Net worth grew $1.515 billion in 1989, or 44.4%, the largest dollar gain in the book so far. Buffett opens by explaining why the two things that produced it will not both be there again.

What counts is intrinsic value, the figure that says what all of Berkshire's businesses are rationally worth, and by that measure Berkshire has compounded somewhat faster than the 23.8% of its book value, because most of the businesses have moved from being worth less than their carrying values to worth far more. But a large portion of book value now sits in stocks carried at market, and at year-end those stocks were priced higher, relative to their own intrinsic values, than ever before. Partly that was the buoyant 1989 market. More important, the virtues of these businesses have been widely recognized. "Whereas once their stock prices were inappropriately low, they are not now."

So far Berkshire has benefited from a double dip (Exhibit 1): exceptional gains in intrinsic value at its portfolio companies, plus the bonus of the market appropriately correcting their prices. He will keep most of the major holdings whatever their price, a 'til-death-do-us-part attitude that, combined with full prices, means they cannot be expected to push Berkshire's value up as sharply as before. The catch-up rewards have been collected. From here it is a single dip.

The other obstacle is size (Exhibit 2). Carl Sagan once mused about bacteria that divide every 15 minutes: after a day of wild asexual abandon their descendants weigh as much as a mountain, in two days more than the sun, until the bugs run out of food, or poison each other, or are shy about reproducing in public. He and Charlie are not shy about reproducing, financially, in public, and have not, to their unending sorrow, found a way to double Berkshire's net worth every 15 minutes. But Sagan's point applies. "From Berkshire's present base of $4.9 billion in net worth, we will find it much more difficult to average 15% annual growth in book value than we did to average 23.8% from the $22 million we began with."

Berkshire would owe more than $1.1 billion of tax if it sold all its securities at year-end prices. Buffett explains what that liability is, what it is not, and why the sleeper beats the trader by 27 to 1.

The 1989 gain came after a charge of about $712 million for income taxes, of which about $172 million is paid currently and $540 million is deferred, almost all of it a 34% reserve against the year's rise in unrealized stock profits. Is the $1.1 billion that a total sale would trigger the same as $1.1 billion owed to a supplier 15 days after year-end? Obviously not, though both reduce audited net worth by exactly that sum. Is it a meaningless fiction, since the stocks are, in very large part, ones he has no intention of selling? Again no (Exhibit 4). "In effect, this deferred tax liability is equivalent to a very large transfer tax that is payable only if we elect to move from one asset to another." Berkshire incurred about $76 million of it in 1989 by selling some relatively small holdings with $224 million of gains.

The arithmetic is extreme (Exhibit 3). Put $1 in a security that doubles by year-end, sell, pay 34% on the gain, and repeat for 20 years, and you end with about $25,250, having handed the government about $13,000 along the way. Make one investment that doubles 20 times and sell at the end, and the dollar grows to $1,048,576, the tax is roughly $356,500, and you keep about $692,000. The only difference is when the tax is paid, and the Treasury likes the second scenario too: it collects 27 times as much, though it has to wait.

He stresses that this is not why Berkshire favors long-term commitments. It might earn more by moving frequently, and many years ago that is exactly what he and Charlie did. "We have found splendid business relationships to be so rare and so enjoyable that we want to retain all we develop." Giving up time with people known to be interesting and admirable for time with others likely to be closer to average would be, he says, like marrying for money: a mistake under most circumstances, insanity if one is already rich.

Berkshire reported about $300 million of operating earnings. Buffett says that had its investees paid out the earnings they kept for it, the figure would have been close to $500 million, and the difference is money the accountants are not allowed to count.

The five major investees paid Berkshire about $45 million of dividends in 1989, after taxes. Berkshire's share of the earnings they retained was about $212 million, not counting large capital gains at GEICO and Coca-Cola (Exhibit 5). Had that $212 million been paid out, Berkshire's own operating earnings, after the extra tax, would have been close to $500 million rather than the $300 million in the table. "The question you must decide is whether these undistributed earnings are as valuable to us as those we report." He believes they are, and may be worth more, since managers like these sometimes have better uses for the money in their own businesses than Berkshire would have. "I would not make such a generous assessment of most managements, but it is appropriate in these cases."

So Berkshire's fundamental earning power is best measured, in his view, by a look-through approach that appends its share of the operating earnings retained by investees to its own reported operating earnings, excluding capital gains in both. For intrinsic value to grow 15% a year, look-through earnings must grow at about that pace, which will take plenty of help from the present investees and a new one from time to time. The year's reported figures were sound in their own right: the $300 million came with a four-segment presentation of the businesses, which the auditors in no way bless, and total earnings of $447.5 million including $147.6 million from sales of securities.

With Borsheim's added, the group needed a new title, alliterative and appropriate. Buffett failed, and settled on the Sainted Seven Plus One. The businesses range from good to superb, and he explains what the best of them have in common.

The eight businesses, with some smaller operations, earned 57% after tax on average equity capital on a historical accounting basis, with no net leverage, since cash equivalents matched funded debt. Most of their managers have no need to work for a living: "they show up at the ballpark because they like to hit home runs." Calling off their names, the Blumkin, Friedman, and Heldman families, Chuck Huggins, Stan Lipsey, and Ralph Schey, gives him the glow Miller Huggins must have felt announcing the lineup of the 1927 Yankees. Ralph Schey's operations alone would make a Fortune 500 company, run in a fashion that would put him high in the top decile by return on equity.

Borsheim's, in its first year with Berkshire, met every expectation, with sales considerably better than twice their level four years ago. The huge volume of one location lets it carry an enormous selection and hold its expense ratio to about one-third of comparable jewelers', which with unusual buying power allows prices far below other jewelers', which brings even more volume (Exhibit 6). Bought without an audit, every surprise has been on the plus side, and "'If you don't know jewelry, know your jeweler' makes sense whether you are buying the whole business or a tiny diamond." Asked to educate a Santa Fe gathering about jewels, Ike Friedman decided to dazzle it and brought about $20 million of fancy merchandise to a lodge that was no Fort Knox, reassured a nervous Buffett about the safe's new combination and the two large armed men guarding it, then leaned closer: "the jewels aren't in the safe."

The Buffalo News leads major metropolitan papers in household penetration, daily and Sunday, keeps the biggest news hole of any comparable paper (50.1%, up from 49.5%), and set its seventh consecutive profit record while profits slipped at many major papers (Exhibit 7). To some extent the three are related: a news hole that size cuts profits by itself, but it draws a wide spectrum of readers, and high penetration gives retailers a single megaphone to the whole community. Mrs. B, who started Nebraska Furniture Mart 52 years ago with $500, quit in May after a disagreement over the carpet department, where sales had fallen 17% over three years while the rest of the store grew 24%. He believes the family was entirely correct, and reports that at 96 she has started a new business selling, what else, carpet and furniture, seven days a week. The Mart's carpet department meanwhile reached a 75.3% share of the Omaha market in the fourth quarter, over six times its nearest competitor's.

Industry premiums grew an estimated 2.1% in 1989 against the 10% needed, and the combined ratio jumped to an estimated 110.4. Buffett explains why the "underwriting cycle," taken as a rhythm, is a misnomer, and then describes the ten days in which Berkshire offered to write $250 million of catastrophe cover.

He expects incurred losses to grow about 10% a year (over the last 25 years they have grown 11%), so premium growth far below that means a rising combined ratio. Last year's forecast is repeated, minus the acceleration, and a 10% rate would only stabilize the ratio, not bring it down. Catastrophes led by Hurricane Hugo probably added about two points in 1989. Commentators talk of an underwriting cycle, and "If that term is used to connote rhythmic qualities, it is in our view a misnomer that leads to faulty thinking about the industry's fundamental economics." The ratio once did swing like alternating current, when industry and regulators ran things in cartel fashion and two lags, in rate-setting and in multi-year policies, produced the rhythm (Exhibit 8). That period is long gone.

