Over the years, Buffett’s common stock purchases have become a part of Berkshire’s folklore. Behind each investment is a unique story. The purchase of the Washington Post Company in 1973 was far different from the 1980 purchase of GEICO. Certainly Buffett’s $500 million investment in Capital Cities, which in turn helped Tom Murphy buy the American Broadcasting Company, was unlike his billion-dollar investment in the Coca-Cola Company. And every one of these stock purchases differed from the investments he made years later in Wells Fargo, General Dynamics, American Express, IBM, and Heinz. But for those of us hoping to understand fully Buffett’s thinking, all these common stock purchases share one very important trait: They allow us to observe his business, management, financial, and market tenets in action.
With the exception of Cap Cities, all of these companies have remained in the Berkshire fold and continue to prosper. Only the Washington Post Company and General Dynamics do not make the list as Berkshire’s top common stock holdings.
In this chapter we examine each purchase in its historical context. This allows us to better analyze Buffett’s thinking at the time of the investment as it relates to the company, the industry, and the stock market.
In 1931, the Washington Post was one of five dailies competing for readers in the nation’s capital. Two years later, the Post, unable to pay for its newsprint, was placed in receivership. That summer, the company was sold at auction to satisfy creditors. Eugene Meyer, a millionaire financier, bought the Washington Post for $825,000. For the next two decades, he supported the paper until it turned a profit. Management of the paper passed to Philip Graham, a brilliant Harvard-educated lawyer, who had married Meyer’s daughter Katharine. In 1954, Philip Graham convinced Eugene Meyer to purchase a rival newspaper, the Times-Herald. Later, Graham purchased Newsweek magazine and two television stations before his tragic suicide in 1963. It is Phil Graham who is credited with transforming the Washington Post from a single newspaper into a media and communications company.
After Graham’s death, control of the Washington Post passed to Katharine Graham. Although she had no experience in managing a major corporation, she quickly distinguished herself by confronting difficult business issues. Much of Katharine Graham’s success can be attributed to her genuine affection for the Post. She had observed how her father and husband had struggled to keep the company viable, and she realized that to be successful, the company would need a decision maker, not a caretaker. “I quickly learned that things don’t stand still,” she said. “You have to make decisions.”1 And she made two doozies, decisions that had a pronounced impact on the newspaper: hiring Ben Bradlee as managing editor and then inviting Warren Buffett to become a director of the company. Bradlee encouraged Katharine Graham to publish the Pentagon Papers and to pursue the Watergate investigation, which earned the Washington Post a reputation for prize-winning journalism. For his part, Buffett taught Katharine Graham how to run a successful business.
Buffett had first met Katharine Graham in 1971. At the time, he owned stock in the New Yorker. Hearing that the magazine might be for sale, he asked Katharine Graham whether the Washington Post would be interested in purchasing it. Although the sale never materialized, Buffett came away very much impressed with the publisher of the Post.
About that time, the Washington Post’s financial structure was headed for profound changes. Under the terms of a trust established by Eugene and Agnes Meyer, Katharine and Philip Graham owned all of the Post’s voting stock. After Phil Graham’s death, Katharine Graham inherited control of the company. Over the years, Eugene Meyer had gifted thousands of shares of private Post stock to several hundred employees in gratitude for their loyalty and service. He also funded the company’s profit-sharing plan with private stock. As the company prospered, the value of the Washington Post skyrocketed from $50 per share in the 1950s to $1,154 in 1971. The profit-sharing plan and the personal holdings of employees required the company to maintain a market for the stock, an arrangement that proved to be an unproductive use of the company’s cash. In addition, the Graham and Meyer family was facing stiff inheritance taxes.
In 1971, Katharine Graham decided to take the Washington Post public, thus erasing the burden of maintaining a market in its own stock, and enabling the family heirs to more profitably plan for their estates. The Washington Post Company was divided into two classes of stock. Class A common stock elected a majority of the board of directors, and class B elected a minority. Katharine Graham held 50 percent of the class A stock, thus effectively controlling the company. In June 1971 the Washington Post Company issued 1,354,000 shares of class B stock. Remarkably, two days later, despite government threats, Katharine Graham gave Ben Bradlee permission to publish the Pentagon Papers. In 1972, the price of both class A and B shares climbed steadily, from $24.75 in January to $38 in December.
But the mood on Wall Street was turning gloomy. In early 1973, the Dow Jones Industrial Average began to slide; by spring, it was down more than 100 points to 921. The Washington Post Company share price was slipping as well; by May it was down to $23. Wall Street brokers were buzzing about IBM—the stock had declined more than 69 points, breaking through its 200-day average; they warned that the technical breakdown was a bad omen for the rest of the market. That same month, gold broke through $100 per ounce, the Federal Reserve boosted the discount rate to 6 percent, and the Dow fell 18 points, its biggest loss in three years. By June, the discount rate was raised again, and the Dow headed down even further, passing through the 900 level.
And all the while, Buffett was quietly buying shares in the Washington Post. By June, he had purchased 467,150 shares at an average price of $22.75, a purchase worth $10,628,000.
At first Katharine Graham was unnerved. The idea of a nonfamily member owning so much Post stock, even though it was noncontrolling, was unsettling. Buffett assured Mrs. Graham that Berkshire’s purchase was for investment purposes only. To reassure her, he suggested that Don Graham, Katharine’s son, be given a proxy to vote Berkshire’s shares. That clinched it. Katharine Graham responded by inviting Buffett to join the board of directors in 1974, and soon made him chairman of the finance committee.
Buffett’s role at the Washington Post is widely known. He helped Katharine Graham persevere during the pressman strikes of the 1970s, and he also tutored Don Graham in business, helping him understand the role of management and its responsibility to its owners. Don, in turn, was an eager student who listened to everything Buffett said. Writing years later, Don Graham promised to “continue to manage the company for the benefit of shareholders, especially long-term shareholders whose perspective extends well beyond quarterly or even yearly results. We will not measure our success by the size of our revenues or the number of companies we control.” Graham vowed always to “manage costs rigorously” and “to be disciplined about the uses we make of our cash.2
Buffett’s grandfather once owned and edited the Cuming County Democrat, a weekly newspaper in West Point, Nebraska. His grandmother helped out at the paper and also set the type at the family’s printing shop. His father, while attending the University of Nebraska, edited the Daily Nebraskan. Buffett himself was once the circulation manager for the Lincoln Journal. It has often been said that if Buffett had not embarked on a business investing career, he most surely would have pursued journalism.
In 1969, Buffett bought his first major newspaper, the Omaha Sun, along with a group of weekly papers. Although he respected high-quality journalism, Buffett thought of newspapers first and always as businesses. He expected profits, not influence, to be the rewards for a paper’s owners. Owning the Omaha Sun taught him the business dynamics of a newspaper. He had four years of hands-on experience owning a newspaper before he bought his first share of the Washington Post Company.
Buffett tells Berkshire’s shareholders that his first financial connection with the Washington Post Company was at age 13. He delivered both the Washington Post and the Times-Herald on his paper route while his father served in Congress. Buffett likes to remind others that with his dual delivery route, he merged the two papers long before Phil Graham bought the Times-Herald.
Obviously Buffett was aware of the newspaper’s rich history, and he considered Newsweek magazine a predictable business. The Washington Post Company had for years been reporting the stellar performance of its broadcast division, and Buffett quickly learned the value of the company’s television stations. Buffett’s personal experience with the company and its own successful history led him to believe that the company was a consistent and dependable business performer.
“The economics of a dominant newspaper,” Buffett wrote in 1984, “are excellent, among the very best in the world.”3 Take note that Buffett said this almost 30 years ago, a full decade before the Internet’s potential was first being realized.
In the early 1980s, there were 1,700 newspapers in the United States and approximately 1,600 operated without any direct competition. The owners of newspapers, Buffett noted, like to believe that the exceptional profits they earn each year are a result of their paper’s journalistic quality. The truth is that even a third-rate newspaper can generate adequate profits if it is the only paper in town. Now it’s true that a high-quality paper will achieve a higher penetration rate, but even a mediocre paper, he explains, is essential to a community for its bulletin-board appeal. Every business in town, every home seller, and every individual who wants to get a message out to a community needs the circulation of a newspaper to do so. Like Canadian media entrepreneur Lord Thomson, Buffett believed that owning a newspaper was like receiving a royalty on every business in town that wanted to advertise.
In addition to their franchise quality, newspapers possess valuable economic goodwill. As Buffett points out, newspapers have low capital needs, so they can easily translate sales into profits. Even when a newspaper installs expensive computer-assisted printing presses and electronic newsroom systems, they are quickly paid for by lower fixed-wage costs. During the 1970s and 1980s, newspapers were also able to increase prices relatively easily, thereby generating above-average returns on invested capital and reducing the harmful effects of inflation.
In 1973, the total market value for the Washington Post Company was $80 million. Yet Buffett claims that “most security analysts, media brokers, and media executives would have estimated WPC’s intrinsic value at $400 to $500 million.”4 How did Buffett arrive at that estimate? Let us walk through the numbers, using Buffett’s reasoning.
We’ll start by calculating owner earnings for the year: net income ($13.3 million) plus depreciation and amortization ($3.7 million) minus capital expenditures ($6.6 million) yields 1973 owner earnings of $10.4 million. If we divide these earnings by the long-term U.S. government bond yield (6.81 percent), the value of the Washington Post Company reaches $150 million, almost twice the market value of the company but well short of Buffett’s estimate.
Buffett tells us that, over time, the capital expenditures of a newspaper will equal depreciation and amortization charges, and therefore net income should approximate owner earnings. Knowing this, we can simply divide net income by the risk-free rate and now reach a valuation of $196 million.
If we stop here, the assumption is that the increase in owner earnings will equal the rise in inflation. But we know that newspapers have unusual pricing power; because most are monopolies in their community, they can raise their prices at rates higher than inflation. If we make one last assumption—the Washington Post has the ability to raise real prices by 3 percent—the value of the company is closer to $350 million. Buffett also knew that the company’s 10 percent pretax margins were below its 15 percent historical average margins, and he knew that Katharine Graham was determined that the Post would once again achieve these margins. If pretax margins improved to 15 percent, the present value of the company would increase by $135 million, bringing the total intrinsic value to $485 million.
Even the most conservative calculation of the company’s value indicates that Buffett bought the Washington Post Company for at least half its intrinsic value. He maintains that he bought the company at less than one-quarter of its value. Either way, he clearly bought the company at a significant discount to its present value. Buffett satisfied Ben Graham’s premise that buying at a discount creates a margin of safety.
When Buffett purchased the stock in the Washington Post, its return on equity was 15.7 percent. That was an average return for most newspapers and only slightly better than that of the Standard & Poor’s (S&P) 500 Industrials index. But within five years, the Post’s return on equity doubled. By then it was twice as high as the S&P Industrials and 50 percent higher than the average newspaper. Over the next 10 years, the Post maintained its supremacy, reaching a high of 36 percent return on equity in 1988.
These above-average returns are more impressive when we observe that the company has, over time, purposely reduced its debt. In 1973, the ratio of long-term debt to shareholders’ equity stood at 37 percent, the second-highest ratio in the newspaper group. Astonishingly, by 1978 Katharine Graham had reduced the company’s debt by 70 percent. In 1983, long-term debt to equity was 2.7 percent—one-tenth the newspaper group average—yet the Post generated a return on equity 10 percent higher than these companies. In 1986, after investing in the cellular telephone systems and purchasing Capital Cities’ 53 cable systems, debt in the company was an uncharacteristic high of $336 million. Within a year, it was reduced to $155 million. By 1992, long-term debt was $51 million and the company’s long-term debt to equity was 5.5 percent compared to the industry average of 42.7 percent.
Six months after the Washington Post Company went public, Katharine Graham met with Wall Street security analysts. The first order of business, she told them, was to maximize profit for the company’s existing operations. Profits continued to rise at the television stations and Newsweek, but profitability at the newspaper was leveling off. Much of the reason, said Mrs. Graham, was high production costs, namely wages. After the Post purchased the Times-Herald, profits had surged. Each time the unions struck the paper (1949, 1958, 1966, 1968, 1969), management had opted to pay their demands rather than risk a shutdown of the paper. During this time, Washington, DC, was still a three-newspaper town. Throughout the 1950s and 1960s, increasing wage costs dampened profits. This problem, Mrs. Graham told the analysts, was going to be solved.
As union contracts began to expire in the 1970s, Mrs. Graham enlisted labor negotiators who took a hard line with the unions. In 1974, the company defeated a strike by the Newspaper Guild and, after lengthy negotiations, the printers settled on a new contract. Mrs. Graham’s firm stance came to a head during the pressmen’s strike in 1975. The strike was violent and bitter. The pressmen lost sympathy when they vandalized the pressroom before striking. Management worked the presses; members of the Guild and the printers’ union crossed the picket lines. After four months, Mrs. Graham announced that the paper was hiring nonunion pressmen. The company had won.
In the early 1970s, the financial press wrote that “the best that could be said about The Washington Post Company’s performance was it rated a gentleman’s C in profitability.”5 Pretax margins in 1973 were 10.8 percent—well below the company’s historical 15 percent margins earned in the 1960s. After successfully renegotiating the union contracts, the Post’s fortunes improved. By 1988, pretax margin reached a high of 31.8 percent, which compared favorably to its newspapers group average of 16.9 percent and the S&P Industrials average of 8.6 percent.
The Washington Post generated substantial cash flow for its owners. Because it generated more cash than it could reinvest in its primary businesses, management was confronted with two choices: return the money to shareholders and/or profitably invest the cash into new opportunities. It is Buffett’s preference to have companies return excess earnings to shareholders. The Washington Post Company, while Katharine Graham was president, was the first newspaper in its industry to repurchase shares in large quantities. Between 1975 and 1991, the company bought an unbelievable 43 percent of its shares at an average price of $60 per share.
A company can also choose to return money to shareholders by increasing the dividend. In 1990, confronted with substantial cash reserves, the Washington Post Company voted to increase the annual dividend to shareholders from $1.84 to $4.00, a 117 percent increase.
In the early 1990s, Buffett concluded that newspapers would remain above-average businesses when compared to American industry in general, but they were destined to become less valuable than he or any other media analyst had predicted years earlier, principally because newspapers had lost their pricing flexibility. In previous years, when the economy slowed and advertisers cut spending, newspapers could maintain profitability by raising linage rates. Today, newspapers are no longer monopolies. Advertisers have found cheaper ways to reach their customers; cable television, direct mail, newspaper inserts, and—most of all—the wide use of the Internet have all taken advertising dollars away from newspapers.
By 1991, Buffett was convinced that the change in profitability represented a long-term secular change as well as a temporary cyclical change. “The fact is,” he confessed, “newspaper, television, and magazine properties have begun to resemble businesses more than franchises in their economic behavior.”6 Cyclical changes hurt short-term earnings but do not reduce a company’s intrinsic value. Secular changes reduce earnings and also reduce intrinsic value. However, the change in intrinsic value of the Washington Post Company, Buffett said, was moderate compared to other media companies. The reasons were twofold. First, the Post’s $50 million long-term debt was more than offset by its $400 million in cash holdings. The Washington Post is the only public newspaper that is essentially free of debt. “As a result,” Buffett said, “the shrinkage in the value of their assets has not been accentuated by the effects of leverage.”7
Buffett’s goal is to select companies in which each dollar of retained earnings is translated into a least one dollar of market value. This test can quickly identify companies whose managers, over time, have been able to optimally invest their company’s capital. If retained earnings are invested in the company and produce above-average returns, the proof will be a proportionally greater rise in the company’s market value.
From 1973 to 1992, the Washington Post Company earned $1.755 billion for its owners. From these earnings, the company paid shareholders $299 million and retained $1.456 billion to reinvest in the company. In 1973, the total market value of the Washington Post Company was $80 million. By 1992, the market value had grown to $2.630 billion. Over those 20 years, for every $1.00 the company retained, it created $1.81 in market value for shareholders.
Still, there is one more way to judge the success of the Washington Post Company under Katharine Graham’s leadership. In his very insightful book The Outsiders: Eight Unconventional CEOs and Their Radically Rational Blueprint for Success, William Thorndike helps us best appreciate how well both company and its CEO actually performed. “From the time of the company’s IPO in 1971 until [Katharine Graham] stepped down as chairman in 1993, the compounded annual return for shareholders was a remarkable 22.3 percent, dwarfing both the S&P (7.4 percent) and her peers (12.4 percent). A dollar invested at the IPO was worth $89 by the time she retired, versus $5 for the S&P and $14 for her peer group. Katharine Graham outperformed the S&P by eighteenfold and her peers by over sixfold. She was simply the best newspaper executive in the country during her twenty-two-year period by a wide margin.”8
The Government Employees Insurance Company (GEICO) was founded in 1936 by Leo Goodwin, an insurance accountant.9 He envisioned a company that insured only preferred-risk drivers and sold this insurance directly by mail. He had discovered that government employees, as a group, had fewer accidents than the general public. He also knew that by selling directly to the driver, the company would eliminate the overhead associated with agents, typically 10 to 25 percent of every premium dollar. Goodwin figured that if he isolated careful drivers and passed along the savings from issuing insurance policies directly, he would have a recipe for success.
Goodwin invited a Fort Worth, Texas, banker named Cleaves Rhea to be his partner. Goodwin invested $25,000 and owned 25 percent of the stock; Rhea invested $75,000 for 75 percent. In 1948, the company moved from Texas to Washington, D.C. That year, the Rhea family decided to sell its interest in the company, and Rhea enlisted Lorimer Davidson, a Baltimore bond salesperson, to help with the sale. In turn, Davidson asked David Kreeger, a Washington, D.C., lawyer, to help him find buyers, and Kreeger approached the Graham-Newman Corporation. Ben Graham decided to buy half of Rhea’s stock for $720,000; Kreeger and Davidson’s Baltimore associates bought the other half. The Securities and Exchange Commission forced Graham-Newman, because it was an investment fund, to limit its holdings of GEICO to 10 percent of the company, so Graham had to distribute GEICO’s stock to the fund’s partners. Years later, when GEICO became a billion-dollar company, Graham’s personal shares were worth millions.
Lorimer Davidson, at Goodwin’s invitation, joined GEICO’s management team. In 1958, he became chairman and led the company until 1970. During this period, the board extended the eligibility for GEICO’s car insurance to include professional, managerial, technical, and administrative workers. GEICO’s insurance market now included 50 percent of all car owners, up from 15 percent. The new strategy was a success. Underwriting profits soared because the new group of drivers turned out to be just as careful as government employees.
These were the company’s golden years. Between 1960 and 1970, insurance regulators were mesmerized by GEICO’s success, and shareholders saw their share price soar. The company’s premium-to-surplus ratio rose above 5:1. This ratio measures the risk that a company takes (premiums written) compared to its policyholders’ surplus (capital that is used to pay claims). Because insurance regulators were so impressed with GEICO, the company was allowed to exceed the industry average ratio.
By the late 1960s, GEICO’s fortunes were beginning to dim. In 1969, the company reported that it had underestimated its reserves for that year by $10 million. Instead of earning $2.5 million, the company actually posted a loss. The adjustment to income was made the next year, but again the company underestimated reserves—this time by $25 million—so 1970’s underwriting profit instead showed a disastrous loss.
The revenues an insurance company receives from policyholders are called earned premiums. From these premiums, the company promises to provide coverage to the automobile driver during the year. Costs to an insurance company include insured losses, which are claims brought by drivers, and loss expenses, the administrative costs of settling the claims. These total costs must reflect not only payments made during the year, but estimates of claims yet to be paid. Estimates, in turn, are divided into two categories: claim costs and expenses, which the company expects to pay during the year, and adjustment reserves, set aside to cover underestimated reserves from earlier years. Because of litigation, some insurance claims are not settled for several years and often involve substantial payments for legal and medical expenses. The problem confronting GEICO was that not only had it written insurance policies that were poised to create an underwriting loss, but its estimates for earlier reserves were inadequate as well.
In 1970, Davidson retired and was replaced by David Kreeger, the Washington lawyer. Running the company fell to Norman Gidden, who had served as president and chief executive officer. What happened next suggests that GEICO was attempting to grow out of its reserve mess created in 1969 and 1970. Between 1970 and 1974, the number of new auto policies grew at an 11 percent annual rate compared to a 7 percent average from 1965 to 1970. In addition, in 1972 the company embarked on an expensive and ambitious decentralization program that required significant investments in real estate, computer equipment, and personnel.
By 1973, the company, facing fierce competition, lowered its eligibility standards to expand its market share. Now GEICO’s automobile drivers, for the first time, included blue-collar workers and drivers under age 21, two groups with checkered histories. Both of these strategic changes, the corporate expansion plan and the plan to insure a greater number of motorists, occurred simultaneously with the lifting of the country’s 1973 price controls. Soon, auto repair and medical care costs exploded.
Underwriting losses at GEICO began to appear in the fourth quarter of 1974. For the year, the company reported a $6 million underwriting loss, its first in 28 years. Amazingly, the premium-to-surplus ratio that year was 5:1. Nonetheless, the company continued to pursue growth, and by the second quarter of 1975, GEICO reported more losses and announced it was eliminating the company’s $0.80 dividend.
Gidden employed the consulting firm of Milliman & Robertson to make recommendations on how GEICO could reverse its slide. The results of the study were not encouraging. The company, the consultants said, was underreserved by $35 million to $70 million and would need a capital infusion to stay viable. The board accepted the consultants’ study and made the announcement to its shareholders. In addition, the board projected that 1975’s underwriting loss would approach a staggering $140 million (the actual result was $126 million). Shareholders and insurance regulators were dumbfounded.
In 1972, GEICO’s share price had reached a high of $61. By 1973, the share price was cut in half, and in 1974, it fell further to $10. In 1975, when the board announced the projected losses, the stock dropped to $7. Several stockholders, charging fraud, filed class-action suits against the company. Executives at GEICO blamed inflation and outrageous legal and medical costs for the company’s woes. But these problems confronted all insurers. GEICO’s problem was that it had moved away from its successful tradition of insuring only careful drivers. Furthermore, it was no longer checking corporate expenses. As the company expanded the list of insured drivers, its earlier loss assumptions were woefully inadequate to cover new and more frequent claims. At a time when a company was underestimating its insured losses, it simultaneously was increasing fixed expenses.
At the March 1976 GEICO annual meeting, Gidden confessed that another president might have handled the company’s problems better. He announced that the company’s board of directors had appointed a committee to seek new management. GEICO’s share price was still weakening—it was now $5 and heading lower.10
After the 1976 annual meeting, GEICO announced that John J. Byrne, a 43-year-old marketing executive from Travelers Corporation, would become the new president. Soon after Byrne’s appointment, the company announced a $76 million preferred stock offering to shore up its capital. But shareholders had lost hope, and the stock drifted down to $2 per share.
During this period, Warren Buffett was quietly and doggedly buying stock in GEICO. As the company teetered on the edge of bankruptcy, he invested $4.1 million, gathering 1,294,308 shares at an average price of $3.18.
When Buffett attended Columbia University in 1950, his teacher, Ben Graham, was a director of GEICO. His curiosity stimulated, Buffett went to Washington, D.C., one weekend to visit the company. On Saturday, he knocked on the company’s door and was let in by a janitor, who led him to the only executive in the office that day, Lorimer Davidson. Buffett peppered him with questions, and Davidson spent the next five hours schooling his young visitor on GEICO’s distinctions. Philip Fisher would have been impressed.
Later, when Buffett returned to Omaha and his father’s brokerage firm, he recommended that the firm’s clients buy GEICO. He himself invested $10,000, approximately two-thirds of his net worth, in its stock. Many investors resisted his recommendation. Even Omaha’s insurance agents complained to Howard Buffett that his son was promoting an “agentless” insurance company. Frustrated, Warren Buffett sold his GEICO shares a year later, at a 50 percent profit, and did not again purchase shares in the company until 1976.
Undaunted, Buffett continued to recommend insurance stocks to his clients. He bought Kansas City Life at three times its earnings. He owned Massachusetts Indemnity & Life Insurance Company in Berkshire Hathaway’s security portfolio, and in 1967 he purchased the controlling interest in National Indemnity. For the next 10 years, Jack Ringwalt, the CEO of National Indemnity, educated Buffett on the mechanics of running an insurance company. That experience, more than any other, helped Buffett understand how an insurance company makes money. It also, despite GEICO’s shaky financial situation, gave him confidence to purchase the company.
In addition to Berkshire’s $4.1 million investment in GEICO’s common stock, Buffett also invested $19.4 million in its convertible preferred stock issue, which raised additional capital for the company. Two years later, Berkshire converted these preferred shares into common, and in 1980, Buffett invested another $19 million of Berkshire’s money in the company. Between 1976 and 1980, Berkshire invested a total of $47 million, purchasing 7.2 million shares of GEICO at an average price of $6.67 per share. By 1980, that investment had appreciated 123 percent. It was now worth $105 million and had become Buffett’s largest holding.
On first reaction, we might assume that Buffett violated his consistency tenet. Clearly, GEICO’s operations in 1975 and 1976 were anything but consistent. When Byrne became president of GEICO, his job was to turn around the company, and turnarounds, Buffett has often said, seldom turn. So how do we explain Berkshire’s purchase of GEICO?
For one thing, it appears to be a turnaround exception. Byrne successfully turned the company and positioned it to compete again for insurance. But more important, Buffett said, GEICO was not terminal, only wounded. Its franchise of providing low-cost agentless insurance was still intact. Furthermore, in the marketplace, there still existed safe drivers who could be insured at rates that would provide a profit for the company. On a price basis, GEICO would always beat its competitors. For decades, GEICO generated substantial profits for its owners by capitalizing on its competitive strengths. These strengths, said Buffett, were still in place. GEICO’s troubles in the 1970s had nothing to do with a diminution of its franchise. Rather, the company, because of operating and financial troubles, became sidetracked. Even with no net worth, GEICO was still worth a lot of money because its franchises were still in one piece.
Although automobile insurance is a commodity product, Buffett says a commodity business can make money if it has a cost advantage that is both sustainable and wide. This description aptly fits GEICO. We also know that management in a commodity business is a crucial variable. GEICO’s leadership, since Berkshire’s purchase, has demonstrated that it, too, has a competitive advantage.
When John (Jack) Byrne took over GEICO in 1976, he convinced both insurance regulators and competitors that if GEICO went bankrupt, it would be bad for the entire industry. His plan for rescuing the company included raising capital, obtaining a reinsurance treaty with other companies to reinsure a portion of GEICO’s business, and cutting costs aggressively. “Operation Bootstrap,” as Byrne called it, was the battle plan aimed at returning the company to profitability.
In his first year, Byrne closed 100 offices, reduced employment from 7,000 to 4,000, and turned in GEICO’s license to sell insurance in both New Jersey and Massachusetts. Byrne told New Jersey regulators he would not renew the 250,000 policies that were costing the company $30 million a year. Next, he did away with the computerized systems that allowed policyholders to renew their insurance without providing updated information. When Byrne required the new information, he found the company was underpricing 9 percent of its renewal policies. When GEICO repriced them, 400,000 policyholders decided to discontinue their insurance. Altogether, Byrne’s actions reduced the number of policyholders from 2.7 million to 1.5 million, and the company went from being the nation’s 18th largest insurer in 1975 to 31st a year later. Despite this reduction, GEICO, after losing $126 million in 1976, earned an impressive $58.6 million on $463 million in revenues in 1977, Byrne’s first full year of responsibility.
Clearly, GEICO’s dramatic recovery was Byrne’s doing, and his steadfast discipline on corporate expenses sustained GEICO’s recovery for years. Byrne told shareholders that the company must return to its first principle of being the low-cost provider of insurance. His reports detailed how the company continually reduced costs. Even in 1981, when GEICO was the country’s seventh-largest writer of automobile insurance, Byrne shared his secretary with two other executives. He boasted how the company serviced 378 policies per GEICO employee, up from 250 years earlier. During his turnaround years, he was always a great motivator. “Byrne,” said Buffett, “is like the chicken farmer who rolls an ostrich egg into the henhouse and says, ‘Ladies, this is what the competition is doing.’”11
Over the years, Byrne happily reported the successful progress of GEICO; he was equally candid with his shareholders when the news turned bad. In 1985, the company temporarily stumbled when it had underwriting losses. Writing in the company’s first-quarter report to shareholders, Byrne “likened the company’s plight to that of the pilot who told his passengers, ‘the bad news is that we are lost, but the good news is that we are making great time.’”12 The company quickly regained its footing and the following year posted profitable underwriting results. But, just as important, the company gained the reputation for being candid with its shareholders.
Over the years, Jack Byrne demonstrated rational behavior managing GEICO’s assets. After he took charge, he positioned the company for controlled growth. It was more profitable, Byrne figured, to grow at a slower rate that allowed the company to carefully monitor its losses and expenses than to grow twice as fast if it meant losing financial control. Even so, this controlled growth continued to generate excess returns for GEICO, and the mark of rationality is what the company did with the cash.
Starting in 1983, the company was unable to invest its cash profitably so it decided to return the money to its shareholders. Between 1983 and 1992, GEICO repurchased, on a postsplit basis, 30 million shares, reducing the company’s total common shares outstanding by 30 percent. In addition to buying back stock, GEICO also began to increase the dividend it paid to its shareholders. In 1980, the company’s split-adjusted dividend was $0.09 per share; in 1992, it was $0.60 per share, a 21 percent annual increase.
In 1980, the return on equity at GEICO was 30.8 percent—almost twice as high as the peer group average. By the late 1980s, the company’s return on equity began to decline, not because the business was floundering but because its equity grew faster than its earnings. Hence, part of the logic of paying out increasing dividends and buying back stock was to reduce capital and maintain an acceptable return on equity.
Investors can compare profitability of insurance companies in several ways. Pretax margins are one of the best measures. Over the 10-year period 1983 to 1992, GEICO’s average pretax margins were the most consistent, with the lowest standard deviation, of any peer group company.
GEICO, as we now understand, paid meticulous attention to all its expenses and closely tracked the expenses associated with settling insurance claims. During this period, corporate expenses as a percentage of premiums written averaged 15 percent—half the industry average. This low ratio partly reflects the cost of insurance agents that GEICO does not have to pay.
GEICO’s combined ratio of corporate expenses and underwriting losses was demonstrably superior to the industry average. From 1977 through 1992, the industry average beat GEICO’s combined ratio only once, in 1977. Since then, GEICO’s combined ratio has averaged 97.1 percent, more than 10 percentage points better than the industry average. GEICO posted an underwriting loss only twice—once in 1985 and again in 1992. The underwriting loss in 1992 was accentuated by the unusual number of natural disasters that struck the country that year. Without Hurricane Andrew and other major storms, GEICO’s combined ratio would have been a low 93.8 percent.
When Buffett first started to buy GEICO for Berkshire Hathaway, the company was close to bankruptcy. But he says GEICO was worth a substantial sum, even with a negative net worth, because of the company’s insurance franchise. Still, in 1976, the company, since it had no earnings, defied a mathematical determination of value as put forth by John Burr Williams, who defined present value as future cash flows discounted at an appropriate rate. Still, despite the uncertainty over GEICO’s future cash flows, Buffett was sure that the company would survive and earn money in the future. How much and when were open to debate.
In 1980, Berkshire owned one-third of GEICO, invested at a cost of $47 million. That year, GEICO’s total market value was $296 million. Even then, Buffett estimated that the company possessed a significant margin of safety. In 1980, the company earned $60 million on $705 million in revenues. Berkshire’s share of GEICO’s earnings was $20 million. According to Buffett, “to buy a similar $20 million of earnings in a business with first-class economic characteristics and bright prospects would cost a minimum of $200 million”—more if the purchase was for a controlling interest in the company.13
Even so, Buffett’s $200 million assumption is realistic, given the Williams valuation theory. Assuming that GEICO could sustain this $60 million in earnings without the aid of any additional capital, the present value of GEICO, discounted by the then-current 12 percent rate for a 30-year U.S. government bond, would have been $500 million—almost twice GEICO’s 1980 market value. If the company could grow this earnings power at 2 percent real, or at 15 percent before current inflation, the present value would increase to $600 million, and Berkshire’s share would equal $200 million. In other words, in 1980, the market value of GEICO’s stock was less than half the discounted present value of its earnings power.
Between 1980 and 1992, the market value of GEICO grew from $296 million to $4.6 billion—an increase of $4.3 billion. During these 13 years, GEICO earned $1.7 billion. It paid shareholders, in common stock dividends, $280 million and retained $1.4 billion for reinvestment. Thus, for every dollar retained, GEICO created $3.12 in market value for its shareholders. This financial accomplishment demonstrates not only GEICO’s superior management and niche marketing, but its ability to reinvest shareholder money at optimal rates.
Further proof of GEICO’s superiority: a $1 investment in GEICO in 1980, excluding dividends, increased to $27.89 by 1992. This is an astonishing 29.2 percent compounded annual rate of return, far greater than the industry average and the S&P 500 index, which both gained 8.9 percent during the same period.
Cap Cities had its beginning in the news business. In 1954, Lowell Thomas, the famous journalist; his business manager, Frank Smith; and a group of associates bought Hudson Valley Broadcasting Company, which included an Albany, New York, television and AM radio station. At that time, Thomas Murphy was a product manager at Lever Brothers. Frank Smith, who was a golfing partner of Murphy’s father, hired the younger Murphy to manage the company’s television station. In 1957, Hudson Valley purchased a Raleigh-Durham television station, and the company’s name was changed to Capital Cities Broadcasting, reflecting that both Albany and Raleigh were capitals of their respective states.
In 1960, Murphy hired Dan Burke to manage the Albany station. Burke was the brother of one of Murphy’s Harvard classmates, Jim Burke, who later became chairman of Johnson & Johnson. Dan Burke, an Albany native, was left in charge of the television station while Murphy returned to New York, where he was named president of Capital Cities in 1964. Thus began one of the most successful corporate partnerships in American business. During the next three decades, Murphy and Burke ran Capital Cities, and together they made more than 30 broadcasting and publishing acquisitions, the most notable being the purchase of ABC in 1985.
Buffett first met Tom Murphy in the late 1960s at a New York luncheon arranged by one of Murphy’s classmates. The story goes that Murphy was so impressed with Buffett that he invited him to join the board of Capital Cities.14 Buffett declined, but he and Murphy became close friends, keeping in touch over the years. Buffett first invested in Capital Cities in 1977; unexplainably, but profitably, he sold the position the following year.
In December 1984, Murphy approached Leonard Goldenson, chairman of American Broadcasting Companies, with the idea of merging the two companies. Although initially rebuffed, Murphy contacted Goldenson again in January 1985. The Federal Communications Commission (FCC) had increased the number of television and radio stations that a single company could own from seven to 12, effective in April that year. This time Goldenson agreed. Goldenson, then 79 years old, was concerned about his successor. Although ABC had several potential candidates, none was, in his opinion, ready for leadership. Murphy and Burke were considered the best managers in the media and communications industry. By agreeing to merge with Cap Cities, Goldenson was ensuring that ABC would remain in strong management hands. American Broadcasting Companies entered the negotiating room with high-priced investment bankers. Murphy, who always negotiated his own deals, brought his trusted friend Warren Buffett. Together they worked out the first-ever sale of a television network and the largest media merger in history up to that point.
Capital Cities offered American Broadcasting Companies a total package worth $121 per ABC share ($118 in cash per share and one-tenth warrant to purchase Capital Cities worth $3 per share). The offer was twice the value at which ABC’s stock traded the day before the announcement. To finance the $3.5 billion deal, Capital Cities would borrow $2.1 billion from a banking consortium, sell overlapping television and radio stations worth approximately $900 million, and also sell restricted properties that a network was not allowed to own, including cable properties subsequently sold to the Washington Post Company. The last $500 million came from Buffett. He agreed that Berkshire Hathaway would purchase three million newly issued shares of Cap Cities at a price of $172.50 per share. Murphy again asked his friend to join the board, and this time Buffett agreed.
After serving on the board of the Washington Post Company for more than 10 years, Buffett understood the business of television broadcasting and newspaper and magazine publishing. Buffett’s business understanding of television networks grew with Berkshire’s own purchase of ABC once in 1978 and again in 1984.
Both Capital Cities and American Broadcasting Companies had profitable operating histories dating back more than 30 years. ABC averaged 17 percent return on equity and 21 percent debt to equity from 1975 through 1984. Capital Cities, during the 10 years before its offer to purchase ABC, averaged 19 percent return on equity and 20 percent debt to capital.
Broadcasting companies and networks are blessed with above-average economics. Like newspapers, and for much the same reason, they generate a great deal of economic goodwill. Once a broadcasting tower is built, capital reinvestment and working capital needs are minor and inventory investment is nonexistent. Movies and programs can be bought on credit and settled later when advertising dollars roll in. Thus, as a general rule, broadcasting companies produce above-average returns on capital and generate substantial cash in excess of their operating needs.
The risks to networks and broadcasters include government regulation, changing technology, and shifting advertising dollars. Governments can deny the renewal of a company’s broadcasting license, but this is rare. Cable programs, in 1985, were a minor threat to networks. Although some viewers tuned in cable shows, the overwhelming majority of television viewers still preferred network programming. Also during the 1980s, advertising dollars for free-spending consumers were growing substantially faster than the country’s gross domestic product. To reach a mass audience, advertisers still counted on network broadcasting. The basic economics of networks, broadcasting companies, and publishers were, in Buffett’s mind, above average, and in 1985, the long-term prospects for these businesses were highly favorable.
Berkshire’s $517 million investment in Capital Cities at that time was the single-largest investment Buffett ever made. How Buffett determined the combined value of Capital Cities and ABC is open for debate. Murphy agreed to sell Buffett three million shares of Capital Cities/ABC for $172.50 per share. But we know that price and value are often two different figures. Buffett’s practice, we have learned, is to acquire a company only when there is a significant margin of safety between the company’s intrinsic value and its purchase price. However, with the purchase of Capital Cities/ABC, he admittedly compromised this principle.
If we discount Buffett’s offer of $172.50 per share by 10 percent (the approximate yield of the 30-year U.S. government bond in 1985) and multiply this value by 16 million shares (Cap Cities had 13 million shares outstanding plus three million issued to Buffett), the present value of this business would need to have earnings power of $276 million. Capital Cities’ 1984 earnings net after depreciation and capital expenditures were $122 million, and ABC’s net income after depreciation and capital expenditures was $320 million, making the combined earnings power $442 million. But the combined company would have substantial debt: the approximately $2.1 billion that Murphy had to borrow would cost the company $220 million a year in interest. So the net earnings of the combined company were approximately $200 million.
There were additional considerations. Murphy’s reputation for improving the cash flow of purchased businesses simply by reducing expenses was legendary. Capital Cities’ operating margins were 28 percent, whereas ABC’s were 11 percent. If Murphy could improve the operating margins of the ABC properties by one-third, to 15 percent, the company would throw off an additional $125 million each year, and the combined earnings power would equal $225 million annually. The per-share present value of a company earning $325 million with 16 million shares outstanding discounted at 10 percent was $203 per share—a 15 percent margin of safety over Buffett’s $172.50 purchase price. “I doubt if Ben’s up there applauding me on this one,” Buffett quipped, in reference to Ben Graham.15
The margin of safety that Buffett accepted could be expanded if we make certain assumptions. Buffett says that conventional wisdom during this period argued that newspapers, magazines, or television stations would be able to increase earnings forever at 6 percent annually—without the need for additional capital.16 The reasoning, he explains, was that capital expenditures would equal depreciation rates and the need for working capital would be minimal. Hence, income could be thought of as freely distributed earnings. This means that an owner of a media company possessed a perpetual annuity that would grow at 6 percent for the foreseeable future without the need of any additional capital. Compare that, Buffett suggests, to a company that is able to grow only if capital is reinvested. If you owned a media company that earned $1 million and expected to grow at 6 percent, it would be appropriate, says Buffett, to pay $25 million for this business ($1 million divided by a risk-free rate of 10 percent less the 6 percent growth rate). Another business that earned $1 million but could not grow earnings without reinvested capital might be worth $10 million ($1 million divided by 10 percent).
If we take this finance lesson and apply it to Cap Cities, the value of Cap Cities increased from $203 per share to $507, or a 66 percent margin of safety over the $172.50 price that Buffett agreed to pay. But there were a lot of “ifs” in these assumptions. Would Murphy be able to sell a portion of Capital Cities/ABC combined properties for $900 million? (He actually got $1.2 billion.) Would he be able to improve operating margins at American Broadcasting Companies? Would he be able to continually count on the growth of advertising dollars?
Buffett’s ability to obtain a significant margin of safety in Capital Cities was complicated by several factors. First, the stock price of Cap Cities had been rising over the years. Murphy and Burke were doing an excellent job of managing the company, and the company’s share price reflected this. So, unlike GEICO, Buffett did not have the opportunity to purchase Cap Cities cheaply because of a temporary business decline. The stock market, which had been steadily rising, didn’t help, either. And, because this was a secondary stock offering, Buffett had to take a price for Cap Cities’ shares that was close to its then-trading value.
If there was any disappointment over the issue of price, Buffett was comforted by the quick appreciation of those same shares. On Friday, March 15, 1985, Capital Cities’ share price was $176. On Monday afternoon, March 18, Capital Cities announced it would purchase American Broadcasting Companies. The next day, by market close, Capital Cities’ share price was $202.75. In four days, the price had risen 26 points, a 15 percent appreciation. Buffett’s profit was $90 million and the deal was not due to close until January 1986.
The margin of safety that Buffett received buying Capital Cities was significantly less than with other purchases. So why did he proceed? The answer was Tom Murphy. Had it not been for Murphy, Buffett admits he would not have invested in the company. Murphy was Buffett’s margin of safety. Capital Cities/ABC was an exceptional business, the kind of business that attracts Buffett. But there is also something special about Murphy. “Warren adores Tom Murphy,” said John Byrne. “Just to be partners with him is attractive to [Buffett].”17
Cap Cities’ management philosophy is decentralization. Murphy and Burke hire the best people possible and then leave them alone to do their job. All decisions are made at the local level. Burke found this out early in his relationship with Murphy. Burke, while managing the Albany TV station, mailed updated reports weekly to Murphy, who never responded. Burke finally got the message. Murphy promised Burke, “I won’t come to Albany unless you invite me—or I have to fire you.”18 Murphy and Burke help set yearly budgets for their companies and review operating performance quarterly. With these two exceptions, managers were expected to operate their businesses as if they owned them. “We expect a great deal from our managers,” wrote Murphy.19
And one thing Capital Cities’ managers were expected to do was control costs. When they failed, Murphy was not shy about getting involved. When Capital Cities purchased ABC, Murphy’s talent for cutting costs was badly needed. Networks tend to think in terms of ratings, not profits. Whatever was needed to increase ratings, the network thought, superseded cost evaluation. This mentality abruptly stopped when Murphy took over. With the help of carefully selected committees at ABC, Murphy pruned payrolls, perks, and expenses. Some 1,500 people, given generous severance packages, were let go. The executive dining rooms and private elevator at ABC were closed. The limousine at ABC Entertainment in Los Angeles that was used to drive Murphy during his first tour of the company’s operation was discharged. On his next trip, he took a cab.
Such cost-consciousness was a way of life at Capital Cities. The company’s Philadelphia television station, WPVI, the number-one station in the city, had a news staff of 100 compared with 150 at the CBS affiliate across town. Before Murphy arrived at ABC, the company employed 60 people to manage ABC’s five television stations. Soon after the Cap Cities’ acquisition, six people managed eight stations. WABC-TV in New York used to employ 600 people and generated 30 percent pretax margins. Once Murphy reconfigured the station, it employed 400 people with pretax margins north of 50 percent. Once a cost crisis was resolved, Murphy depended on Burke to manage operating decisions. He concentrated on acquisitions and shareholder assets.
The basic economics of the broadcasting and network business assured Cap Cities it would generate ample cash flow. However, the industry’s basic economics, coupled with Murphy’s penchant for controlling costs, meant Cap Cities would have overwhelming cash flow. From 1988 through 1992, Cap Cities generated $2.3 billion in unencumbered cash. Given these resources, some managers might be unable to resist the temptation to spend the money, buying businesses and expanding the corporate domain. Murphy, too, bought a few businesses. In 1990, he spent $61 million acquiring small properties. At the time, the general market for most media properties was priced too high, he said.
Acquisitions had always been very important to Cap Cities in the development of its growth. Murphy was always on the lookout for media properties, but he remained steadfast in his discipline not to overpay for a company. Cap Cities, with its enormous cash flow, could easily gobble up other media properties, but as BusinessWeek reported, “Murphy would sometimes wait for years until he found the right property. He never made a deal just because he had the resources available to do it.”20 Murphy and Burke also realized that the media business was cyclical, and if a purchase was built on too much leverage, the risk to shareholders would be unacceptable. “Murphy never did a deal that either of us thought was capable of mortally wounding us,” Burke said.21
A company that generates more cash than can be profitably reinvested in its business can buy growth, reduce leverage, or return the money to shareholders. Since Murphy was unwilling to pay the high asking prices for media companies, he chose instead to reduce leverage and buy back stock. In 1986, after the acquisition of ABC, total long-term debt at Cap Cities was $1.8 billion and the debt-to-capital ratio was 48.6 percent. Cash and cash equivalents at 1986 year-end amounted to $16 million. By 1992, long-term debt at the company was $964 million and the debt-to-capital ratio had dropped to 20 percent. Furthermore, cash and cash equivalents increased to $1.2 billion, making the company essentially debt free.
Murphy’s strengthening of Cap Cities’ balance sheet substantially reduced the company’s risk. What he did next substantially increased its value.
From 1985 through 1992, the market value of Cap Cities/ABC grew from $2.9 billion to $8.3 billion. During this period, the company retained $2.7 billion in earnings, thereby creating $2.01 in market value for every $1 reinvested. This accomplishment was especially noteworthy considering that the company endured both a cyclical downturn in earnings, in 1990–1991, and a decline in its intrinsic value from secular changes in the network-broadcasting business. Even so, Berkshire’s investment in Capital Cities/ABC grew from $517 million to $1.5 billion, a 14.5 percent compounded annual rate of return—better than both CBS and the Standard & Poor’s 500 index.
In 1988, Cap Cities announced that it had authorized the repurchase of up to two million shares, 11 percent of the company’s outstanding stock. In 1989, the company spent $233 million purchasing 523,000 shares of stock at an average price of $445—7.3 times the company’s operating cash flow, compared to the asking prices of other media companies that were selling at 10 to 12 times cash flow. The following year, the company purchased 926,000 shares at an average price of $4,777, or 7.6 times operating cash flow. In 1992, the company continued to buy back its stock. That year it purchased 270,000 shares at an average cost of $434 per share, or 8.2 times cash flow. The price it paid for itself, Murphy reiterated, was still less than the price of other advertiser-supported media companies that he and Burke considered attractive. From 1988 through 1992, Cap Cities purchased a total of 1,953,000 shares of stock, investing $866 million.
In November 1993, the company announced a Dutch auction to purchase up to two million shares at prices between $590 and $630 per share. Berkshire participated in the auction, submitting one million of its three million shares. This act alone caused widespread speculation. Was the company unable to find an appropriate acquisition, putting itself up for sale? Was Buffett, by selling a third of his position, giving up on the company? Cap Cities denied the rumors. Opinions surfaced that Buffett would not have tendered stock that surely would have fetched a higher price if indeed the company was for sale. Cap Cities/ABC eventually purchased 1.1 million shares of stock—one million of them from Berkshire—at an average price of $630 per share. Buffett was able to redeploy $630 million without disrupting the marketplace for Cap Cities’ shares while remaining the largest shareholder of the company, owning 13 percent of the shares outstanding.
Buffett has observed the operations and management of countless businesses over the years. But according to him, Cap Cities was the best-managed publicly owned company in the country. To prove his point, when Buffett invested in Cap Cities, he assigned all voting rights for the next 11 years to Murphy and Burke, as long as either one continued to manage the company. And if that was not enough to convince you of the high regard Buffett held for these men, consider this: “Tom Murphy and Dan Burke are not only great managers,” Buffett said, “they are precisely the sort of fellows that you would want your daughter to marry.”22
In the fall of 1988, Donald Keough, president of Coca-Cola, could not help but notice that someone was buying shares of the company’s stock in a big way. Just a year after the 1987 stock market crash, Coca-Cola’s shares were still trading 25 percent below their precrash high. But the share price had finally found a floor because “some mysterious investor was gulping down shares by the caseload.” When Keough discovered that the broker who was doing all the buying hailed from the Midwest, he immediately thought of his friend Warren Buffett, and decided to give him a call.
“Well, Warren, what’s going on?” Keough began. “You don’t happen to be buying any shares of Coca-Cola?” Buffett paused and then said, “It so happens that I am, but I would appreciate it if you would stay quiet about it until I disclose my ownership.”23 If word had ever gotten out that Buffett was buying Coca-Cola shares, it would have created a rush of buying, ultimately driving the share price higher, and he was not done adding to Berkshire’s position.
By the spring of 1989, Berkshire Hathaway shareholders learned that Buffett had spent $1.02 billion buying Coca-Cola shares. He had bet a third of the Berkshire portfolio, and now owned 7 percent of the company. It was the single-largest Berkshire investment to date, and already Wall Street was scratching its head. Buffett had paid five times book value and over 15 times earnings, then a premium to the stock market, for a hundred-year-old company that sold soda pop. What did the Wizard of Omaha see that everyone else missed?
Coca-Cola is the world’s largest beverage company. It sells more than 500 different sparkling and still refreshments in over 200 countries worldwide. Of those 500, 15 are billion-dollar brands, including Coca-Cola, Diet Coke, Fanta, Sprite, Vitaminwater, Powerade, Minute Maid, Simply, Georgia, and Dell Valle.
Buffett’s relationship with Coca-Cola dates back to his childhood. He drank his first Coke when he was five years old, and soon afterward he started the entrepreneurial venture you may remember from Chapter 1, buying six Cokes for 25 cents and reselling them for five cents each. For the next 50 years, he observed the phenomenal growth of Coca-Cola but purchased instead textile mills, department stores, and farming equipment manufacturers. Even in 1986, when he formally announced that Cherry Coke would become the official soft drink of Berkshire Hathaway’s annual meetings, Buffett had still not purchased one share of Coca-Cola. It wasn’t until two years later, in the summer of 1988, that Buffett began to buy.
The business of Coca-Cola is relatively simple. The company purchases commodity inputs and mixes them to manufacture a concentrate that is sold to bottlers, who combine the concentrate with other ingredients. The bottlers then sell the finished product to retail outlets, including minimarts, supermarkets, and vending machines. The company also provides soft drink syrups to restaurants and fast-food retailers, who then sell soft drinks to consumers in cups and glasses.
No other company can match Coca-Cola’s consistent operating history. The business was started in 1886, selling one beverage product. Today, almost 130 years later, Coca-Cola is selling the same beverage—plus a few others. The only significant difference is the company’s size and geographic reach.
At the turn of the twentieth century, the company employed 10 traveling salesmen to cover the entire United States. At that point, the company was selling 116,492 gallons of syrup a year, for annual sales of $148,000. Fifty years after inception, the company was selling 207 million cases of soft drinks annually (having converted sales from gallons to cases). “It would be hard,” Buffett has noted, “to name any company comparable to Coca-Cola and selling, as Coca-Cola does, an unchanged product that can point to a ten-year record anything like Coca-Cola’s.”24 Today, with 1.7 billion servings daily, the Coca-Cola Company is the number-one global provider of beverages, ready-to-drink coffees, juices, and juice drinks.
Shortly after Berkshire’s 1989 public announcement that it owned 6.3 percent of the Coca-Cola Company, Buffett was interviewed by Melissa Turner, a business writer for the Atlanta Constitution. She asked Buffett a question he has been asked often: Why hadn’t he purchased shares in the company sooner? By way of an answer, Buffett related what he was thinking at the time he finally made the decision.
“Let’s say you are going away for ten years,” he said, “and you wanted to make one investment and you know everything that you know now, but you couldn’t change it while you were gone. What would you think about?” Of course the business would have to be simple and understandable. Of course the business would have to have demonstrated a great deal of business consistency over the years. And of course, the long-term prospects would have to be favorable. “If I came up with anything in terms of certainty, where I knew the market was going to grow, where I knew the leader was going to be the leader—I mean worldwide—and where I knew there would be big unit growth, I just don’t know anything like Coke,” Buffett explained. “I’d be relatively sure that when I came back they would be doing a hell of a lot more business than they do now.”25
But why purchase at that particular time? Coca-Cola’s business attributes, as described by Buffett, had existed for several decades. What caught his eye, he said, were the changes occurring at Coca-Cola during the 1980s under the leadership of Roberto Goizueta, chairman and CEO, and president Donald Keough.
Change was critical, and overdue. The 1970s were dismal years for Coca-Cola. The decade was marred by disputes with bottlers, accusations of mistreatment of migrant workers at the company’s Minute Maid groves, environmentalists’ claim that Coke’s one-way containers contributed to the country’s growing pollution problems, and the Federal Trade Commission’s charges that the company’s exclusive franchise system violated the Sherman Antitrust Act. Coca-Cola’s international business was reeling as well. The Arab boycott of Coke, begun when the company issued an Israeli franchise, dismantled years of investment. Japan, where the company’s earnings were growing the fastest, was a battlefield of corporate mistakes. Coke’s 26-ounce take-home bottles were exploding—literally—on store shelves. In addition, Japanese consumers angrily objected to the company’s use of artificial coal-tar coloring in Fanta Grape. When the company developed a new version using real grape skins, the contents fermented and the grape soda was tossed into Tokyo Bay.
Coca-Cola in the 1970s was a fragmented and reactive company rather than an innovator setting the pace within the beverage industry. Despite its problems, the company continued to generate millions of dollars in earnings. But instead of reinvesting in Coca-Cola’s own beverage market, Paul Austin, appointed chairman in 1971 after serving as president since 1962, decided to diversify. He invested in water projects and shrimp farms, despite their slim profit margins. He also purchased a winery. Shareholders bitterly opposed this move, arguing that Coca-Cola should not be associated with alcohol. To deflect criticism, Austin directed unprecedented amounts of money for advertising campaigns.
Meanwhile Coca-Cola earned 20 percent on equity, but pretax margins were slipping. The market value of the company at the end of the bear market of 1974 was $3.1 billion. Six years later, that value had increased to $4.1 billion. In other words, from 1974 to 1980, the company’s market value increased at an average annual rate of 5.6 percent, significantly underperforming the S&P 500 index. For every dollar the company retained in those six years, it created only $1.02 in market value.
Coca-Cola’s corporate woes were exacerbated by Austin’s intimidating and unapproachable behavior.26 To make matters worse, his wife, Jeane, was a disruptive influence within the company. She redecorated corporate headquarters with modern art, shunning the company’s classic Norman Rockwell paintings and even using a corporate jet for her art-buying trips. But it was her last order that contributed to her husband’s downfall.
In May 1980, Mrs. Austin ordered the company’s park closed to employee luncheons. Their food droppings, she complained, attracted pigeons on the well-manicured lawns. Employee morale hit an all-time low. Robert Woodruff, the company’s 91-year-old patriarch, who had led Coca-Cola from 1923 until 1955 and was still chairman of the board’s finance committee, had heard enough. He demanded Austin’s resignation and replaced him with Roberto Goizueta.
Goizueta, raised in Cuba, was Coca-Cola’s first foreign chief executive officer. He was as outgoing as Austin was reticent. One of his first acts was to bring together Coca-Cola’s top 50 managers for a meeting in Palm Springs, California. “Tell me what we’re doing wrong,” he said. “I want to know it all and once it is settled, I want 100 percent loyalty. If anyone is not happy, we will make you a good settlement and say goodbye.”27 From this meeting evolved the company’s “Strategy for the 1980s,” a 900-word pamphlet outlining the corporate goals for Coca-Cola.
Goizueta encouraged his managers to take intelligent risks. He wanted Coca-Cola to initiate action rather than to be reactive. He began by cutting costs, and he demanded that any business that Coca-Cola owned must optimize its return on assets. These actions translated, immediately, into increasing profit margins.
In 1980, Coca-Cola’s pretax profit margins were a low 12.9 percent. Margins had been falling for five straight years and were substantially below the company’s 1973 margins of 18 percent. In Goizueta’s first year, pretax margins rose to 13.7 percent; by 1988, when Buffett bought his Coca-Cola shares, margins had climbed to a record 19 percent.
In “Strategy for the 1980s,” Goizueta pointed out that the company would divest any business that no longer generated acceptable returns on equity. Any new business venture must have sufficient real growth potential to justify an investment. Coca-Cola was no longer interested in battling for share in a stagnant market. “Increasing earnings per share and effecting increased return on equity are still the name of the game,” Goizueta announced.28 His words were followed by actions: Coca-Cola’s wine business was sold to Seagram’s in 1983.
Although the company had earned a respectable 20 percent on equity during the 1970s, Goizueta was not impressed. He demanded better returns, and the company obliged. By 1988, Coca-Cola’s return on equity had increased to 31 percent.
By any measurement, Goizueta’s Coca-Cola was doubling and tripling the financial accomplishments of Austin’s Coca-Cola. The results could be seen in the market value of the company. In 1980, that value was $4.1 billion. By the end of 1987, even after the stock market crash in October, market value had risen to $14.1 billion. In just seven years, Coca-Cola’s market value increased at an average annual rate of 19.3 percent. For every dollar Coca-Cola retained during this period, it gained $4.66 in market value.
Goizueta’s strategy for the 1980s pointedly included shareholders. “We shall, during the next decade, remain totally committed to our shareholders and to the protection and enhancement of their investment,” he wrote. “In order to give our shareholders an above-average total return on their investment,” he explained, “we must choose businesses that generate returns in excess of inflation.”29
Goizueta not only had to grow the business, which required capital investment; he also was obliged to increase shareholder value. To do so, Coca-Cola, by increasing profit margins and return on equity, was able to pay dividends while simultaneously reducing the dividend payout ratio. Dividends to shareholders in the 1980s were increasing 10 percent per year while the payout ratio was declining from 65 percent to 40 percent. This enabled Coca-Cola to reinvest a greater percentage of the company’s earnings to help sustain its growth rate while not shortchanging shareholders.
Under Goizueta’s leadership, Coca-Cola’s mission statement became crystal clear; management’s primary objective was to maximize shareholder value over time. To do so, the company focused on the high-return soft drink business. If successful, the evidence would be an increase in the growth of cash flow and increased return on equity, and ultimately an increased total return to shareholders.
The growth in net cash flow not only allowed Coca-Cola to increase its dividend to shareholders, but also enabled the company to initiate its first ever buy-back program. In 1984, Goizueta announced that the company would repurchase six million shares of stock in the open market. Repurchasing stock is rational only if the intrinsic value of the company is higher than the market price. The strategic changes initiated by Goizueta, with emphasis on increasing the return on equity for shareholders, suggested to him that Coca-Cola had reached that tipping point.
In 1973, owner earnings (net income plus depreciation minus capital expenditures) were $152 million. By 1980, owner earnings were $262 million, an 8 percent annual compounded growth rate. From 1981 through 1988, owner earnings grew from $262 million to $828 million, a 17.8 percent average annual compounded growth rate.
The growth in owner earnings was reflected in the share price of Coca-Cola. This is particularly obvious if we look at 10-year periods. From 1973 to 1982, the total return of Coca-Cola grew at an average annual rate of 6.3 percent. The following 10 years, from 1983 to 1992, when Goizueta’s management approach was clearly visible, the average annual total return for the stock was 31.1 percent.
When Goizueta took over Coca-Cola, one of his first moves was to jettison the unrelated businesses that Paul Austin had developed, and return the company to its core business: selling syrup. It was a clear demonstration of Coca-Cola’s ability to resist the institutional imperative.
Reducing the company to a single-product business was undeniably a bold move. What made Goizueta’s strategy even more remarkable was his willingness to take action at a time when others in the industry were doing the exact opposite. Several leading beverage companies were investing their profits in other unrelated businesses. Anheuser-Busch used the profits from its beer business to invest in theme parks. Brown Forman, a producer and distributor of wine and spirits, invested its profits in china, crystal, silver, and luggage businesses, all of which had much lower returns. Seagram Company, Ltd., a global spirits and wine business, bought Universal Studios. Pepsi, Coca-Cola’s chief beverage rival, bought snack businesses (Frito-Lay) and restaurants, including Taco Bell, Kentucky Fried Chicken, and Pizza Hut.
It is important to note that not only did Goizueta’s action focus the company’s attention on its largest and most important product, but it worked to reallocate the company’s resources into its most profitable business. Because the economic returns of selling syrup far outweighed the economic returns of the other businesses, the company was now reinvesting its profits in its highest-return business.
When Buffett first purchased Coca-Cola in 1988, people asked: “Where is the value of Coke?” The company’s stock price was 15 times earnings and 12 times cash flow—a 30 percent and 50 percent premium to the market average. Buffett paid five times book value for a company with 6.6 percent earnings yield at a time when long-term bonds were yielding 9 percent. He was willing to do that because of Coke’s extraordinary level of economic goodwill. The company was earning 31 percent on equity while employing relatively little in capital investment. Buffett has explained that price tells you nothing about value. The value of Coca-Cola, he said, like that of any other company, is determined by the total owner earnings expected to occur over the life of the business, discounted by the appropriate interest rate.
In 1988, owner earnings of Coca-Cola equaled $828 million. The 30-year U.S. Treasury bond (the risk-free rate) at that time traded near a 9 percent yield. Coca-Cola’s 1988 owner earnings, discounted by 9 percent, would produce an intrinsic value of $9.2 billion. When Buffett purchased Coca-Cola, the market value was $14.8 billion. At first glance this seems to suggest that Buffett might have overpaid for the company. But remember that the $9.2 billion represents the discounted value of Coca-Cola’s then-current owner earnings. If buyers were willing to pay a price for Coca-Cola that was 60 percent higher than $9.2 billion, it must have been because they perceived part of the value of Coca-Cola to be its future growth opportunities.
Analyzing Coca-Cola, we find that owner earnings from 1981 through 1988 grew at a 17.8 percent annual rate—faster than the risk-free rate of return. When this occurs, analysts use a two-stage discount model. It permits them to calculate future earnings when a company has extraordinary growth for a limited number of years, followed by a period of constant growth at a slower rate.
We can use this two-stage process to calculate the 1988 present value of the company’s future cash flows. In 1988, Coca-Cola’s owner earnings were $828 million. If we assume that Coca-Cola would be able to grow owner earnings at 15 percent per year for the next 10 years (a reasonable assumption, since that rate is lower than the company’s previous seven-year average), by year 10 owner earnings would equal $3.349 billion. Let’s further assume that, starting in year 11, that growth rate will slow to 5 percent per year. Using a discount rate of 9 percent (the long-term bond rate at the time), we can back-calculate the intrinsic value of Coca-Cola in 1988 to be $48.377 billion.
We can repeat this exercise using different growth rate assumptions. If we assume that Coca-Cola can grow owner earnings at 12 percent for 10 years, followed by 5 percent growth, the present value of the company, discounted at 9 percent, would be $38.163 billion. At 10 percent growth for 10 years and 5 percent thereafter, the value would be $32.497 billion. And even if we assume that Coca-Cola could only grow at a steady state of 5 percent throughout, the company would still be worth at least $20.7 billion.
In June 1988, the price of Coca-Cola was approximately $10 per share (split adjusted). Over the next 10 months, Buffett acquired 93,400,000 shares, for a total investment of $1.023 billion. His average cost per share was $10.96. At the end of 1989, Coca-Cola represented 35 percent of Berkshire’s common stock portfolio.
From the time Goizueta took control of Coca-Cola in 1980, the company’s stock price had increased every year. In the five years before Buffett purchased his first shares, the average annual gain in share price was 18 percent. The company’s fortunes were so good that Buffett was unable to purchase any shares at distressed prices. Still, he charged ahead. Price, he reminds us, has nothing to do with value.
The stock market’s value of Coca-Cola in 1988 and 1989, during Buffett’s purchase period, averaged $15.1 billion. But by Buffett’s estimation, the intrinsic value of the company was anywhere from $20.7 billion (assuming a 5 percent growth in owner earnings) to $32.4 billion (assuming 10 percent growth), or $38.1 billion (assuming 12 percent growth), or $48.3 billion (assuming 15 percent growth). Buffett’s margin of safety—the discount to intrinsic value—could be as low as a conservative 27 percent or as high as 70 percent.
The best business to own, says Buffett, is one that, over a long period of time, can employ ever-larger amounts of capital at sustainably high rates of return. In Buffett’s mind, this was the perfect description of Coca-Cola. Ten years after Berkshire began investing in Coca-Cola, the market value of the company had grown from $25.8 billion to $143 billion. Over that time period, the company produced $26.9 billion in profits, paid out $10.5 billion in dividends to shareholders, and retained $16.4 billion for reinvestment. For every dollar the company retained, it created $7.20 in market value. At year-end 1999, Berkshire’s original $1.023 billion investment in Coca-Cola was worth $11.6 billion. The same amount invested in the S&P 500 index would have been worth $3 billion.
In 1990, General Dynamics was the country’s second-largest defense contractor behind McDonnell Douglas Corporation. General Dynamics provided missile systems (Tomahawk, Sparrow, Stinger, and other advanced cruise missiles) in addition to air defense systems, space-launched vehicles, and fighter planes (F-16s) for the U.S. armed forces. In 1990, the company had combined sales of more than $10 billion. By 1993, sales had dropped to $3.5 billion. Despite that, shareholder value during this period increased sevenfold.
In 1990, the Berlin Wall crumbled, signaling the beginning of the end of the long and expensive Cold War. The following year, communism collapsed in the Soviet Union. With each hard-earned victory, from World War I to the Vietnam War, the United States had to reshape the massive concentration of its defense resources. Now that the Cold War was over, the U.S. military-industrial complex was in the midst of another reorganization.
In January 1991, General Dynamics appointed William Anders as chief executive. At the time, the share price was at a decade low of $19. Initially, Anders attempted to convince Wall Street that even with a shrinking defense budget, the company could earn higher valuations. Hoping to remove any financial uncertainty that would prejudice analysts, he began to restructure the company. He cut capital expenditures and research development by $1 billion, cut employment by the thousands, and instituted an executive compensation program that was based on the performance of General Dynamics’ share price.
It was not long before Anders began to realize that the defense industry had fundamentally changed and that to be successful, General Dynamics would have to take more dramatic steps than just pinching pennies. There simply was not enough defense business to go around. A smaller defense budget would ultimately require companies to downsize, diversify into nondefense businesses, or dominate what little defense business was available.
In October 1991, Anders commissioned a consultant’s study of the defense industry. The conclusions were sobering: When defense companies acquired nondefense businesses, failures occurred 80 percent of the time. As long as the defense industry was burdened with overcapacity, none of the defense companies would achieve efficiencies. Anders concluded that to be successful, General Dynamics would have to rationalize its business. He decided that General Dynamics would keep only those businesses that (1) demonstrated a market acceptance of its franchise-like product and (2) could achieve critical mass, the balance between research and development and production capacity that produces economies of scale and financial strength. Where critical mass could not be achieved, Anders said, the business would be sold.
Initially, Anders believed that General Dynamics would focus on its four core operations: submarines, tanks, aircraft, and space systems. These businesses were market leaders, and Anders figured they would remain viable even in a shrinking defense market. The rest of General Dynamics’ businesses would be sold. So, in November 1991, General Dynamics sold its Data Systems to Computer Sciences for $200 million. The next year the company sold Cessna Aircraft to Textron for $600 million and its missile business to Hughes Aircraft for $450 million. In less than six months, the company raised $1.25 billion by selling noncore businesses.
Anders’s actions woke up Wall Street. General Dynamics’ share price in 1991 rose 112 percent. What Anders did next got Buffett’s attention.
With the cash holdings, Anders declared that the company would first meet its liquidity needs and then bring down debt to ensure financial strength. After reducing debt, General Dynamics was still generating cash well in excess of its needs. Knowing that adding capacity to a shrinking defense budget did not make sense and that diversification into nondefense businesses invited failure, Anders decided to use the excess cash to benefit shareholders. In July 1992, under the terms of its Dutch auction, General Dynamics purchased 13.2 million shares at prices between $65.37 and $72.25, reducing its shares outstanding by 30 percent.
On the morning of July 22, 1992, Buffett called Anders to tell him that Berkshire had purchased 4.3 million shares of General Dynamics. Buffett told Anders that he was impressed with General Dynamics and that he bought the shares for investment purposes. In September, Buffett granted General Dynamics’ board a proxy to vote Berkshire’s shares so long as Anders remained chief executive.
Of all Berkshire’s common stock purchases, none has caused as much confusion as General Dynamics. It had none of the traditional markings of Buffett’s earlier purchases. It was not a company that was simple and understandable, it was not a consistent performer, and it did not have favorable long-term prospects. Not only was the company in an industry that was controlled by the government (90 percent of sales came from government contracts), but that industry was shrinking in size. General Dynamics had pitiful profit margins and below-average returns on equity. Furthermore, its future cash flows were unknown, so how could Buffett determine its value? The answer is that Buffett did not initially purchase General Dynamics as a long-term common stock holding. He purchased it as an arbitrage opportunity, so the usual financial and business requirements did not apply.
“We were lucky in our General Dynamics purchase,” Buffett wrote. “I had paid little attention to the company until last summer, when it announced it would repurchase about 30 percent of its shares by way of Dutch tender. Seeing an arbitrage opportunity, I began buying the stock for Berkshire, expecting to tender our holdings at a small profit.”30
But he changed his mind. The original plan had been to tender Berkshire’s shares at the Dutch auction “but then I began studying the company,” said Buffett, “and the accomplishments of Bill Anders in the brief time he’d been CEO. And what I saw made my eyes pop. Bill had a clearly articulated and rational strategy; he had been focused and imbued with a sense of urgency in carrying it out; and the results were truly remarkable.”31 Buffett abandoned his thoughts of arbitraging General Dynamics and instead decided to become a long-term shareholder.
Clearly, Buffett’s investment in General Dynamics was a testament to Bill Anders’s ability to resist the institutional imperative. Although critics have argued that Anders liquidated a great company, Anders argues he simply monetized the unrealized value of the company. When he took charge in 1991, the stock of General Dynamics was trading at a 60 percent discount to book value. In the 10 previous years, General Dynamics had returned to shareholders a compounded annual return of 9 percent compared to the 17 percent for the 10 other defense companies and a 17.6 percent return for the Standard & Poor’s 500 index. Buffett saw a company that was trading below book value, generating cash flow, and embarking on a divestiture program. Additionally, and most importantly, management was shareholder-oriented.
Although General Dynamics had earlier thought that the aircraft and space systems divisions would remain core holdings, Anders decided to sell these businesses. The aircraft business was sold to Lockheed. At the time, General Dynamics, Lockheed, and Boeing were one-third partners in the development of the next generation of tactical fighter, the F-22. By buying General Dynamics’ aircraft division, Lockheed acquired the mature F-16 business and became a two-thirds partner with Boeing on the F-22 project. The space systems business was sold to Martin Marietta, maker of the Titan family of space launch vehicles. Together, selling the two businesses provided General Dynamics with $1.72 billion.
Flush with cash, the company again returned the money to its shareholders. In April 1993, the company issued a $20 per share special dividend to shareholders. In July the company issued an $18 special dividend, and in October gave $12 per share to its owners. In 1993, the company returned $50 in special dividends and raised the quarterly dividend from $0.40 to $0.60 per share. From July 1992 through the end of 1993, for its investment of $72 per share, Berkshire received $2.60 in common dividends, $50 in special dividends, and a share price that rose to $103. It amounted to a 116 percent return over 18 months. Not surprisingly, over this time period General Dynamics not only outperformed its peer group, but it also soundly beat the Standard & Poor’s 500 index.
If General Dynamics was the most confusing investment Buffett ever made, then the Wells Fargo & Company investment would certainly qualify as the most controversial. In October 1990, Buffett announced that Berkshire had purchased five million shares of Wells Fargo, investing $289 million in the company at an average $57.88 per share. Berkshire was now the largest shareholder of the bank, owning 10 percent of the shares outstanding.
Earlier in the year, Wells Fargo had traded as high as $86 per share, but then investors began abandoning California banks and thrifts. They feared the recession that was gripping the West Coast would soon cause wide loan losses in the commercial and residential real estate market. Since Wells Fargo had the most commercial real estate of any California bank, investors sold their stock and short sellers added to the downside pressure. The short interest in Wells Fargo’s stock jumped 77 percent in the month of October, about the same time that Buffett began purchasing shares in the company.
In the months following the announcement that Berkshire had become a major shareholder, the battle for Wells Fargo resembled a heavyweight fight. In one corner, Buffett was the bull, betting $289 million that Wells Fargo would increase in value. In the other corner, short sellers were the bears, betting that Wells Fargo, already down 49 percent for the year, was destined to fall further. The Feshbach brothers, the nation’s biggest short sellers, were betting against Buffett. “Wells Fargo is a dead duck,” said Tom Barton, a Dallas money manager for the Feshbachs. “I don’t think it’s right to call them a bankruptcy candidate but I think it’s a teenager.”32 By that, Barton meant he thought that Wells Fargo would trade down into the teens. Buffett “is a famous bargain hunter and long-term investor,” said George Salem, an analyst with Prudential Securities, but “California could become another Texas.”33 Salem was referring to the bank failures that had occurred in Texas during the decline in energy prices. Buffett “won’t have to worry about who spends his fortune much longer,” said John Liscio at Barron’s, “not if he keeps trying to pick a bottom in bank stocks.”34
Buffett was very familiar with the business of banking. In 1969, Berkshire Hathaway purchased 98 percent of the holdings of Illinois National Bank and Trust Company. Before the Bank Holding Act required Berkshire to divest its interest in the bank in 1979, Buffett reported the sales and earnings of the bank each year in Berkshire’s annual reports. The bank took its place beside Berkshire’s other controlled holdings.
Just as Jack Ringwalt helped Buffett understand the intricacy of the insurance business, Gene Abegg, who was chairman of Illinois National Bank, taught Buffett about the banking business. What he learned was banks were profitable businesses if they issued loans responsibly and curtailed costs. “Our experience has been that the manager of an already high-cost operation frequently is uncommonly resourceful in finding new ways to add to overhead,” said Buffett, “while the manager of a tightly run operation usually continues to find additional methods to curtail costs, even when costs are already well below those of its competitors. No one has demonstrated this latter ability better than Gene Abegg.”35
Wells Fargo is not Coca-Cola, Buffett says. Under most circumstances, it is hard to imagine how Coca-Cola could fail as a business. But the banking business is different. Banks can fail and have, on many occasions. Most bank closures can be traced to management mistakes, Buffett points out, usually when they foolishly issue loans that a rational banker would never have considered. When assets are 20 times equity, which is common in the banking industry, any managerial foolishness involving even a small amount of assets can destroy a company’s equity.
Still, it is not impossible for banks to be good investments, Buffett says. If management does its job, banks can generate a 20 percent return on equity. Although this is below what a Coca-Cola might earn, it is above the average return for most businesses. It is not necessary to be number one in your industry if you are a bank, Buffett explains. What counts is how you manage your assets, liabilities, and costs. Like insurance, banking is very much a commodity business. And as we know, in a commodity-like business, the actions of management are frequently the most distinguishing trait. In this respect, Buffett picked the best management team in banking. “With Wells Fargo,” he states, “we think we have obtained the best managers in the business, Carl Reichardt and Paul Hazen. In many ways, the combination of Carl and Paul reminds me of another—Tom Murphy and Dan Burke at Capital Cities/ABC. Each pair is stronger than the sum of the parts.”36
When Carl Reichardt became chairman of Wells Fargo in 1983, he began to transform the sluggish bank into a profitable business. From 1983 through 1990, Wells Fargo averaged returns of 1.3 percent on assets and 15.2 percent on equity. By 1990, it had become the tenth-largest bank in the country, with $56 billion in assets. Reichardt, like many managers that Buffett admires, is rational. Although he had not instigated stock buy-back programs or passed along special dividends, all of which reward shareholders, he does run Wells Fargo for the benefit of its owners. Like Tom Murphy at Capital Cities/ABC, he was legendary when it came to controlling costs. