• 1986年致股东信
  • • • 报告盈利来源
      • Sources of Reported Earnings
      • 费奇海默兄弟公司
      • The Fechheimer Bros. Co.
      • 保险业务
      • Insurance Operations
      • NHP公司
      • NHP, Inc.
      • 税收
      • Taxation
      • 杂项
      • Miscellaneous
    • 购买价格会计调整与“现金流”谬误
    • Purchase-Price Accounting Adjustments and the “Cash Flow” Fallacy

To the Shareholders of Berkshire Hathaway Inc.:

Our gain in net worth during 1986 was $492.5 million, or 26.1%. Over the last 22 years (that is, since present management took over), our per-share book value has grown from $19.46 to $2,073.06, or 23.3% compounded annually. Both the numerator and denominator are important in the per-share book value calculation: during the 22-year period our corporate net worth has increased 10,600% while shares outstanding have increased less than 1%.

In past reports I have noted that book value at most companies differs widely from intrinsic business value—the number that really counts for owners. In our own case, however, book value has served for more than a decade as a reasonable if somewhat conservative proxy for business value. That is, our business value has moderately exceeded our book value, with the ratio between the two remaining fairly steady.

The good news is that in 1986 our percentage gain in business value probably exceeded the book value gain. I say “probably” because business value is a soft number: in our own case, two equally well-informed observers might make judgments more than 10% apart.

A large measure of our improvement in business value relative to book value reflects the outstanding performance of key managers at our major operating businesses. These managers—the Blumkins, Mike Goldberg, the Heldmans, Chuck Huggins, Stan Lipsey, and Ralph Schey—have over the years improved the earnings of their businesses dramatically while, except in the case of insurance, utilizing little additional capital. This accomplishment builds economic value, or “Goodwill,” that does not show up in the net worth figure on our balance sheet, nor in our per-share book value. In 1986 this unrecorded gain was substantial.

So much for the good news. The bad news is that my performance did not match that of our managers. While they were doing a superb job in running our businesses, I was unable to skillfully deploy much of the capital they generated. Charlie Munger, our Vice Chairman, and I really have only two jobs. One is to attract and keep outstanding managers to run our various operations. This hasn’t been all that difficult. Usually the managers came with the companies we bought, having demonstrated their talents throughout careers that spanned a wide variety of business circumstances. They were managerial stars long before they knew us, and our main contribution has been to not get in their way. This approach seems elementary: if my job were to manage a golf team—and if Jack Nicklaus or Arnold Palmer were willing to play for me—neither would get a lot of directives from me about how to swing.

Some of our key managers are independently wealthy (we hope they all become so), but that poses no threat to their continued interest: they work because they love what they do and relish the thrill of outstanding performance. They unfailingly think like owners (the highest compliment we can pay a manager) and find all aspects of their business absorbing.

(Our prototype for occupational fervor is the Catholic tailor who used his small savings of many years to finance a pilgrimage to the Vatican. When he returned, his parish held a special meeting to get his first-hand account of the Pope. “Tell us,” said the eager faithful, “just what sort of fellow is he?” Our hero wasted no words: “He’s a forty-four, medium.”)

Charlie and I know that the right players will make almost any team manager look good. We subscribe to the philosophy of Ogilvy & Mather’s founding genius, David Ogilvy: “If each of us hires people who are smaller than we are, we shall become a company of dwarfs. But, if each of us hires people who are bigger than we are, we shall become a company of giants.”

A by-product of our managerial style is the ability it gives us to easily expand Berkshire’s activities. We’ve read management treatises that specify exactly how many people should report to any one executive, but they make little sense to us. When you have able managers of high character running businesses about which they are passionate, you can have a dozen or more reporting to you and still have time for an afternoon nap. Conversely, if you have even one person reporting to you who is deceitful, inept or uninterested, you will find yourself with more than you can handle. Charlie and I could work with double the number of managers we now have, so long as they had the rare qualities of the present ones.

We intend to continue our practice of working only with people whom we like and admire. This policy not only maximizes our chances for good results, it also ensures us an extraordinarily good time. On the other hand, working with people who cause your stomach to churn seems much like marrying for money—probably a bad idea under any circumstances, but absolute madness if you are already rich.

The second job Charlie and I must handle is the allocation of capital, which at Berkshire is a considerably more important challenge than at most companies. Three factors make that so: we earn more money than average; we retain all that we earn; and, we are fortunate to have operations that, for the most part, require little incremental capital to remain competitive and to grow. Obviously, the future results of a business earning 23% annually and retaining it all are far more affected by today’s capital allocations than are the results of a business earning 10% and distributing half of that to shareholders. If our retained earnings—and those of our major investees, GEICO and Capital Cities/ABC, Inc.—are employed in an unproductive manner, the economics of Berkshire will deteriorate very quickly. In a company adding only, say, 5% to net worth annually, capital-allocation decisions, though still important, will change the company’s economics far more slowly.

Capital allocation at Berkshire was tough work in 1986. We did make one business acquisition—The Fechheimer Bros. Company, which we will discuss in a later section. Fechheimer is a company with excellent economics, run by exactly the kind of people with whom we enjoy being associated. But it is relatively small, utilizing only about 2% of Berkshire’s net worth.

Meanwhile, we had no new ideas in the marketable equities field, an area in which once, only a few years ago, we could readily employ large sums in outstanding businesses at very reasonable prices. So our main capital allocation moves in 1986 were to pay off debt and stockpile funds. Neither is a fate worse than death, but they do not inspire us to do handsprings either. If Charlie and I were to draw blanks for a few years in our capital-allocation endeavors, Berkshire’s rate of growth would slow significantly.

We will continue to look for operating businesses that meet our tests and, with luck, will acquire such a business every couple of years. But an acquisition will have to be large if it is to help our performance materially. Under current stock market conditions, we have little hope of finding equities to buy for our insurance companies. Markets will change significantly—you can be sure of that and some day we will again get our turn at bat. However, we haven’t the faintest idea when that might happen.

It can’t be said too often (although I’m sure you feel I’ve tried) that, even under favorable conditions, our returns are certain to drop substantially because of our enlarged size. We have told you that we hope to average a return of 15% on equity and we maintain that hope, despite some negative tax law changes described in a later section of this report. If we are to achieve this rate of return, our net worth must increase $7.2 billion in the next ten years. A gain of that magnitude will be possible only if, before too long, we come up with a few very big (and good) ideas. Charlie and I can’t promise results, but we do promise you that we will keep our efforts focused on our goals.