Now hundreds of participants sell a commodity-like product at independently set prices, which guarantees subnormal profits in all circumstances but one, a shortage of usable capacity. Capacity is described in dollars of net worth, but regulators, brokers, and customers are slow to discipline companies that strain their resources. "At bottom, therefore, the amount of industry capacity at any particular moment primarily depends on the mental state of insurance managers." Good profits come only when capacity is short, shortages come only when insurers are frightened, and that happens rarely and is most assuredly not happening now. Berkshire's premium volume may drop to $150 million or so in 1990, from a high of $1 billion in 1986, which will not disturb him: "We have no interest in writing insurance that carries a mathematical expectation of loss", and Berkshire gets enough disappointments from business it expects to profit on.

Its appetite for properly priced business is ample, as 1989 showed. CAT covers protect an insurer against a single catastrophe above the losses it keeps for itself, usually with one automatic reinstatement, and an event is usually limited to any 72 hours the insurer designates (Exhibit 9). Hugo did its Carolina damage slightly more than 72 hours after its Caribbean damage, the earthquake hit within weeks, and sellers of CAT policies had paid out perhaps twice while many insurers were left actually or possibly bare. Prices instantly became attractive. Just as instantly Berkshire offered to write up to $250 million of catastrophe cover, advertised it in the trade press, and in a busy ten days booked a substantial amount, all of it retained (Exhibit 10). "Our willingness to put such a huge sum on the line for a loss that could occur tomorrow sets us apart from any reinsurer in the world." It can afford to: about $6 billion of net worth sits in its insurers, and it does not care what it reports in a quarter. Had the full $250 million been written, all of it could be lost in one quarter, about $165 million after tax, a blow to pride rather than well-being.

The five holdings worth over $100 million are the same as last year's, with one change in share count: Coca-Cola, up from 14,172,500 shares to 23,350,000. Buffett explains what took him so long.

He believes he had his first Coca-Cola in 1935 or 1936, and in 1936 he was buying Cokes at six for 25 cents from Buffett & Son, the family grocery, to sell around the neighborhood for 5 cents each, an excursion into high-margin retailing in which he duly observed the product's extraordinary consumer attractiveness. For the next 52 years he noted those qualities while Coke blanketed the world and carefully avoided buying a single share, preferring street railways, windmill makers, anthracite producers, textile businesses, and trading-stamp issuers. "Only in the summer of 1988 did my brain finally establish contact with my eyes." What he then saw was a company remade since 1981 by Roberto Goizueta and Don Keough, once his across-the-street neighbor in Omaha, with overseas sales virtually exploding and a rare blend of marketing and financial skill at the top. The holdings are listed in Exhibit 11, at year-end prices he calls much higher relative to intrinsic value than before: not yet nosebleed levels, but vulnerable, so a one-year fall in Berkshire's net worth is almost certain in at least one of the next three years. Arbitrage was nearly idle, mainly because the deals on offer made no economic sense to him. Bond sales produced about $51 million of pre-tax profit, and their proceeds, with spare cash, went into three convertible preferreds (Lesson 06).

Berkshire put $1.26 billion into three convertible preferred stocks and raised $400 million with a zero-coupon bond of its own. Then Buffett takes apart the zero-coupon bonds that Wall Street has been using to make bad deals look good.

In July Berkshire bought $600 million of Gillette preferred, then $358 million of USAir and $300 million of Champion International, each with a fixed dividend, mandatory redemption in ten years, a conversion price, and voting rights as if converted (Exhibit 12). Gillette is very much the kind of business he likes. Airlines, paper, and the investment banking of the 1987 Salomon preferred are industries he cannot forecast: "This does not mean that we predict a negative future for these industries: we're agnostics, not atheists." In one respect the purchases are not different from any other: "We only want to link up with people whom we like, admire, and trust." The structures should return the money plus dividends under almost any conditions, give a mediocre return if industry economics hinder the investees, and a reasonably attractive one if they earn like American industry in general. "The only way Berkshire can achieve satisfactory results from its four preferred issues is to have the common stocks of the investee companies do well."

In September Berkshire sold $902.6 million face value of zero-coupon convertible debentures for $400 million, due in 15 years at the equivalent of 5.5% a year, and since the accrual is tax-deductible each year without being paid, the net effect is positive cash flow. These are not the zeros he and Charlie have attacked. The benign kind began with the wartime Series E Savings Bond and the stripped Treasuries that gave pension funds a guaranteed compound rate (Exhibit 13). Then, as it tends to on Wall Street, the fools followed the wise. Ever-junkier credits issued zeros and PIK bonds in enormous quantities, because a promise to pay nothing cannot be defaulted on, and borrowers introduced EBDIT to judge whether a deal could carry its interest, ignoring depreciation on the theory that it costs no cash this month. Even a high school dropout financing a car knows he needs income to cover interest, running costs, and an honest charge for depreciation: "He would be laughed out of the bank if he started talking about EBDIT."

Then they measured EBDIT against cash interest only, so a company earning $100 million could carry $90 million of cash interest plus $60 million more accruing on a zero, $69 million the next year, creating $150 million of bondholders' "income" from $100 million of earning power (Exhibit 14). Many professional money managers went along, usually with clients' money rather than their own, which he thinks makes "professionals" too kind a word: "promotees" is nearer the mark. The time between folly and failure stretches, and certain savings and loans that bought such bonds with federally insured deposits recorded ultra-high interest they did not receive, a bill the taxpayer will pick up where those loans fail. "A base business can not be transformed into a golden business by tricks of accounting or capital structure." His advice when a banker mentions EBDIT is to zip up your wallet, and his test for the promoter is to offer him zero-coupon fees.

It is a good idea, Buffett says, to review past mistakes before committing new ones. His first was buying control of Berkshire. The condensed version runs to seven lessons, and he hopes the 50-year edition in 2015, if he is around to write it, will need many more pages.

He knew Berkshire's textile business was unpromising in 1965 and bought it because the price looked cheap, the cigar-butt approach: a butt found on the street with one puff left makes that puff all profit. Unless you are a liquidator it is foolish (Exhibit 15). The bargain is rarely the steal it looked, since in a difficult business the problems keep coming: "never is there just one cockroach in the kitchen." And the low return erodes any initial edge. A business bought for $8 million and sold for $10 million at once is a fine deal, and a disappointment if the sale comes ten years later with a few percent a year in between. Hochschild Kohn, a Baltimore department store bought below book value with first-class people, taught him that again. Charlie understood early. Buffett was a slow learner.

A good jockey on a broken-down nag goes nowhere, he says: Berkshire's textile mill and Hochschild Kohn had able, honest people who would have made fine records in a business with good economics, and made none running in quicksand. "I've said many times that when a management with a reputation for brilliance tackles a business with a reputation for bad economics, it is the reputation of the business that remains intact." He wishes he had been less energetic in creating examples, and borrows Mae West: "I was Snow White, but I drifted." Easy does it, too (Exhibit 16). After 25 years he and Charlie have not learned to solve difficult business problems, only to avoid them, concentrating on one-foot hurdles they could step over, with the occasional exception such as the Sunday paper in Buffalo or a marvelous business with a one-time solvable problem, as at American Express and GEICO years back.

His most surprising discovery was the institutional imperative (Exhibit 17): "Instead, rationality frequently wilts when the institutional imperative comes into play." He has tried to run Berkshire to minimize it and to invest in companies that appear alert to it. He learned to go into business only with people he likes, trusts, and admires, which does not by itself ensure success, but "We've never succeeded in making a good deal with a bad person." Some of his worst mistakes were invisible: big purchases served up on a platter that he understood and did not make, a thumb-sucking whose cost to shareholders has been huge. And the conservative financial policy was no mistake. Higher but still conventional leverage would have beaten the 23.8%, perhaps with 99:1 odds of nothing but good, and he would not have liked those odds then or now. "A small chance of distress or disgrace cannot, in our view, be offset by a large chance of extra returns."