Berkshire Hathaway Letters to Shareholders: 1965-2024

by Warren Buffett

This change underscores the need for companies also to report segmented data: The greater the number of economically diverse business operations lumped together in conventional financial statements, the less useful those presentations are and the less able investors are to answer the three questions posed earlier. Indeed, the only reason we ever prepare consolidated figures at Berkshire is to meet outside requirements. On the other hand, Charlie and I constantly study our segment data.

I referred earlier to a major change in GAAP that is expected in 1990. This change relates to the calculation of deferred taxes, and is both complicated and controversial—so much so that its imposition, originally scheduled for 1989, was postponed for a year. When implemented, the new rule will affect us in various ways. Most important, we will be required to change the way we calculate our liability for deferred taxes on the unrealized appreciation of stocks held by our insurance companies. Right now, our liability is layered. For the unrealized appreciation that dates back to 1986 and earlier years, $1.2 billion, we have booked a 28% tax liability. For the unrealized appreciation built up since, $600 million, the tax liability has been booked at 34%. The difference reflects the increase in tax rates that went into effect in 1987. It now appears, however, that the new accounting rule will require us to establish the entire liability at 34% in 1990, taking the charge against our earnings. Assuming no change in tax rates by 1990, this step will reduce our earnings in that year (and thereby our reported net worth) by $71 million. The proposed rule will also affect other items on our balance sheet, but these changes will have only a minor impact on earnings and net worth.

We have no strong views about the desirability of this change in calculation of deferred taxes. We should point out, however, that neither a 28% nor a 34% tax liability precisely depicts economic reality at Berkshire since we have no plans to sell the stocks in which we have the great bulk of our gains. To those of you who are uninterested in accounting, I apologize for this dissertation. I realize that many of you do not pore over our figures, but instead hold Berkshire primarily because you know that: (1) Charlie and I have the bulk of our money in Berkshire; (2) we intend to run things so that your gains or losses are in direct proportion to ours; and (3) the record has so far been satisfactory. There is nothing necessarily wrong with this kind of “faith” approach to investing. Other shareholders, however, prefer an “analysis” approach and we want to supply the information they need. In our own investing, we search for situations in which both approaches give us the same answer.

our experience with newly-minted MBAs has not been that great. Their academic records always look terrific and the candidates always know just what to say; but too often they are short on personal commitment to the company and general business savvy. It’s difficult to teach a new dog old tricks.

At Fechheimer, the Heldmans—Bob, George, Gary, Roger and Fred—are the Cincinnati counterparts of the Blumkins. Neither furniture retailing nor uniform manufacturing has inherently attractive economics. In these businesses, only exceptional managements can deliver high returns on invested capital.

Take the breakfast cereal industry, whose return on invested capital is more than double that of the auto insurance industry (which is why companies like Kellogg and General Mills sell at five times book value and most large insurers sell close to book).

The cereal companies regularly impose price increases, few of them related to a significant jump in their costs. Yet not a peep is heard from consumers. But when auto insurers raise prices by amounts that do not even match cost increases, customers are outraged. If you want to be loved, it’s clearly better to sell high-priced corn flakes than low-priced auto insurance.

If voters insist that auto insurance be priced below cost, it eventually must be sold by government. Stockholders can subsidize policyholders for a short period, but only taxpayers can subsidize them over the long term. At most property-casualty companies, socialized auto insurance would be no disaster for shareholders. Because of the

commodity characteristics of the industry, most insurers earn mediocre returns and therefore have little or no economic goodwill to lose if they are forced by government to leave the auto insurance business. But GEICO, because it is a low-cost producer able to earn high returns on equity, has a huge amount of economic goodwill at risk. In turn, so do we.

We just continuously search among them for the highest after-tax returns as measured by “mathematical expectation,” limiting ourselves always to investment alternatives we think we understand. Our criteria have nothing to do with maximizing immediately reportable earnings; our goal, rather, is to maximize eventual net worth.