Sources of Reported Earnings

The table on the next page shows the major sources of Berkshire’s reported earnings. This table differs in several ways from the one presented last year. We have added four new lines of business because of the Scott Fetzer and Fechheimer acquisitions. In the case of Scott Fetzer, the two major units acquired were World Book and Kirby, and each is presented separately. Fourteen other businesses of Scott Fetzer are aggregated in Scott Fetzer—Diversified Manufacturing. SF Financial Group, a credit company holding both World Book and Kirby receivables, is included in “Other.” This year, because Berkshire is so much larger, we also have eliminated separate reporting for several of our smaller businesses.

In the table, amortization of Goodwill is not charged against the specific businesses but, for reasons outlined in the Appendix to my letter in the 1983 Annual Report, is aggregated as a separate item. (A Compendium of earlier letters, including the Goodwill discussion, is available upon request.) Both the Scott Fetzer and Fechheimer acquisitions created accounting Goodwill, which is why the amortization charge for Goodwill increased in 1986.

Additionally, the Scott Fetzer acquisition required other major purchase-price accounting adjustments, as prescribed by generally accepted accounting principles (GAAP). The GAAP figures, of course, are the ones used in our consolidated financial statements. But, in our view, the GAAP figures are not necessarily the most useful ones for investors or managers. Therefore, the figures shown for specific operating units are earnings before purchase-price adjustments are taken into account. In effect, these are the earnings that would have been reported by the businesses if we had not purchased them.

A discussion of our reasons for preferring this form of presentation is in the Appendix to this letter. This Appendix will never substitute for a steamy novel and definitely is not required reading. However, I know that among our 6,000 shareholders there are those who are thrilled by my essays on accounting—and I hope that both of you enjoy the Appendix.

In the Business Segment Data on pages 41-43 and in the Management’s Discussion section on pages 45-49, you will find much additional information about our businesses. I urge you to read those sections, as well as Charlie Munger’s letter to Wesco shareholders, describing the various businesses of that subsidiary, which starts on page 50.

 As you can see, operating earnings substantially improved during 1986. Some of the improvement came from the insurance operation, whose results I will discuss in a later section. Fechheimer also will be discussed separately. Our other major businesses performed as follows:

• Operating results at The Buffalo News continue to reflect a truly superb managerial job by Stan Lipsey. For the third year in a row, man-hours worked fell significantly and other costs were closely controlled. Consequently, our operating margins improved materially in 1986, even though our advertising rate increases were well below those of most major newspapers.

  Our cost-control efforts have in no way reduced our commitment to news. We continue to deliver a 50% “news hole” (the portion of the total space in the paper devoted to news), a higher percentage, we believe, than exists at any dominant newspaper in this country of our size or larger.

  The average news hole at papers comparable to the News is about 40%. The difference between 40% and 50% is more important than it might first seem: a paper with 30 pages of ads and a 40% news hole delivers 20 pages of news a day, whereas our paper matches 30 pages of ads with 30 pages of news. Therefore, given ad pages equal in number, we end up delivering our readers no less than 50% more news.

  We believe this heavy commitment to news is one of the reasons The Buffalo News has the highest weekday penetration rate (the percentage of households in the paper’s primary marketing area purchasing it each day) among any of the top 50 papers in the country. Our Sunday penetration, where we are also number one, is even more impressive. Ten years ago, the only Sunday paper serving Buffalo (the Courier-Express) had circulation of 271,000 and a penetration ratio of about 63%. The Courier-Express had served the area for many decades and its penetration ratio—which was similar to those existing in many metropolitan markets—was thought to be a “natural” one, accurately reflecting the local citizenry’s appetite for a Sunday product.

  Our Sunday paper was started in late 1977. It now has a penetration ratio of 83% and sells about 100,000 copies more each Sunday than did the Courier-Express ten years ago—even though population in our market area has declined during the decade. In recent history, no other city that has long had a local Sunday paper has experienced a penetration gain anywhere close to Buffalo’s.

  Despite our exceptional market acceptance, our operating margins almost certainly have peaked. A major newsprint price increase took effect at the end of 1986, and our advertising rate increases in 1987 will again be moderate compared to those of the industry. However, even if margins should materially shrink, we would not reduce our news-hole ratio.

  As I write this, it has been exactly ten years since we purchased The News. The financial rewards it has brought us have far exceeded our expectations and so, too, have the non-financial rewards. Our respect for the News—high when we bought it—has grown consistently ever since the purchase, as has our respect and admiration for Murray Light, the editor who turns out the product that receives such extraordinary community acceptance. The efforts of Murray and Stan, which were crucial to the News during its dark days of financial reversals and litigation, have not in the least been lessened by prosperity. Charlie and I are grateful to them.

• The amazing Blumkins continue to perform business miracles at Nebraska Furniture Mart. Competitors come and go (mostly go), but Mrs. B. and her progeny roll on. In 1986 net sales increased 10.2% to $132 million. Ten years ago sales were $44 million and, even then, NFM appeared to be doing just about all of the business available in the Greater Omaha Area. Given NFM’s remarkable dominance, Omaha’s slow growth in population and the modest inflation rates that have applied to the goods NFM sells, how can this operation continue to rack up such large sales gains? The only logical explanation is that the marketing territory of NFM’s one-and-only store continues to widen because of its ever-growing reputation for rock-bottom everyday prices and the broadest of selections. In preparation for further gains, NFM is expanding the capacity of its warehouse, located a few hundred yards from the store, by about one-third.

  Mrs. B, Chairman of Nebraska Furniture Mart, continues at age 93 to outsell and out-hustle any manager I’ve ever seen. She’s at the store seven days a week, from opening to close. Competing with her represents a triumph of courage over judgment.

  It’s easy to overlook what I consider to be the critical lesson of the Mrs. B saga: at 93, Omaha based Board Chairmen have yet to reach their peak. Please file this fact away to consult before you mark your ballot at the 2024 annual meeting of Berkshire.

• At See’s, sales trends improved somewhat from those of recent years. Total pounds sold rose about 2%. (For you chocaholics who like to fantasize, one statistic: we sell over 12,000 tons annually.) Same-store sales, measured in pounds, were virtually unchanged. In the previous six years, same store poundage fell, and we gained or maintained poundage volume only by adding stores. But a particularly strong Christmas season in 1986 stemmed the decline. By stabilizing same-store volume and making a major effort to control costs, See’s was able to maintain its excellent profit margin in 1986 though it put through only minimal price increases. We have Chuck Huggins, our long-time manager at See’s, to thank for this significant achievement.

  See’s has a one-of-a-kind product “personality” produced by a combination of its candy’s delicious taste and moderate price, the company’s total control of the distribution process, and the exceptional service provided by store employees. Chuck rightfully measures his success by the satisfaction of our customers, and his attitude permeates the organization. Few major retailing companies have been able to sustain such a customer-oriented spirit, and we owe Chuck a great deal for keeping it alive and well at See’s.

