伯克希尔·哈撒韦 2005 年股东信

To the Shareholders of Berkshire Hathaway Inc.:

Our gain in net worth during 2005 was $5.6 billion, which increased the per-share book value of both our Class A and Class B stock by 6.4%. Over the last 41 years (that is, since present management took over) book value has grown from $19 to $59,377, a rate of 21.5% compounded annually.*

* All figures used in this report apply to Berkshire’s A shares, the successor to the only stock that the company had outstanding before 1996. The B shares have an economic interest equal to 1/30th that of the A.

Berkshire had a decent year in 2005. We initiated five acquisitions (two of which have yet to close) and most of our operating subsidiaries prospered. Even our insurance business in its entirety did well, though Hurricane Katrina inflicted record losses on both Berkshire and the industry. We estimate our loss from Katrina at $2.5 billion — and her ugly sisters, Rita and Wilma, cost us an additional $.9 billion.

Credit GEICO — and its brilliant CEO, Tony Nicely — for our stellar insurance results in a disaster-ridden year. One statistic stands out: In just two years, GEICO improved its productivity by 32%. Remarkably, employment fell by 4% even as policy count grew by 26% — and more gains are in store. When we drive unit costs down in such a dramatic manner, we can offer ever-greater value to our customers. The payoff: Last year, GEICO gained market-share, earned commendable profits and strengthened its brand. If you have a new son or grandson in 2006, name him Tony.

My goal in writing this report is to give you the information you need to estimate Berkshire’s intrinsic value. I say “estimate” because calculations of intrinsic value, though all-important, are necessarily imprecise and often seriously wrong. The more uncertain the future of a business, the more possibility there is that the calculation will be wildly off-base. (For an explanation of intrinsic value, see pages 77 — 78.) Here Berkshire has some advantages: a wide variety of relatively-stable earnings streams, combined with great liquidity and minimum debt. These factors mean that Berkshire’s intrinsic value can be more precisely calculated than can the intrinsic value of most companies.

Yet if precision is aided by Berkshire’s financial characteristics, the job of calculating intrinsic value has been made more complex by the mere presence of so many earnings streams. Back in 1965, when we owned only a small textile operation, the task of calculating intrinsic value was a snap. Now we own 68 distinct businesses with widely disparate operating and financial characteristics. This array of unrelated enterprises, coupled with our massive investment holdings, makes it impossible for you to simply examine our consolidated financial statements and arrive at an informed estimate of intrinsic value.

We have attempted to ease this problem by clustering our businesses into four logical groups, each of which we discuss later in this report. In these discussions, we will provide the key figures for both the group and its important components. Of course, the value of Berkshire may be either greater or less than the sum of these four parts. The outcome depends on whether our many units function better or worse by being part of a larger enterprise and whether capital allocation improves or deteriorates when it is under the direction of a holding company. In other words, does Berkshire ownership bring anything to the party, or would our shareholders be better off if they directly owned shares in each of our 68 businesses? These are important questions but ones that you will have to answer for yourself.

Before we look at our individual businesses, however, let’s review two sets of figures that show where we’ve come from and where we are now. The first set is the amount of investments (including cash and cash-equivalents) we own on a per-share basis. In making this calculation, we exclude investments held in our finance operation because these are largely offset by borrowings:

In addition to these marketable securities, which with minor exceptions are held in our insurance companies, we own a wide variety of non-insurance businesses. Below, we show the pre-tax earnings (excluding goodwill amortization) of these businesses, again on a per-share basis:

When growth rates are under discussion, it will pay you to be suspicious as to why the beginning and terminal years have been selected. If either year was aberrational, any calculation of growth will be distorted. In particular, a base year in which earnings were poor can produce a breathtaking, but meaningless, growth rate. In the table above, however, the base year of 1965 was abnormally good; Berkshire earned more money in that year than it did in all but one of the previous ten.

As you can see from the two tables, the comparative growth rates of Berkshire’s two elements of value have changed in the last decade, a result reflecting our ever-increasing emphasis on business acquisitions. Nevertheless, Charlie Munger, Berkshire’s Vice Chairman and my partner, and I want to increase the figures in both tables. In this ambition, we hope — metaphorically — to avoid the fate of the elderly couple who had been romantically challenged for some time. As they finished dinner on their 50th anniversary, however, the wife — stimulated by soft music, wine and candlelight — felt a long-absent tickle and demurely suggested to her husband that they go upstairs and make love. He agonized for a moment and then replied, “I can do one or the other, but not both.”

