict on growth stocks; it is an honest assessment of your own odds.

Hard point 2: How do you use the "7-year average earnings P/E ≤ 25" rule today?

Three essentials:

Use "multi-year average earnings," not "latest year's earnings," as the denominator—this is the soul of the rule. A single year's earnings can be inflated or deflated by cycles, one-off items, and accounting choices (Chapter 12 covers this in depth). Using a 7-year average automatically discounts cyclical industries and companies that "just happened to blow out this year."
The two thresholds are not alternatives; they are double protection: 7-year-average P/E ≤ 25 and trailing-12-month P/E ≤ 20. The former guards against "one good year masking years of mediocrity"; the latter against "an ancient low base masking how expensive things are today."
The closest ready-made tool today is the Shiller P/E (CAPE, the P/E on 10-year inflation-adjusted average earnings)—which already appeared in Chapter 3. Graham used 7 years in 1973; Shiller later used 10. The idea is identical.

Hard point 3: How should "fluctuation is not risk" actually be understood? It is not permission to ignore declines.

Graham's narrowing of the definition is very precise. He applies "risk" only to a loss of value, in three situations:

Situation	Genuine risk?	Notes
The security is actually sold (below the purchase price)	✅ Yes	The loss is realized
The company's position seriously deteriorates	✅ Yes	The value is truly gone
The purchase price was too high relative to intrinsic value	✅ Yes, and "most commonly"	Even if the market recovers its lost ground years later, overpaying was in itself already a loss
A cyclical, temporary price decline, with no need to sell	❌ No	"In any practical sense, we do not consider this genuine risk"

Note that row 4 holds only under one hard precondition: "and the holder is under no necessity of selling at that point." What guarantees that precondition?—The previous chapter's 25% bond floor (Zweig: "the cushion of bond holdings will give you the courage to keep holding the rest of your stocks through a bad market"), plus this chapter's diversification and "not needing investment income to cover living expenses."

In other words, "fluctuation is not risk" is not an article of faith; it is a conclusion with preconditions. If you are fully invested, leveraged, and might need the money at any moment, then fluctuation is genuine, hard risk for you—because you will be forced to sell, and row 4 becomes row 1.

Hard point 4: Are "buy what you know" and "invest within your circle of competence" the same thing?

No. The difference lies precisely in the amputated half-sentence:

	Familiarity	Study the financials and business value?	Result
Lynch's original meaning	Provides the candidate pool	✅ Mandatory ("the next step is researching it")	Works
The popular misreading	Treated directly as the reason to buy	❌ Skipped	Streisand buys Starbucks, the Landis fund falls 73.2%, Enron employees' retirement savings go to zero

And psychology's mechanism is even more counterintuitive: greater knowledge "does not appreciably reduce people's tendency to exaggerate how much they actually know"—in fact "the more information you hold, the laxer your policing of that stock's weaknesses may become."

So the correct use of the "circle of competence" is: familiarity determines which companies you're entitled to research, not that you may skip the research. Familiarity is an admission ticket, not a waiver of inspection.

The One-Sentence Takeaway

The defensive investor selects stocks by only four rules: 10–30 stocks diversified across industries; every one large, prominent, and conservatively financed; a long record of continuous dividends; and a P/E no higher than 25 on 7-year average earnings (no higher than 20 on the trailing 12 months). The fourth rule sweeps all the hottest growth stocks of the day out of the portfolio—because they require you to get three things right in a row: "pick the right company + buy at the right price + sell after the big rise but before the fall," and Texas Instruments' stock rose 5× as fast as its earnings; when it turned, earnings fell 50% and the stock fell 80%—which Cisco replayed verbatim thirty years later. Real risk is not price fluctuation but paying more than intrinsic value—"even if a badly falling market recovers its lost ground after many years," you have already lost (those who bought in 1929 had to wait until 1954, and many didn't make it). As for "buy what you know," please restore Lynch's amputated second half: "Finding a promising company is only the first step. The next step is researching it." Familiarity breeds complacency—and Enron's employees were the people on earth most familiar with Enron.

Questions to Leave You With

Run your current holdings through this chapter's four rules: ① How many stocks do you own, across how many industries? ② Is every one of them "large, prominent, and conservatively financed" (using Graham's quantitative test: book value of common stock ≥ half of total capital)? ③ Does every one have 10+ consecutive years of dividends? ④ At a P/E computed on average earnings per share over the past 7 years, how many exceed 25? How many of your stocks pass all four?
Graham says the problem with growth stocks is that they "demand you get three things right in a row." Recall the most expensive stock you ever bought: at the time, did you have a concrete plan for the third thing (selling after the big rise, before the fall), or were you planning to "decide when the time comes"? If you had no plan, what were you actually betting on when you bought it?
Work out the split between "valuation contribution" and "earnings contribution" in your holdings: how much has a stock risen over the past 3 years, and how much did its earnings per share rise over the same period? If the stock rose more than 3× as fast as earnings, that is by Graham's account "a general characteristic of this type of hot issue"—if the multiple simply reverted to its level of 3 years ago, how far would the stock have to fall?
Zweig offers a quantitative red line: more than two trades a year, or more than an hour or two per month spent on stocks, means something is wrong. Count your actual trades over the past 12 months and your monthly hours spent watching the market and reading financial news. By this standard, are you currently a defensive investor or a speculator?
Home bias in three numbers (3× as much local stock; fund holdings 115 miles closer; 25%–30% of retirement money in the employer's own stock)—what proportion of your portfolio is "bought because it's familiar": your own company's or industry's stock, the companies behind products you use every day, tips from friends where you feel "I know this business"? Of those, how many full annual reports have you actually read?
The worst starting point, September 1929: $12,000 invested in a lump sum left $7,223 after ten years; $100 a month invested steadily became $15,571. If a three-year decline started tomorrow, would your current finances let you keep buying every month—or would you stop, planning to "wait until it's over"? The answer to that question determines whether dollar-cost averaging is, for you, a strategy or a slogan.

Next episode: Chapter 6, "Portfolio Policy for the Enterprising Investor: Negative Approach"—the defensive map is now drawn; next comes the enterprising investor's turn. Interestingly, Graham's first chapter on the enterprising investor is not about "what to buy," but about "what not to buy": high-grade preferred stocks, second-tier bonds, foreign government bonds, new issues (IPOs)... He will offer a principle that sounds strange at first—the enterprising investor is not "a defensive investor with more guts." He must first cross a whole batch of "higher-yielding" instruments off the menu, because the extra yield on those things compensates for someone else's risk, not yours.

Reading with AI 05 | The Intelligent Investor: Chapter 5 — The Defensive Investor and Common Stocks