  See’s profits should stay at about their present level. We will continue to increase prices very modestly, merely matching prospective cost increases.

• World Book is the largest of 17 Scott Fetzer operations that joined Berkshire at the beginning of 1986. Last year I reported to you enthusiastically about the businesses of Scott Fetzer and about Ralph Schey, its manager. A year’s experience has added to my enthusiasm for both. Ralph is a superb businessman and a straight shooter. He also brings exceptional versatility and energy to his job: despite the wide array of businesses that he manages, he is on top of the operations, opportunities and problems of each. And, like our other managers, Ralph is a real pleasure to work with. Our good fortune continues.

  World Book’s unit volume increased for the fourth consecutive year, with encyclopedia sales up 7% over 1985 and 45% over 1982. Childcraft’s unit sales also grew significantly.

  World Book continues to dominate the U.S. direct-sales encyclopedia market—and for good reasons. Extraordinarily well-edited and priced at under 5 cents per page, these books are a bargain for youngster and adult alike. You may find one editing technique interesting: World Book ranks over 44,000 words by difficulty. Longer entries in the encyclopedia include only the most easily comprehended words in the opening sections, with the difficulty of the material gradually escalating as the exposition proceeds. As a result, youngsters can easily and profitably read to the point at which subject matter gets too difficult, instead of immediately having to deal with a discussion that mixes up words requiring college-level comprehension with others of fourth-grade level.

  Selling World Book is a calling. Over one-half of our active salespeople are teachers or former teachers, and another 5% have had experience as librarians. They correctly think of themselves as educators, and they do a terrific job. If you don’t have a World Book set in your house, I recommend one.

• Kirby likewise recorded its fourth straight year of unit volume gains. Worldwide, unit sales grew 4% from 1985 and 33% from 1982. While the Kirby product is more expensive than most cleaners, it performs in a manner that leaves cheaper units far behind (“in the dust,” so to speak). Many 30- and 40-year-old Kirby cleaners are still in active duty. If you want the best, you buy a Kirby.

  Some companies that historically have had great success in direct sales have stumbled in recent years. Certainly the era of the working woman has created new challenges for direct sales organizations. So far, the record shows that both Kirby and World Book have responded most successfully.

The businesses described above, along with the insurance operation and Fechheimer, constitute our major business units. The brevity of our descriptions is in no way meant to diminish the importance of these businesses to us. All have been discussed in past annual reports and, because of the tendency of Berkshire owners to stay in the fold (about 98% of the stock at the end of each year is owned by people who were owners at the start of the year), we want to avoid undue repetition of basic facts. You can be sure that we will immediately report to you in detail if the underlying economics or competitive position of any of these businesses should materially change. In general, the businesses described in this section can be characterized as having very strong market positions, very high returns on capital employed, and the best of operating managements.

The Fechheimer Bros. Co.

Every year in Berkshire’s annual report I include a description of the kind of business that we would like to buy. This “ad” paid off in 1986.

On January 15th of last year I received a letter from Bob Heldman of Cincinnati, a shareholder for many years and also Chairman of Fechheimer Bros. Until I read the letter, however, I did not know of either Bob or Fechheimer. Bob wrote that he ran a company that met our tests and suggested that we get together, which we did in Omaha after their results for 1985 were compiled.

He filled me in on a little history: Fechheimer, a uniform manufacturing and distribution business, began operations in 1842. Warren Heldman, Bob’s father, became involved in the business in 1941 and his sons, Bob and George (now President), along with their sons, subsequently joined the company. Under the Heldmans’ management, the business was highly successful.

In 1981 Fechheimer was sold to a group of venture capitalists in a leveraged buy out (an LBO), with management retaining an equity interest. The new company, as is the case with all LBOS, started with an exceptionally high debt/equity ratio. After the buy out, however, operations continued to be very successful. So by the start of last year debt had been paid down substantially and the value of the equity had increased dramatically. For a variety of reasons, the venture capitalists wished to sell and Bob, having dutifully read Berkshire’s annual reports, thought of us.

Fechheimer is exactly the sort of business we like to buy. Its economic record is superb; its managers are talented, high-grade, and love what they do; and the Heldman family wanted to continue its financial interest in partnership with us. Therefore, we quickly purchased about 84% of the stock for a price that was based upon a $55 million valuation for the entire business.

The circumstances of this acquisition were similar to those prevailing in our purchase of Nebraska Furniture Mart: most of the shares were held by people who wished to employ funds elsewhere; family members who enjoyed running their business wanted to continue both as owners and managers; several generations of the family were active in the business, providing management for as far as the eye can see; and the managing family wanted a purchaser who would not re-sell, regardless of price, and who would let the business be run in the future as it had been in the past. Both Fechheimer and NFM were right for us, and we were right for them.

You may be amused to know that neither Charlie nor I have been to Cincinnati, headquarters for Fechheimer, to see their operation. (And, incidentally, it works both ways: Chuck Huggins, who has been running See’s for 15 years, has never been to Omaha.) If our success were to depend upon insights we developed through plant inspections, Berkshire would be in big trouble. Rather, in considering an acquisition, we attempt to evaluate the economic characteristics of the business—its competitive strengths and weaknesses—and the quality of the people we will be joining. Fechheimer was a standout in both respects. In addition to Bob and George Heldman, who are in their mid-60s—spring chickens by our standards—there are three members of the next generation, Gary, Roger and Fred, to insure continuity.

As a prototype for acquisitions, Fechheimer has only one drawback: size. We hope our next acquisition is at least several times as large but a carbon copy in all other respects. Our threshold for minimum annual after-tax earnings of potential acquisitions has been moved up to $10 million from the $5 million level that prevailed when Bob wrote to me.

Flushed with success, we repeat our ad. If you have a business that fits, call me or, preferably, write.

Here’s what we’re looking for:

1. large purchases (at least $10 million of after-tax earnings),

2. demonstrated consistent earning power (future projections are of little interest to us, nor are “turn-around” situations),

3. businesses earning good returns on equity while employing little or no debt.

4. management in place (we can’t supply it),

5. simple businesses (if there’s lots of technology, we won’t understand it),

6. an offering price (we don’t want to waste our time or that of the seller by talking, even preliminarily, about a transaction when price is unknown).