However, a CEO who doesn’t perform is frequently carried indefinitely. One reason is that performance standards for his job seldom exist. When they do, they are often fuzzy or they may be waived or explained away, even when the performance shortfalls are major and repeated. At too many companies, the boss shoots the arrow of managerial performance and then hastily paints the bullseye around the spot where it lands.

when we own portions of outstanding businesses with outstanding managements, our favorite holding period is forever. We are just the opposite of those who hurry to sell and book profits when companies perform well but who tenaciously hang on to businesses that disappoint. Peter Lynch aptly likens such behavior to cutting the flowers and watering the weeds.

We have not lost our aversion to long-term bonds. We will become enthused about such securities only when we become enthused about prospects for long-term stability in the purchasing power of money. And that kind of stability isn’t in the cards: Both society and elected officials simply have too many higher-ranking priorities that conflict with purchasing-power stability.

Once, the word applied only to the simultaneous purchase and sale of securities or foreign exchange in two different markets. The goal was to exploit tiny price differentials that might exist between, say, Royal Dutch stock trading in guilders in Amsterdam, pounds in London, and dollars in New York. Some people might call this scalping; it won’t surprise you that practitioners opted for the French term, arbitrage.

Since World War I the definition of arbitrage—or “risk arbitrage,” as it is now sometimes called—has expanded to include the pursuit of profits from an announced corporate event such as sale of the company, merger, recapitalization, reorganization, liquidation, self-tender, etc. In most cases the arbitrageur expects to profit regardless of the behavior of the stock market. The major risk he usually faces instead is that the announced event won’t happen.

In recent years, most arbitrage operations have involved takeovers, friendly and unfriendly. With acquisition fever rampant, with anti-trust challenges almost non-existent, and with bids often ratcheting upward, arbitrageurs have prospered mightily.

To evaluate arbitrage situations you must answer four questions: (1) How likely is it that the promised event will indeed occur? (2) How long will your money be tied up? (3) What chance is there that something still better will transpire—a competing takeover bid, for example? and (4) What will happen if the event does not take place because of anti-trust action, financing glitches, etc.?

Arcata Corp., one of our more serendipitous arbitrage experiences, illustrates the twists and turns of the business. On September 28, 1981 the directors of Arcata agreed in principle to sell the company to Kohlberg, Kravis, Roberts & Co. (KKR), then and now a major leveraged-buy out firm. Arcata was in the printing and forest products businesses and had one other thing going for it: In 1978 the U.S. Government had taken title to 10,700 acres of Arcata timber, primarily old-growth redwood, to expand Redwood National Park. The government had paid $97.9 million, in several installments, for this acreage, a sum Arcata was contesting as grossly inadequate. The parties also disputed the interest rate that should apply to the period between the taking of the property and final payment for it. The enabling legislation stipulated 6% simple interest; Arcata argued for a much higher and compounded rate. Buying a company with a highly-speculative, large-sized claim in litigation creates a negotiating problem, whether the claim is on behalf of or against the company. To solve this problem, KKR offered $37.00 per Arcata share plus two-thirds of any additional amounts paid by the government for the redwood lands. Appraising this arbitrage opportunity, we had to ask ourselves whether KKR would consummate the transaction since, among other things, its offer was contingent upon its obtaining “satisfactory financing.” A clause of this kind is always dangerous for the seller: It offers an eas...

The other way we differ from some arbitrage operations is that we participate only in transactions that have been publicly announced. We do not trade on rumors or try to guess takeover candidates. We just read the newspapers, think about a few of the big propositions, and go by our own sense of probabilities.

One pleasant reason is that our cash holdings are down—because our position in equities that we expect to hold for a very long time is substantially up. As regular readers of this report know, our new commitments are not based on a judgment about short-term prospects for the stock market. Rather, they reflect an opinion about long-term business prospects for specific companies. We do not have, never have had, and never will have an opinion about where the stock market, interest rates, or business activity will be a year from now.

We have no idea how long the excesses will last, nor do we know what will change the attitudes of government, lender and buyer that fuel them. But we do know that the less the prudence with which others conduct their affairs, the greater the prudence with which we should conduct our own affairs. We have no desire to arbitrage transactions that reflect the unbridled—and, in our view, often unwarranted—optimism of both buyers and lenders. In our activities, we will heed the wisdom of Herb Stein: “If something can’t go on forever, it will end.” Efficient

Amazingly, EMT was embraced not only by academics, but by many investment professionals and corporate managers as well. Observing correctly that the market was frequently efficient, they went on to conclude incorrectly that it was always efficient. The difference between these propositions is night and day.

Over the 63 years, the general market delivered just under a 10% annual return, including dividends. That means $1,000 would have grown to $405,000 if all income had been reinvested. A 20% rate of return, however, would have produced $97 million. That strikes us as a statistically-significant differential that might, conceivably, arouse one’s curiosity.

Naturally the disservice done students and gullible investment professionals who have swallowed EMT has been an extraordinary service to us and other followers of Graham. In any sort of a contest—financial, mental, or physical—it’s an enormous advantage to have opponents who have been taught that it’s useless to even try. From a selfish point of view, Grahamites should probably endow chairs to ensure the perpetual teaching of EMT.

In two respects our goals probably differ somewhat from those of most listed companies. First, we do not want to maximize the price at which Berkshire shares trade. We wish instead for them to trade in a narrow range centered at intrinsic business value (which we hope increases at a reasonable—or, better yet, unreasonable—rate). Charlie and I are bothered as much by significant overvaluation as significant undervaluation. Both extremes will inevitably produce results for many shareholders that will differ sharply from Berkshire’s business results. If our stock price instead consistently mirrors business value, each of our shareholders will receive an investment result that roughly parallels the business results of Berkshire during his holding period.

Second, we wish for very little trading activity. If we ran a private business with a few passive partners, we would be disappointed if those partners, and their replacements, frequently wanted to leave the partnership. Running a public company, we feel the same way.

Our interest in new ventures, turnarounds, or auction-like sales can best be expressed by another Goldwynism: “Please include me out.”

We received some good news a few weeks ago: Standard & Poor’s raised our credit rating to AAA, which is the highest rating it bestows. Only 15 other U.S. industrial or property-casualty companies are rated AAA, down from 28 in 1980. Corporate bondholders have taken their lumps in the past few years from “event risk.” This term refers to the overnight degradation of credit that accompanies a heavily-leveraged purchase or recapitalization of a business whose financial policies, up to then, had been conservative. In a world of takeovers inhabited by few owner-managers, most corporations present such a risk. Berkshire does not. Charlie and I promise bondholders the same respect we afford shareholders.

Our gain in net worth during 1989 was $1.515 billion, or 44.4%. Over the last 25 years (that is, since present management took over) our per-share book value has grown from $19.46 to $4,296.01, or at a rate of 23.8% compounded annually.

What counts, however, is intrinsic value—the figure indicating what all of our constituent businesses are rationally worth. With perfect foresight, this number can be calculated by taking all future cash flows of a business—in and out—and discounting them at prevailing interest rates. So valued, all businesses, from manufacturers of buggy whips to operators of cellular phones, become economic equals.

We face another obstacle: In a finite world, high growth rates must self-destruct. If the base from which the growth is taking place is tiny, this law may not operate for a time. But when the base balloons, the party ends: A high growth rate eventually forges its own anchor. Carl Sagan has entertainingly described this phenomenon, musing about the destiny of bacteria that reproduce by dividing into two every 15 minutes. Says Sagan: “That means four doublings an hour, and 96 doublings a day. Although a bacterium weighs only about a trillionth of a gram, its descendants, after a day of wild asexual abandon, will collectively weigh as much as a mountain…in two days, more than the sun—and before very long, everything in the universe will be made of bacteria.” Not to worry, says Sagan: Some obstacle always impedes this kind of exponential growth. “The bugs run out of food, or they poison each other, or they are shy about reproducing in public.”

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.

Our 1989 gain of $1.5 billion was achieved after we took a charge of about $712 million for income taxes.

Of this year’s tax charge, about $172 million will be paid currently; the remainder, $540 million, is deferred. Almost all of the deferred portion relates to the 1989 increase in unrealized profits in our common stock holdings. Against this increase, we have reserved a 34% tax.

As you can see from our balance sheet on page 27, we would owe taxes of more than $1.1 billion were we to sell all of our securities at year-end market values. Is this $1.1 billion liability equal, or even similar, to a $1.1 billion liability payable to a trade creditor 15 days after the end of the year? Obviously not—despite the fact that both items have exactly the same effect on audited net worth, reducing it by $1.1 billion. On the other hand, is this liability for deferred taxes a meaningless accounting fiction because its payment can be triggered only by the sale of stocks that, in very large part, we have no intention of selling? Again, the answer is no. In economic terms, the liability resembles an interest-free loan from the U.S. Treasury that comes due only at our election (unless, of course, Congress moves to tax gains before they are realized). This “loan” is peculiar in other respects as well: It can be used only to finance the ownership of the particular, appreciated stocks and it fluctuates in size—daily as market prices change and periodically if tax rates change. 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.

Because of the way the tax law works, the Rip Van Winkle style of investing that we favor—if successful—has an important mathematical edge over a more frenzied approach. Let’s look at an extreme comparison. Imagine that Berkshire had only $1, which we put in a security that doubled by yearend and was then sold. Imagine further that we used the after-tax proceeds to repeat this process in each of the next 19 years, scoring a double each time. At the end of the 20 years, the 34% capital gains tax that we would have paid on the profits from each sale would have delivered about $13,000 to the government and we would be left with about $25,250. Not bad. If, however, we made a single fantastic investment that itself doubled 20 times during the 20 years, our dollar would grow to $1,048,576. Were we then to cash out, we would pay a 34% tax of roughly $356,500 and be left with about $692,000. The sole reason for this staggering difference in results would be the timing of tax payments. Interestingly, the government would gain from Scenario 2 in exactly the same 27:1 ratio as we—taking in taxes of $356,500 vs. $13,000—though, admittedly, it would have to wait for its money.

The table below shows the major sources of Berkshire’s reported earnings. In this presentation, amortization of Goodwill and other major purchase-price accounting adjustments are not charged against the specific businesses to which they apply, but are instead aggregated and shown separately. This procedure lets you view the earnings of our businesses as they would have been reported had we not purchased them. I’ve explained in past reports why this form of presentation seems to us to be more useful to investors and managers than one utilizing generally accepted accounting principles (GAAP), which require purchase-price adjustments to be made on a business-by-business basis.

The question you must decide is whether these undistributed earnings are as valuable to us as those we report. We believe they are—and even think they may be more valuable. The reason for this a-bird-in-the-bush-may-be-worth-two-in-the-hand conclusion is that earnings retained by these investees will be deployed by talented, owner-oriented managers who sometimes have better uses for these funds in their own businesses than we would have in ours. I would not make such a generous assessment of most managements, but it is appropriate in these cases.

page 49, illustrate just how outstanding their performance is. On an historical accounting basis, after-tax earnings of these operations were 57% on average equity capital. Moreover, this return was achieved with no net leverage: Cash equivalents have matched funded debt.

We bought the business without an audit, and all of our surprises have been on the plus side. “If you don’t know jewelry, know your jeweler” makes sense whether you are buying the whole business or a tiny diamond.

NFM and Borsheim’s follow precisely the same formula for success: (1) unparalleled depth and breadth of merchandise at one location; (2) the lowest operating costs in the business; (3) the shrewdest of buying, made possible in part by the huge volumes purchased; (4) gross margins, and therefore prices, far below competitors’; and (5) friendly personalized service with family members on hand at all times.