Acquisitions

Over the years, our current businesses, in aggregate, should deliver modest growth in operating earnings. But they will not in themselves produce truly satisfactory gains. We will need major acquisitions to get that job done.

In this quest, 2005 was encouraging. We agreed to five purchases: two that were completed last year, one that closed after yearend and two others that we expect to close soon. None of the deals involve the issuance of Berkshire shares. That’s a crucial, but often ignored, point: When a management proudly acquires another company for stock, the shareholders of the acquirer are concurrently selling part of their interest in everything they own. I’ve made this kind of deal a few times myself — and, on balance, my actions have cost you money.

Here are last year’s purchases:

• On June 30 we bought Medical Protective Company (“MedPro”), a 106-year-old medical malpractice insurer based in Fort Wayne. Malpractice insurance is tough to underwrite and has proved to be a graveyard for many insurers. MedPro nevertheless should do well. It will have the attitudinal advantage that all Berkshire insurers share, wherein underwriting discipline trumps all other goals. Additionally, as part of Berkshire, MedPro has financial strength far exceeding that of its competitors, a quality assuring doctors that long-to-settle claims will not end up back on their doorstep because their insurer failed. Finally, the company has a smart and energetic CEO, Tim Kenesey, who instinctively thinks like a Berkshire manager.

• Forest River, our second acquisition, closed on August 31. A couple of months earlier, on June 21, I received a two-page fax telling me — point by point — why Forest River met the acquisition criteria we set forth on page 25 of this report. I had not before heard of the company, a recreational vehicle manufacturer with $1.6 billion of sales, nor of Pete Liegl, its owner and manager. But the fax made sense, and I immediately asked for more figures. These came the next morning, and that afternoon I made Pete an offer. On June 28, we shook hands on a deal.

Pete is a remarkable entrepreneur. Some years back, he sold his business, then far smaller than today, to an LBO operator who promptly began telling him how to run the place. Before long, Pete left, and the business soon sunk into bankruptcy. Pete then repurchased it. You can be sure that I won’t be telling Pete how to manage his operation.

Forest River has 60 plants, 5,400 employees and has consistently gained share in the RV business, while also expanding into other areas such as boats. Pete is 61 — and definitely in an acceleration mode. Read the piece from RV Business that accompanies this report, and you’ll see why Pete and Berkshire are made for each other.

• On November 12, 2005, an article ran in The Wall Street Journal dealing with Berkshire’s unusual acquisition and managerial practices. In it Pete declared, “It was easier to sell my business than to renew my driver’s license.”

In New York, Cathy Baron Tamraz read the article, and it struck a chord. On November 21, she sent me a letter that began, “As president of Business Wire, I’d like to introduce you to my company, as I believe it fits the profile of Berkshire Hathaway subsidiary companies as detailed in a recent Wall Street Journal article.”

By the time I finished Cathy’s two-page letter, I felt Business Wire and Berkshire were a fit. I particularly liked her penultimate paragraph: “We run a tight ship and keep unnecessary spending under wraps. No secretaries or management layers here. Yet we’ll invest big dollars to gain a technological advantage and move the business forward.”

I promptly gave Cathy a call, and before long Berkshire had reached agreement with Business Wire’s controlling shareholder, Lorry Lokey, who founded the company in 1961 (and who had just made Cathy CEO). I love success stories like Lorry’s. Today 78, he has built a company that disseminates information in 150 countries for 25,000 clients. His story, like those of many entrepreneurs who have selected Berkshire as a home for their life’s work, is an example of what can happen when a good idea, a talented individual and hard work converge.

• In December we agreed to buy 81% of Applied Underwriters, a company that offers a combination of payroll services and workers’ compensation insurance to small businesses. A majority of Applied’s customers are located in California.