We will not engage in unfriendly takeovers. We can promise complete confidentiality and a very fast answer—customarily within five minutes—as to whether we’re interested. We prefer to buy for cash, but will consider issuing stock when we receive as much in intrinsic business value as we give. Indeed, following recent advances in the price of Berkshire stock, transactions involving stock issuance may be quite feasible. We invite potential sellers to check us out by contacting people with whom we have done business in the past. For the right business—and the right people—we can provide a good home.

On the other hand, we frequently get approached about acquisitions that don’t come close to meeting our tests: new ventures, turnarounds, auction-like sales, and the ever-popular (among brokers) “I’m-sure-something-will-work-out-if-you-people-get-to-know-each-other.” None of these attracts us in the least.

* * *

Besides being interested in the purchases of entire businesses as described above, we are also interested in the negotiated purchase of large, but not controlling, blocks of stock, as in our Cap Cities purchase. Such purchases appeal to us only when we are very comfortable with both the economics of the business and the ability and integrity of the people running the operation. We prefer large transactions: in the unusual case we might do something as small as $50 million (or even smaller), but our preference is for commitments many times that size.

Insurance Operations

We present our usual table of industry figures, expanded this year to include data about incurred losses and the GNP inflation index. The contrast in 1986 between the growth in premiums and growth in incurred losses will show you why underwriting results for the year improved materially:

Source: Best’s Insurance Management Reports

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 on to policyholders’ funds (“the float”) is taken into account, a combined ratio in the 107-112 range typically produces an overall break-even result, exclusive of earnings on the funds provided by shareholders.

The math of the insurance business, encapsulated by the table, is not very complicated. In years when the industry’s annual gain in revenues (premiums) pokes along at 4% or 5%, underwriting losses are sure to mount. This is not because auto accidents, fires, windstorms and the like are occurring more frequently, nor has it lately been the fault of general inflation. Today, social and judicial inflation are the major culprits; the cost of entering a courtroom has simply ballooned. Part of the jump in cost arises from skyrocketing verdicts, and part from the tendency of judges and juries to expand the coverage of insurance policies beyond that contemplated by the insurer when the policies were written. Seeing no let-up in either trend, we continue to believe that the industry’s revenues must grow at close to 10% annually for it to just hold its own in terms of profitability, even though general inflation may be running only 2%—4%.

In 1986, as noted, the industry’s premium volume soared even faster than loss costs. Consequently, the underwriting loss of the industry fell dramatically. In last year’s report we predicted this sharp improvement but also predicted that prosperity would be fleeting. Alas, this second prediction is already proving accurate. The rate of gain in the industry’s premium volume has slowed significantly (from an estimated 27.1% in 1986’s first quarter, to 23.5% in the second, to 21.8% in the third, to 18.7% in the fourth), and we expect further slowing in 1987. Indeed, the rate of gain may well fall below my 10% “equilibrium” figure by the third quarter.

Nevertheless, underwriting results in 1987, assuming they are not dragged down by a major natural catastrophe, will again improve materially because price increases are recognized in revenues on a lagged basis. In effect, the good news in earnings follows the good news in prices by six to twelve months. But the improving trend in earnings will probably end by late 1988 or early 1989. Thereafter the industry is likely to head south in a hurry.

Pricing behavior in the insurance industry continues to be exactly what can be expected in a commodity-type business. Only under shortage conditions are high profits achieved, and such conditions don’t last long. When the profit sun begins to shine, long-established insurers shower investors with new shares in order to build capital. In addition, newly-formed insurers rush to sell shares at the advantageous prices available in the new-issue market (prices advantageous, that is, to the insiders promoting the company but rarely to the new shareholders). These moves guarantee future trouble: capacity soars, competitive juices flow, and prices fade.

It’s interesting to observe insurance leaders beseech their colleagues to behave in a more “statesmanlike” manner when pricing policies. “Why,” they ask, “can’t we learn from history, even out the peaks and valleys, and consistently price to make reasonable profits?” What they wish, of course, is pricing that resembles, say, that of The Wall Street journal, whose prices are ample to start with and rise consistently each year.

Such calls for improved behavior have all of the efficacy of those made by a Nebraska corn grower asking his fellow growers, worldwide, to market their corn with more statesmanship. What’s needed is not more statesmen, but less corn. By raising large amounts of capital in the last two years, the insurance industry has, to continue our metaphor, vastly expanded its plantings of corn. The resulting increase in “crop”—i.e., the proliferation of insurance capacity—will have the same effect on prices and profits that surplus crops have had since time immemorial.

Our own insurance operation did well in 1986 and is also likely to do well in 1987. We have benefited significantly from industry conditions. But much of our prosperity arises from the efforts and ability of Mike Goldberg, manager of all insurance operations.

Our combined ratio (on a statutory basis and excluding structured settlements and financial reinsurance) fell from 111 in 1985 to 103 in 1986. In addition, our premium growth has been exceptional: although final figures aren’t available, I believe that over the past two years we were the fastest growing company among the country’s top 100 insurers. Some of our growth, it is true, came from our large quota-share contract with Fireman’s Fund, described in last year’s report and updated in Charlie’s letter on page 54. But even if the premiums from that contract are excluded from the calculation, we probably still ranked first in growth.

Interestingly, we were the slowest-growing large insurer in the years immediately preceding 1985. In fact, we shrank—and we will do so again from time to time in the future. Our large swings in volume do not mean that we come and go from the insurance marketplace. Indeed, we are its most steadfast participant, always standing ready, at prices we believe adequate, to write a wide variety of high-limit coverages. The swings in our volume arise instead from the here-today, gone-tomorrow behavior of other insurers. When most insurers are “gone,” because their capital is inadequate or they have been frightened by losses, insureds rush to us and find us ready to do business. But when hordes of insurers are “here,” and are slashing prices far below expectable costs, many customers naturally leave us in order to take advantage of the bargains temporarily being offered by our competition.

Our firmness on prices works no hardship on the consumer: he is being bombarded by attractively priced insurance offers at those times when we are doing little business. And it works no hardship on our employees: we don’t engage in layoffs when we experience a cyclical slowdown at one of our generally-profitable insurance operations. This no-layoff practice is in our self-interest. Employees who fear that large layoffs will accompany sizable reductions in premium volume will understandably produce scads of business through thick and thin (mostly thin).

The trends in National Indemnity’s traditional business—the writing of commercial auto and general liability policies through general agents—suggest how gun-shy other insurers became for a while and how brave they are now getting. In the last quarter of 1984, NICO’s monthly volume averaged $5 million, about what it had been running for several years. By the first quarter of 1986, monthly volume had climbed to about $35 million. In recent months, a sharp decline has set in. Monthly volume is currently about $20 million and will continue to fall as new competitors surface and prices are cut. Ironically, the managers of certain major new competitors are the very same managers that just a few years ago bankrupted insurers that were our old competitors. Through state-mandated guaranty funds, we must pay some of the losses these managers left unpaid, and now we find them writing the same sort of business under a new name. C’est la guerre.