The Buffalo News continued to star in 1989 in three important ways: First, among major metropolitan papers, both daily and Sunday, the News is number one in household penetration—the percentage of local households that purchase it each day. Second, in “news hole”—the portion of the paper devoted to news—the paper stood at 50.1% in 1989 vs. 49.5% in 1988, a level again making it more news-rich than any comparable American paper. Third, in a year that saw profits slip at many major papers, the News set its seventh consecutive profit record. To some extent, these three factors are related, though obviously a high-percentage news hole, by itself, reduces profits significantly. A large and intelligently-utilized news hole, however, attracts a wide spectrum of readers and thereby boosts penetration. High penetration, in turn, makes a newspaper particularly valuable to retailers since it allows them to talk to the entire community through a single “megaphone.” A low-penetration paper is a far less compelling purchase for many advertisers and will eventually suffer in both ad rates and profits.

Aggregate earnings of these businesses have increased every year since our purchase and returns on invested capital continue to be exceptional.

The combined ratio represents total insurance costs (losses incurred plus expenses) compared to revenue from premiums: A ratio below 100 indicates an underwriting profit, and one above 100 indicates a loss. When the investment income that an insurer earns from holding policyholders’ funds (“the float”) is taken into account, a combined ratio in the 107-111 range typically produces an overall breakeven result, exclusive of earnings on the funds provided by shareholders.

The cartel period is long gone. Now the industry has hundreds of participants selling a commodity-like product at independently-established prices. Such a configuration—whether the product being sold is steel or insurance policies—is certain to cause subnormal profitability in all circumstances but one: a shortage of usable capacity. Just how often these periods occur and how long they last determines the average profitability of the industry in question.

The industry will meantime say it needs higher prices to achieve profitability matching that of the average American business. Of course it does. So does the steel business. But needs and desires have nothing to do with the long-term profitability of industries. Instead, economic fundamentals determine the outcome. Insurance profitability will improve only when virtually all insurers are turning away business despite higher prices. And we’re a long way from that point.

But because the 1989 disasters left many insurers either actually or possibly bare, and also left most CAT writers licking their wounds, there was an immediate shortage after the earthquake of much-needed catastrophe coverage. Prices instantly became attractive, particularly for the reinsurance that CAT writers themselves buy. Just as instantly, Berkshire Hathaway offered to write up to $250 million of catastrophe coverage, advertising that proposition in trade publications. Though we did not write all the business we sought, we did in a busy ten days book a substantial amount.

Our method of operation, incidentally, makes us a stabilizing force in the industry. We add huge capacity when capacity is short and we become less competitive only when capacity is abundant. Of course, we don’t follow this policy in the interest of stabilization—we follow it because we believe it to be the most sensible and profitable course of action. Nevertheless, our behavior steadies the market. In this case, Adam Smith’s invisible hand works as advertised.

Our holdings of Coca-Cola increased from 14,172,500 shares at the end of 1988 to 23,350,000. This Coca-Cola investment provides yet another example of the incredible speed with which your Chairman responds to investment opportunities, no matter how obscure or well-disguised they may be. I believe I had my first Coca-Cola in either 1935 or 1936. Of a certainty, it was in 1936 that I started buying Cokes at the rate of six for 25 cents from Buffett & Son, the family grocery store, to sell around the neighborhood for 5 cents each. In this excursion into high-margin retailing, I duly observed the extraordinary consumer attractiveness and commercial possibilities of the product. I continued to note these qualities for the next 52 years as Coke blanketed the world. During this period, however, I carefully avoided buying even a single share, instead allocating major portions of my net worth to street railway companies, windmill manufacturers, anthracite producers, textile businesses, trading-stamp issuers, and the like. (If you think I’m making this up, I can supply the names.) Only in the summer of 1988 did my brain finally establish contact with my eyes.

As I mentioned earlier, the yearend prices of our major investees were much higher relative to their intrinsic values than theretofore. While those prices may not yet cause nosebleeds, they are clearly vulnerable to a general market decline. A drop in their prices would not disturb us at all—it might in fact work to our eventual benefit—but it would cause at least a one-year reduction in Berkshire’s net worth. We think such a reduction is almost certain in at least one of the next three years.