In 1998, though, when the company had 12 employees, it acquired an Omaha-based operation with 24 employees that offered a somewhat-similar service. Sid Ferenc and Steve Menzies, who have built Applied’s remarkable business, concluded that Omaha had many advantages as an operational base — a brilliant insight, I might add — and today 400 of the company’s 479 employees are located here.

Less than a year ago, Applied entered into a large reinsurance contract with Ajit Jain, the extraordinary manager of National Indemnity’s reinsurance division. Ajit was impressed by Sid and Steve, and they liked Berkshire’s method of operation. So we decided to join forces. We are pleased that Sid and Steve retain 19% of Applied. They started on a shoestring only 12 years ago, and it will be fun to see what they can accomplish with Berkshire’s backing.

• Last spring, MidAmerican Energy, our 80.5% owned subsidiary, agreed to buy PacifiCorp, a major electric utility serving six Western states. An acquisition of this sort requires many regulatory approvals, but we’ve now obtained these and expect to close this transaction soon. Berkshire will then buy $3.4 billion of MidAmerican’s common stock, which MidAmerican will supplement with $1.7 billion of borrowing to complete the purchase. You can’t expect to earn outsized profits in regulated utilities, but the industry offers owners the opportunity to deploy large sums at fair returns — and therefore, it makes good sense for Berkshire. A few years back, I said that we hoped to make some very large purchases in the utility field. Note the plural — we’ll be looking for more.

In addition to buying these new operations, we continue to make “bolt-on” acquisitions. Some aren’t so small: Shaw, our carpet operation, spent about $550 million last year on two purchases that furthered its vertical integration and should improve its profit margin in the future. XTRA and Clayton Homes also made value-enhancing acquisitions.

Unlike many business buyers, Berkshire has no “exit strategy.” We buy to keep. We do, though, have an entrance strategy, looking for businesses in this country or abroad that meet our six criteria and are available at a price that will produce a reasonable return. If you have a business that fits, give me a call. Like a hopeful teenage girl, I’ll be waiting by the phone.

Insurance

Let’s now talk about our four sectors and start with insurance, our core business. What counts here is the amount of “float” and its cost over time.

For new readers, let me explain. “Float” is money that doesn’t belong to us but that we temporarily hold. Most of our float arises because (1) premiums are paid upfront though the service we provide — insurance protection — is delivered over a period that usually covers a year and; (2) loss events that occur today do not always result in our immediately paying claims, because it sometimes takes many years for losses to be reported (asbestos losses would be an example), negotiated and settled. The $20 million of float that came with our 1967 entry into insurance has now increased — both by way of internal growth and acquisitions — to $49 billion.

Float is wonderful — if it doesn’t come at a high price. Its cost is determined by underwriting results, meaning how the expenses and losses we will ultimately pay compare with the premiums we have received. When an insurer earns an underwriting profit — as has been the case at Berkshire in about half of the 39 years we have been in the insurance business — float is better than free. In such years, we are actually paid for holding other people’s money. For most insurers, however, life has been far more difficult: In aggregate, the property-casualty industry almost invariably operates at an underwriting loss. When that loss is large, float becomes expensive, sometimes devastatingly so.

In 2004 our float cost us less than nothing, and I told you that we had a chance — absent a mega-catastrophe — of no-cost float in 2005. But we had the mega-cat, and as a specialist in that coverage, Berkshire suffered hurricane losses of $3.4 billion. Nevertheless, our float was costless in 2005 because of the superb results we had in our other insurance activities, particularly at GEICO.

* * * * * * * * * * * *

Auto policies in force grew by 12.1% at GEICO, a gain increasing its market share of U.S. private passenger auto business from about 5.6% to about 6.1%. Auto insurance is a big business: Each share-point equates to $1.6 billion in sales.

While our brand strength is not quantifiable, I believe it also grew significantly. When Berkshire acquired control of GEICO in 1996, its annual advertising expenditures were $31 million. Last year we were up to $502 million. And I can’t wait to spend more.

Our advertising works because we have a great story to tell: More people can save money by insuring with us than is the case with any other national carrier offering policies to all comers. (Some specialized auto insurers do particularly well for applicants fitting into their niches; also, because our national competitors use rating systems that differ from ours, they will sometimes beat our price.) Last year, we achieved by far the highest conversion rate — the percentage of internet and phone quotes turned into sales — in our history. This is powerful evidence that our prices are more attractive relative to the competition than ever before. Test us by going to GEICO.com or by calling 800-847-7536. Be sure to indicate you are a shareholder because that fact will often qualify you for a discount.

I told you last year about GEICO’s entry into New Jersey in August, 2004. Drivers in that state love us. Our retention rate there for new policyholders is running higher than in any other state, and by sometime in 2007, GEICO is likely to become the third largest auto insurer in New Jersey. There, as elsewhere, our low costs allow low prices that lead to steady gains in profitable business.

That simple formula immediately impressed me 55 years ago when I first discovered GEICO. Indeed, at age 21, I wrote an article about the company — it’s reproduced on page 24 — when its market value was $7 million. As you can see, I called GEICO “The Security I Like Best.” And that’s what I still call it.

* * * * * * * * * * * *

We have major reinsurance operations at General Re and National Indemnity. The former is run by Joe Brandon and Tad Montross, the latter by Ajit Jain. Both units performed well in 2005 considering the extraordinary hurricane losses that battered the industry.

It’s an open question whether atmospheric, oceanic or other causal factors have dramatically changed the frequency or intensity of hurricanes. Recent experience is worrisome. We know, for instance, that in the 100 years before 2004, about 59 hurricanes of Category 3 strength, or greater, hit the Southeastern and Gulf Coast states, and that only three of these were Category 5s. We further know that in 2004 there were three Category 3 storms that hammered those areas and that these were followed by four more in 2005, one of them, Katrina, the most destructive hurricane in industry history. Moreover, there were three Category 5s near the coast last year that fortunately weakened before landfall.

Was this onslaught of more frequent and more intense storms merely an anomaly? Or was it caused by changes in climate, water temperature or other variables we don’t fully understand? And could these factors be developing in a manner that will soon produce disasters dwarfing Katrina?

Joe, Ajit and I don’t know the answer to these all-important questions. What we do know is that our ignorance means we must follow the course prescribed by Pascal in his famous wager about the existence of God. As you may recall, he concluded that since he didn’t know the answer, his personal gain/loss ratio dictated an affirmative conclusion.

So guided, we’ve concluded that we should now write mega-cat policies only at prices far higher than prevailed last year — and then only with an aggregate exposure that would not cause us distress if shifts in some important variable produce far more costly storms in the near future. To a lesser degree, we felt this way after 2004 — and cut back our writings when prices didn’t move. Now our caution has intensified. If prices seem appropriate, however, we continue to have both the ability and the appetite to be the largest writer of mega-cat coverage in the world.

* * * * * * * * * * * *

Our smaller insurers, with MedPro added to the fold, delivered truly outstanding results last year. However, what you see in the table below does not do full justice to their performance. That’s because we increased the loss reserves of MedPro by about $125 million immediately after our purchase.

No one knows with any precision what amount will be required to pay the claims we inherited. Medical malpractice insurance is a “long-tail” line, meaning that claims often take many years to settle. In addition, there are other losses that have occurred, but that we won’t even hear about for some time. One thing, though, we have learned — the hard way — after many years in the business: Surprises in insurance are far from symmetrical. You are lucky if you get one that is pleasant for every ten that go the other way. Too often, however, insurers react to looming loss problems with optimism. They behave like the fellow in a switchblade fight who, after his opponent has taken a mighty swipe at his throat, exclaimed, “You never touched me.” His adversary’s reply: “Just wait until you try to shake your head.”

Excluding the reserves we added for prior periods, MedPro wrote at an underwriting profit. And our other primary companies, in aggregate, had an underwriting profit of $324 million on $1,270 million of volume. This is an extraordinary result, and our thanks go to Rod Eldred of Berkshire Hathaway Homestate Companies, John Kizer of Central States Indemnity, Tom Nerney of U. S. Liability, Don Towle of Kansas Bankers Surety and Don Wurster of National Indemnity.

Here’s the overall tally on our underwriting and float for each major sector of insurance:

Regulated Utility Business

We have an 80.5% (fully diluted) interest in MidAmerican Energy Holdings, which owns a wide variety of utility operations. The largest of these are (1) Yorkshire Electricity and Northern Electric, whose

3.7 million electric customers make it the third largest distributor of electricity in the U.K.; (2) MidAmerican Energy, which serves 706,000 electric customers, primarily in Iowa; and (3) Kern River and Northern Natural pipelines, which carry 7.8% of the natural gas consumed in the U.S. When our PacifiCorp acquisition closes, we will add 1.6 million electric customers in six Western states, with Oregon and Utah providing us the most business. This transaction will increase MidAmerican’s revenues by $3.3 billion and its assets by $14.1 billion.

The Public Utility Holding Company Act (“PUHCA”) was repealed on August 8, 2005, a milestone that allowed Berkshire to convert its MidAmerican preferred stock into voting common shares on February 9, 2006. This conversion ended a convoluted corporate arrangement that PUHCA had forced upon us. Now we have 83.4% of both the common stock and the votes at MidAmerican, which allows us to consolidate the company’s income for financial accounting and tax purposes. Our true economic interest, however, is the aforementioned 80.5%, since there are options outstanding that are sure to be exercised within a few years and that upon exercise will dilute our ownership.

Though our voting power has increased dramatically, the dynamics of our four-party ownership have not changed at all. We view MidAmerican as a partnership among Berkshire, Walter Scott, and two terrific managers, Dave Sokol and Greg Abel. It’s unimportant how many votes each party has; we will make major moves only when we are unanimous in thinking them wise. Five years of working with Dave, Greg and Walter have underscored my original belief: Berkshire couldn’t have better partners.

You will notice that this year we have provided you with two balance sheets, one representing our actual figures per GAAP on December 31, 2005 (which does not consolidate MidAmerican) and one that reflects the subsequent conversion of our preferred. All future financial reports of Berkshire will include MidAmerican’s figures.

Somewhat incongruously, MidAmerican owns the second largest real estate brokerage firm in the U.S. And it’s a gem. The parent company’s name is HomeServices of America, but our 19,200 agents operate through 18 locally-branded firms. Aided by three small acquisitions, we participated in $64 billion of transactions last year, up 6.5% from 2004.

Currently, the white-hot market in residential real estate of recent years is cooling down, and that should lead to additional acquisition possibilities for us. Both we and Ron Peltier, the company’s CEO, expect HomeServices to be far larger a decade from now.

Here are some key figures on MidAmerican’s operations:

Finance and Financial Products

The star of our finance sector is Clayton Homes, masterfully run by Kevin Clayton. He does not owe his brilliant record to a rising tide: The manufactured-housing business has been disappointing since Berkshire purchased Clayton in 2003. Industry sales have stagnated at 40-year lows, and the recent uptick from Katrina-related demand will almost certainly be short-lived. In recent years, many industry participants have suffered losses, and only Clayton has earned significant money.

In this brutal environment Clayton has bought a large amount of manufactured-housing loans from major banks that found them unprofitable and difficult to service. Clayton’s operating expertise and Berkshire’s financial resources have made this an excellent business for us and one in which we are preeminent. We presently service $17 billion of loans, compared to $5.4 billon at the time of our purchase. Moreover, Clayton now owns $9.6 billion of its servicing portfolio, a position built up almost entirely since Berkshire entered the picture.

To finance this portfolio, Clayton borrows money from Berkshire, which in turn borrows the same amount publicly. For the use of its credit, Berkshire charges Clayton a one percentage-point markup on its borrowing cost. In 2005, the cost to Clayton for this arrangement was $83 million. That amount is included in “Other” income in the table on the facing page, and Clayton’s earnings of $416 million are after deducting this payment.

On the manufacturing side, Clayton has also been active. To its original base of twenty plants, it first added twelve more in 2004 by way of the bankruptcy purchase of Oakwood, which just a few years earlier was one of the largest companies in the business. Then in 2005 Clayton purchased Karsten, a four-plant operation that greatly strengthens Clayton’s position on the West Coast.

* * * * * * * * * * * *

Long ago, Mark Twain said: “A man who tries to carry a cat home by its tail will learn a lesson that can be learned in no other way.” If Twain were around now, he might try winding up a derivatives business. After a few days, he would opt for cats.

We lost $104 million pre-tax last year in our continuing attempt to exit Gen Re’s derivative operation. Our aggregate losses since we began this endeavor total $404 million.

Originally we had 23,218 contracts outstanding. By the start of 2005 we were down to 2,890. You might expect that our losses would have been stemmed by this point, but the blood has kept flowing. Reducing our inventory to 741 contracts last year cost us the $104 million mentioned above.

Remember that the rationale for establishing this unit in 1990 was Gen Re’s wish to meet the needs of insurance clients. Yet one of the contracts we liquidated in 2005 had a term of 100 years! It’s difficult to imagine what “need” such a contract could fulfill except, perhaps, the need of a compensation-conscious trader to have a long-dated contract on his books. Long contracts, or alternatively those with multiple variables, are the most difficult to mark to market (the standard procedure used in accounting for derivatives) and provide the most opportunity for “imagination” when traders are estimating their value. Small wonder that traders promote them.

A business in which huge amounts of compensation flow from assumed numbers is obviously fraught with danger. When two traders execute a transaction that has several, sometimes esoteric, variables and a far-off settlement date, their respective firms must subsequently value these contracts whenever they calculate their earnings. A given contract may be valued at one price by Firm A and at another by Firm B. You can bet that the valuation differences — and I’m personally familiar with several that were huge — tend to be tilted in a direction favoring higher earnings at each firm. It’s a strange world in which two parties can carry out a paper transaction that each can promptly report as profitable.

I dwell on our experience in derivatives each year for two reasons. One is personal and unpleasant. The hard fact is that I have cost you a lot of money by not moving immediately to close down Gen Re’s trading operation. Both Charlie and I knew at the time of the Gen Re purchase that it was a problem and told its management that we wanted to exit the business. It was my responsibility to make sure that happened. Rather than address the situation head on, however, I wasted several years while we attempted to sell the operation. That was a doomed endeavor because no realistic solution could have extricated us from the maze of liabilities that was going to exist for decades. Our obligations were particularly worrisome because their potential to explode could not be measured. Moreover, if severe trouble occurred, we knew it was likely to correlate with problems elsewhere in financial markets.

So I failed in my attempt to exit painlessly, and in the meantime more trades were put on the books. Fault me for dithering. (Charlie calls it thumb-sucking.) When a problem exists, whether in personnel or in business operations, the time to act is now.

The second reason I regularly describe our problems in this area lies in the hope that our experiences may prove instructive for managers, auditors and regulators. In a sense, we are a canary in this business coal mine and should sing a song of warning as we expire. The number and value of derivative contracts outstanding in the world continues to mushroom and is now a multiple of what existed in 1998, the last time that financial chaos erupted.

Our experience should be particularly sobering because we were a better-than-average candidate to exit gracefully. Gen Re was a relatively minor operator in the derivatives field. It has had the good fortune to unwind its supposedly liquid positions in a benign market, all the while free of financial or other pressures that might have forced it to conduct the liquidation in a less-than-efficient manner. Our accounting in the past was conventional and actually thought to be conservative. Additionally, we know of no bad behavior by anyone involved.

It could be a different story for others in the future. Imagine, if you will, one or more firms (troubles often spread) with positions that are many multiples of ours attempting to liquidate in chaotic markets and under extreme, and well-publicized, pressures. This is a scenario to which much attention should be given now rather than after the fact. The time to have considered — and improved — the reliability of New Orleans’ levees was before Katrina.

When we finally wind up Gen Re Securities, my feelings about its departure will be akin to those expressed in a country song, “My wife ran away with my best friend, and I sure miss him a lot.”

* * * * * * * * * * * *

Below are the results of our various finance and financial products activities:

Manufacturing, Service and Retailing Operations

Our activities in this part of Berkshire cover the waterfront. Let’s look, though, at a summary balance sheet and earnings statement for the entire group.

This eclectic collection, which sells products ranging from Dilly Bars to fractional interests in Boeing 737s, earned a very respectable 22.2% on average tangible net worth last year. It’s noteworthy also that these operations used only minor financial leverage in achieving that return. Clearly, we own some terrific businesses. We purchased many of them, however, at substantial premiums to net worth — a point reflected in the goodwill item shown on the balance sheet — and that fact reduces the earnings on our average carrying value to 10.1%.