The business we call “large risks” expanded significantly during 1986, and will be important to us in the future. In this operation, we regularly write policies with annual premiums of $1—$3 million, or even higher. This business will necessarily be highly volatile—both in volume and profitability—but our premier capital position and willingness to write large net lines make us a very strong force in the market when prices are right. On the other hand, our structured settlement business has become near-dormant because present prices make no sense to us.

The 1986 loss reserve development of our insurance group is chronicled on page 46. The figures show the amount of error in our yearend 1985 liabilities that a year of settlements and further evaluation has revealed. As you can see, what I told you last year about our loss liabilities was far from true—and that makes three years in a row of error. If the physiological rules that applied to Pinocchio were to apply to me, my nose would now draw crowds.

When insurance executives belatedly establish proper reserves, they often speak of “reserve strengthening,” a term that has a rather noble ring to it. They almost make it sound as if they are adding extra layers of strength to an already-solid balance sheet. That’s not the case: instead the term is a euphemism

尽管商业和金融界热衷于活跃交易，我们将坚持我们的‘至死不分离’政策。这是查理和我唯一感到舒适的政策，它产生了体面的结果，并让我们的经理人和被投资公司的管理者们能够心无旁骛地经营业务。

NHP公司

去年我们支付了2370万美元购买了NHP公司约50%的股份，该公司是一家多户租赁住房的开发商、联合体组织者、业主和管理者。如果所有已授权的管理层股票期权都被授予并行使，我们的股权比例将降至略高于45%。

NHP公司拥有一个极不寻常的起源。1967年，约翰逊总统任命了一个由商界和民间领袖组成的委员会，由埃德加·凯泽领导，研究增加为中低收入租户提供多户住宅供应的方法。该委员会的某些成员随后组建并推广了两个商业实体以促进这一目标。这两个实体现在都由NHP公司拥有，其中一个按照不同寻常的基本规则运营：其三名董事必须由总统任命并经过参议院咨询和同意，同时法律要求其向总统提交年度报告。

超过260家大型公司，其动机更多是出于公共服务而非利润，向这两个原始实体投资了4200万美元，它们随即通过合伙关系开始开发政府补贴的租赁房产。典型的合伙企业拥有单个房产，主要通过无追索权抵押贷款融资。每个合伙企业的多数股权资金由一群有限合伙人提供，他们主要被投资带来的巨额税收减免所吸引。NHP担任普通合伙人，并购买了每个合伙企业的一小部分股权。

政府的住房政策当然已经转变，NHP不可避免地将其业务扩展到包括收取市场租金的无补贴公寓。此外，NHP的一家子公司在华盛顿特区地区建造独户住宅，年收入约为5000万美元。

NHP目前监督着约500个合伙房产，这些房产分布在40个州、哥伦比亚特区和波多黎各，包括约80,000套住房单元。这些房产的成本超过25亿美元，并且维护良好。NHP直接管理约55,000套住房单元，并监督其余部分的管理工作。公司的管理收入每年约为1600万美元，并且还在增长。

除了在每个合伙企业成立时购买的股权权益外，NHP还拥有各种剩余权益，这些权益在处置财产并向有限合伙人进行分配时发挥作用。NHP许多“深度补贴”房产的剩余权益可能价值不大。但某些其他房产的剩余权益可能非常有价值，尤其是在通货膨胀升温的情况下。

1986年的《税收改革法案》已叫停了向个人进行房产的税收导向型联合。总的来说，NHP目前正试图在高质量、大规模（通常200至500套单元）的无补贴租赁房产中取得股权头寸或重要的剩余权益。在这类项目中，NHP通常与一个或多个大型机构投资者或贷款机构合作。NHP将继续寻求开发中低收入公寓住房的方式，但除非政府政策改变，否则不太可能成功。

除了我们自己，NHP的大股东是惠好公司（持股约25%）以及由NHP首席执行官罗德·海勒领导的管理团队。还有约60家大型公司继续持有少量股份，均未超过2%。

税收

1986年的《税收改革法案》以重要且不同的方式影响着我们的各项业务。尽管我们发现该法案有许多值得称赞之处，但对伯克希尔来说，净财务影响是负面的：在新法律下，我们的企业价值增长率很可能会比在老法律下至少适度放缓。对我们股东的影响甚至更为负面：每股业务价值的每一美元增长，假设伴随着伯克希尔股票市场价值同等金额的增长，将为我们的股东带来72美分的税后收益，而不是老法律下的80美分。当然，这一结果反映了个人资本利得最高税率从20%上升到28%。

以下是影响伯克希尔的主要税收变化：

• 公司普通收入的税率计划从1986年的46%下降到1988年的34%。这一变化显然对我们有积极影响——并且对我们三大被投资公司中的两家——大都会/ABC公司和华盛顿邮报公司——也有显著的积极影响。

  我这样说，是知道多年来关于谁真正支付公司税——是企业还是它们的客户——一直存在许多模糊且往往带有党派色彩的评论。当然，争论通常围绕税收增加而非减少展开。那些抵制提高公司税率的人经常争辩说，公司实际上并未支付对其征收的任何税款，而是充当某种经济管道，将所有税款转嫁给消费者。根据这些倡导者的说法，任何公司税的增加都只会导致价格上涨，对公司而言，这抵消了增加的影响。采取这种立场后，“管道”理论的拥护者也必须得出结论，对公司减税不会帮助利润，而是会传导下去，导致消费者价格相应降低。

  相反，其他人则认为，公司不仅支付对其征收的税款，而且还吸收了这些税款。这个学派说，消费者将不会受到公司税率变化的影响。

  实际情况如何？当公司税率降低时，伯克希尔、华盛顿邮报、大都会等公司是自己吸收了利益，还是以降低价格的形式将这些利益转给了客户？这对投资者、管理者以及政策制定者来说都是一个重要问题。

  我们的结论是，在某些情况下，降低公司税的好处完全或几乎完全落入了公司及其股东的口袋；而在其他情况下，这些好处则完全或几乎完全转嫁给了客户。决定结果的因素是公司业务特许经营权的实力，以及该特许经营权的盈利能力是否受到监管。

  例如，当特许经营权强大且税后利润受到相对精确的监管时，就像电力公用事业那样，公司税率的变化在很大程度上反映在价格上，而不是利润上。当税收削减时，价格通常会在短期内降低。当税收增加时，价格会上升，尽管往往不那么迅速。

  类似的结果出现在第二个领域——价格竞争激烈的行业，这些行业的公司通常拥有非常薄弱的业务特许经营权。在这样的行业中，自由市场以延迟且不规则但通常有效的方式“调节”税后利润。实际上，市场在处理价格竞争行业时，其作用与公用事业委员会在处理电力公用事业时非常相似。因此，在这些行业中，税收变化最终更多地影响价格而非利润。

  然而，对于拥有强大特许经营权的未受监管企业来说，情况则不同：公司和其股东是减税的主要受益者。这些公司从减税中获益，很像电力公司如果缺乏监管机构来压低价格时受益的情况。

  我们的许多业务，无论是全资拥有的还是部分拥有的，都拥有此类特许经营权。因此，它们的税收削减最终大部分落入了我们的口袋，而不是我们客户的口袋。虽然这样说可能不合时宜，但无法否认。如果你倾向于相信相反的情况，请想一想你所在地区最优秀的外科医生或律师。既然最高个人税率正从50%降至28%，你真的期望这位专家（其专业领域的当地“特许经营权持有者”）的费用会降低吗？

  然而，我们对较低税率有利于我们许多运营企业和被投资公司的结论的喜悦，应被我们的另一个信念严重冲淡：计划中的1988年税率，无论是个人还是公司税率，对我们来说似乎完全不切实际。这些税率很可能会给华盛顿带来财政问题，最终被证明与价格稳定不相容。因此，我们相信，最终——比如说，五年内——要么更高的税率，要么更高的通胀率几乎肯定会成为现实。而且，如果两者同时出现，我们也不会感到惊讶。

• 公司资本利得税率已从28%提高到34%，自1987年生效。这一变化将对伯克希尔产生重大不利影响，因为我们预期未来我们业务价值增长的很大一部分，如同过去一样，将来自资本利得。例如，我们三大主要投资——大都会、GEICO和华盛顿邮报——在年底的市场价值超过17亿美元，接近伯克希尔总净值的75%，但它们每年仅为我们带来约900万美元的收入。相反，这三家公司都保留了非常高比例的收益，我们预计这些收益最终将为我们带来资本利得。

  新法提高了未来所有已实现收益的税率，包括法律颁布前存在的未实现收益。年底，我们在股权投资中拥有12亿美元的此类未实现收益。新法对我们资产负债表的影响将被延迟，因为GAAP规则规定，适用于未实现收益的递延税负债应按去年28%的税率而非当前的34%税率列报。预计该规则将很快改变。一旦改变，约7300万美元将从我们的GAAP净值中消失，并添加到递延税账户中。

• 新法下，我们的保险公司收到的股息和利息收入将被征收更重的税。首先，所有公司从其他国内公司收到的股息中，20%将被征税，高于旧法下的15%。其次，有一个仅适用于财产/意外伤害保险公司的关于剩余80%部分的修改：如果支付股息的股票是在1986年8月7日之后购买的，则该剩余部分的15%将被征税。第三项修改，同样仅适用于财产/意外伤害保险公司，涉及免税债券：保险公司在1986年8月7日之后购买的债券的利息将只有85%是免税的。

  最后两项修改非常重要。它们意味着我们在未来年份投资所获得的收入将显著低于旧法下的情况。我最好的猜测是，仅这些变化最终就会使我们的保险业务的盈利能力比我们以前可能预期的水平至少降低10%。

• 新税法还对财产/意外伤害保险公司纳税的时间安排进行了实质性修改。一项新规则要求我们在纳税申报表中对我们的损失准备金进行折现，这一变化将减少扣除额并增加应纳税所得额。另一项规则将在六年内逐步实施，要求我们将未赚保费准备金的20%纳入应纳税所得额。

  这两项规则都不会改变我们向你们报告的年度应计税额，但每一项都大幅加快了支付时间表。也就是说，以前递延的税款现在将被前置，这一变化将显著降低我们业务的盈利能力。一个类比可以说明其影响：如果你在21岁时被要求立即支付你在整个生命过程中应得的全部收入的税款，那么你的终生财富和遗产将仅是如果你所有收入税都去世时才需支付时的极小一部分。

  细心的读者可能会发现我们说法中的不一致之处。早些时候，在讨论处于价格竞争行业的公司时，我们提出税收增加或减少对这些公司影响相对较小，而是主要转嫁给了它们的客户。但现在我们说，税收增加将影响伯克希尔财产/意外伤害保险公司的利润，尽管它们也处于激烈价格竞争的行业。

  该行业可能成为我们一般规则例外情况的原因是，并非所有主要保险公司都将遵循相同的税务公式。存在重要差异的原因有几个：一项新的替代性最低税将对某些公司产生重大影响，而对另一些公司则不然；某些大型保险公司拥有巨额亏损结转，至少在几年内将基本保护其收入免受重大税收影响；一些大型保险公司的业绩将被并入拥有非保险业务的公司的合并报表。这些不同的情况将在财产/意外伤害保险行业产生差异巨大的边际税率。然而，在大多数其他价格竞争行业，如铝业、汽车业和百货业，主要参与者通常将面对相似的税务公式，情况则不会如此。

  财产/意外伤害保险公司缺乏共同的税务计算方法，意味着该行业面临的增税可能不会像在典型的竞争行业那样转嫁给客户。换句话说，保险公司自身将承担大部分新的税收负担。

• 对这些负担的部分抵消是1987年1月1日生效的“新起点”调整，届时我们1986年12月31日的损失准备金数字将为了税务目的转换为新要求的折现基础。（然而，在给你们的报告中，准备金将完全保持与过去相同的基础——不折现，除非是结构式结算等特殊情况。）“新起点”的净效果是给我们带来双重扣除：我们将在1987年及未来年份对我们已经作为成本在1986年及更早年份完全扣除的部分已发生但未支付的保险损失获得税收扣除。

  这一变化带来的净值增加尚未反映在我们的财务报表中。相反，根据现行GAAP规则（可能会改变），该利益将通过减少的税收费用在未来几年内流入收益表，并因此进入净值。我们预计新起点调整的总收益将在3000万至4000万美元范围内。但应注意，这是一次性收益，而其他与保险相关的税收变化的负面影响不仅是持续性的，而且在重要方面，随着时间的推移将变得更加严重。

• 新税法废除了《通用公用事业原则》。这意味着在1987年及以后，公司清算将面临双重征税，一层在公司层面，另一层在股东层面。过去，公司层面的税收可以避免。例如，如果伯克希尔被清算——这肯定不会发生——在新法下，股东从出售我们资产中获得的收益将远低于过去资产出售所能获得的收益（假设每次出售价格相同）。尽管这在我们情况下是理论上的，但法律的这一变化将对许多公司产生非常重大的影响。因此，它也会影响我们对潜在投资的评估。以希望出售业务的油气生产公司、特定媒体公司、房地产公司等为例。其股东能够实现的价值很可能因《通用公用事业原则》被废除而显著降低——尽管这些公司的经营经济状况并未发生任何不利变化。我的印象是，无论是投资者还是管理者，对法律的这一重要变化尚未完全理解。

我们报告的这部分内容比我原本希望的要冗长和复杂得多。但法律的变化是众多且重要的，特别是对财产/意外伤害保险公司而言。正如我所指出的，新法律将损害伯克希尔的业绩，但其负面影响无法精确量化。

杂项

去年我们买了一架公务机。你听说的关于这种飞机的一切都是真的：它们非常昂贵，对于像我们这样很少需要前往偏远地方出差的情况来说是一种奢侈品。飞机不仅运营成本高昂，而且仅仅是停放着的成本也很高。税前，一架价值1500万美元的新飞机的资金成本加折旧大约每年300万美元。对于我们那架以85万美元购买的二手飞机，这样的成本每年接近20万美元。

深知这些数字，不幸的是，你们的董事长过去对公司飞机发表过一些相当激烈的言论。因此，在购买之前，我被迫进入了伽利略模式。我迅速体验了必要的“反向启示”，现在旅行比以前方便多了——也贵多了。伯克希尔能否从这架飞机上收回本钱还是个问题，但我会努力取得一些商业成功，以便（无论多么可疑地）将其归功于它。恐怕本·富兰克林看穿了我。他说：“做一个理性动物是如此方便，因为你可以为自己想做的任何事情找到或制造一个理由。”

大约97%的合格股份参与了伯克希尔1986年的股东指定捐赠计划。通过该计划进行的捐款为400万美元，共有1,934家慈善机构受益。

我们敦促新股东阅读第58和59页关于我们股东指定捐赠计划的描述。如果您希望参与未来的计划，我们强烈敦促您立即确保您的股份以实际所有者的名义登记，而不是以“券商名义”或代名人名义。未能在1987年9月30日如此登记的股份将没有资格参加1987年的计划。

去年大约有450人参加了我们的股东大会，高于前年的约250人（以及十年前的大约十几人）。我希望您能在5月19日加入我们在奥马哈的会议。查理和我喜欢回答与所有者相关的问题，我可以向您保证，我们的股东会提出许多好问题。回答问题可能需要相当长的时间——去年我们大约有65个问题，所以一旦您自己的问题得到解答，您就可以随时离开。

去年，会后，来自新泽西和纽约的两位股东去了家具城，每个人从B夫人那里购买了一块5000美元的东方地毯。（准确地说，他们购买的地毯在其他地方可能要花费10000美元，而他们只被收取了大约5000美元。）B夫人很高兴——但并不满足——她将在今年的会后在商店里等着你们。除非我们的股东超过去年的记录，否则我会有麻烦。所以，帮我（也帮你自己）一个忙，去看看她。

沃伦·E·巴菲特 董事会主席

1987年2月27日

附录

购买价格会计调整与“现金流”谬误

首先一个小测验：以下是两家公司1986年的简要损益表。哪家企业更有价值？

（数字(1)至(4)表示本节后面讨论的项目。）

正如你可能猜到的，公司O和公司N是同一家企业——斯科特·费策公司。在“O”（代表“旧”）列中，我们展示了如果伯克希尔没有收购它，该公司1986年的GAAP收益；在“N”（代表“新”）列中，我们展示了伯克希尔实际报告的斯科特·费策的GAAP收益。

应该强调的是，两列显示了相同的经济实质——即相同的销售额、工资、税收等。并且两个“公司”为所有者创造相同数量的现金。只是会计处理不同。

那么，同行的哲学家们，哪一列反映了真相？管理者和投资者应该关注哪一组数字？

在我们解决这些问题之前，让我们看看是什么导致了O和N之间的差异。我们将在某些方面简化讨论，但这种简化不应该导致分析或结论出现任何不准确之处。

O和N之间的差异源于我们为斯科特·费策支付的价格与其账面净值不同。根据GAAP，此类差异——此类溢价或折价——必须通过“购买价格调整”来处理。在斯科特·费策的案例中，我们支付了3.15亿美元购买其账面价值为1.724亿美元的净资产。所以我们支付了1.426亿美元的溢价。

核算任何已支付溢价的第一步是，将流动资产的价值调整为当前价值。在实践中，这一要求通常不影响应收账款，因为应收账款通常已经按当前价值列示，但会影响存货。由于2290万美元的LIFO储备和其他会计上的复杂性，斯科特·费策的存货账户的账面价值比当前价值低3730万美元。因此，作为我们的第一个会计动作，我们从1.426亿美元的溢价中使用了3730万美元来增加存货的账面价值。

假设在流动资产调整后仍有剩余溢价，下一步是将固定资产调整至当前价值。在我们的案例中，这一调整还需要一些与递延税相关的会计技巧。由于这被标榜为简化讨论，我将跳过细节，直接给出底线：6800万美元被添加到固定资产，1300万美元从递延税负债中消除。在完成这8100万美元的调整后，我们剩下2430万美元的溢价需要分配。

如果情况需要，接下来还会有两个步骤：将商誉以外的其他无形资产调整为当前公允价值，以及将负债调整为当前公允价值，这一要求通常只影响长期债务和未备基金的养老金负债。然而，在斯科特·费策的案例中，这两个步骤都不需要。

在记录了所有资产和负债的公允市场价值之后，我们需要做的最后一个会计调整是将剩余的溢价分配给商誉（技术上称为“成本超过所购净资产公允价值的差额”）。这个剩余金额为2430万美元。因此，下表中O栏所总结的收购前立即呈现的斯科特·费策资产负债表，通过购买转变成了N栏所示的资产负债表。实际上，两个资产负债表都描述了相同的资产和负债——但正如你所见，某些数字差异显著。

N栏所示的较高资产负债表数字导致了前面已呈现的收益表N栏中较低的收入数字。这是资产增值的结果，因为部分增值资产必须计提折旧或摊销。资产数字越高，每年的折旧或摊销费用对收益的影响就越大。由于资产负债表增值而流入收益表的费用，在前面呈现的收益表中已编号为：

(1) 4,979,000美元的非现金存货成本，主要源于斯科特·费策在1986年减少其库存；此类费用在未来几年可能会很小或不存在。 (2) 5,054,000美元的额外折旧，归因于固定资产的增值；类似金额的费用可能在未来12年内每年计提。 (3) 595,000美元的商誉摊销；此费用将在未来39年内每年计提，金额会稍大，因为我们的收购是在1月6日完成的，因此1986年的数字仅适用于全年的98%。 (4) 998,000美元用于递延税调整，其复杂程度我无法简要解释（甚至可能无法冗长解释）；类似金额的费用可能在未来12年内每年计提。

重要的是要理解，这些新产生的会计成本，总计1160万美元，均不能在所得税前扣除。“新”斯科特·费策与“旧”斯科特·费策缴纳的税款完全相同，尽管两个实体的GAAP收益差异很大。并且，就营业收益而言，未来也是如此。然而，在斯科特·费策出售其一项业务的不太可能发生的情况下，“旧”公司和“新”公司的税务后果可能会有很大差异。

到1986年底，“旧”和“新”斯科特·费策的净值之间的差额已从1.426亿美元减少到1.310亿美元，这是因为新实体额外计提了1160万美元的费用到收益中。随着时间推移，类似的收益费用将导致大部分溢价消失，两张资产负债表将趋同。然而，新资产负债表上建立的高土地价值和大部分高存货价值将保持不变，除非土地被处置或库存水平进一步降低。

所有这些对所有者意味着什么？伯克希尔的股东是购买了一家在1986年盈利4020万美元的企业，还是购买了一家盈利2860万美元的企业？那1160万美元的新费用对我们来说是真实的经济成本吗？投资者应该为公司O的股票支付比公司N更高的价格吗？而且，如果一家企业价值是其收益的某个给定倍数，斯科特·费策在我们购买前一天的价值是否远比购买后一天的价值高得多？

如果我们思考这些问题，我们可以对所谓的“所有者收益”获得一些见解。这表示为(a)报告收益加上(b)折旧、损耗、摊销以及某些其他非现金费用，例如公司N的项目(1)和(4)，减去(c)企业为充分维持其长期竞争地位和单位产量所需的厂房和设备等的年平均资本化支出。(如果企业需要额外的营运资金来维持其竞争地位和单位产量，增量也应包含在(c)中。然而，采用LIFO存货方法的企业在单位产量不变时通常不需要额外的营运资金。)

我们的所有者收益公式并未产生GAAP所提供的看似精确的数字，因为(c)必须是一个猜测——有时非常难以做出。尽管存在这个问题，我们认为所有者收益数字，而非GAAP数字，是用于估值目的的相关项目——既适用于购买股票的投资者，也适用于购买整个企业的管理者。我们同意凯恩斯的观察：“我宁愿模糊地对，也不愿精确地错。”

我们概述的方法为O公司和N公司产生了相同的“所有者收益”，这意味着估值也相同，正如常识告诉你的那样。这个结果之所以达成，是因为(a)和(b)的总和在O栏和N栏中是相同的，并且因为(c)在两种情况下必然相同。

那么，作为所有者和经理人，查理和我认为斯科特·费策的正确的所有者收益数字是多少？在目前情况下，我们认为(c)非常接近“旧”公司的(b)数字830万美元，并且远低于“新”公司的(b)数字1990万美元。因此，我们认为所有者收益在O栏报告的收益中得到了远比N栏报告的收益好得多的描述。换句话说，我们感觉斯科特·费策的所有者收益远大于我们报告的GAAP数字。

这显然是一种令人高兴的状况。但这种计算通常不会带来如此令人愉快的消息。大多数管理者可能会承认，从长期来看，他们需要在其业务上花费比(b)更多的费用，仅仅是为了在单位产量和竞争地位上保持现有水平。当这种必要性存在时——即当(c)超过(b)时——GAAP收益高估了所有者收益。这种高估常常是巨大的。石油行业在近年来提供了这种现象的一个显著例子。如果大多数大型石油公司每年只花费(b)金额，它们将注定在实际上萎缩。

这一切都凸显了华尔街报告中经常出现的“现金流”数字的荒谬性。这些数字通常包括(a)加上(b)——但不减去(c)。大多数投资银行的销售说明书也以这种具有欺骗性的陈述为特色。这暗示着被出售的企业是金字塔的商业对应物——永远是最先进的，永远不需要更换、改进或翻新。事实上，如果所有美国公司都通过我们领先的投资银行同时出售——并且如果描述它们的销售说明书可信的话——那么政府对国家厂房和设备支出的预测将需要削减90%。

“现金流”确实可能在描述某些房地产企业或其他需要巨额初始投资且后续投资很小的企业时，作为一种有用的简略表达。一个只拥有一座桥梁或一个寿命极长的气田的公司就是一个例子。但对于制造业、零售业、采掘业和公用事业等企业来说，“现金流”是毫无意义的，因为对于它们来说，(c)始终是重要的。当然，这类企业可能在某个特定年份能够推迟资本支出。但超过五年或十年，它们必须进行投资——否则企业就会衰败。

那么，为什么“现金流”数字在当今如此流行？作为回答，我们承认我们的愤世嫉俗：我们认为这些数字经常被企业和证券的营销者用来试图证明不合理的事情是合理的（并以此销售本应无法销售的东西）。当(a)——即GAAP收益——本身看起来不足以偿还垃圾债券的债务或证明愚蠢的股票价格合理时，销售人员专注于(a)+(b)是多么方便啊。但是，你不应该在不减去(c)的情况下加上(b)：虽然牙医正确地声称，如果你忽视你的牙齿，它们会离你而去，但(c)却不是这样。相信企业的偿债能力或股权估值可以通过合计(a)和(b)而忽略(c)来衡量的公司或投资者，注定会遇到麻烦。

总结一下：对于斯科特·费策和我们的其他业务，我们觉得基于历史成本的(b)——即排除无形资产摊销和其他购买价格调整后——与(c)的金额相当接近。（当然，这两个项目并不完全相同。例如，在喜诗糖果，我们每年进行资本性支出，超出折旧50万到100万美元，仅仅是为了保持我们的竞争地位。）我们对此的信念是我们将摊销和其他购买价格调整项目在第8页表格中单独列示的原因，也是我们将那里报告的各业务收益视为比GAAP数字更接近所有者收益的原因。

质疑GAAP数字对一些读者来说可能显得不敬。毕竟，如果会计师不负责向我们提供关于我们业务的“真相”，我们付钱给他们做什么？但会计师的工作是记录，而不是评估。评估工作落到了投资者和管理者身上。

当然，会计数字是商业的语言，因此对于任何评估企业价值和追踪其进展的人来说，都是巨大的帮助。没有这些数字，查理和我会迷失方向：它们始终是我们评估自己和他人业务的起点。然而，管理者和所有者需要记住，会计只是商业思维的辅助工具，而绝不能替代它。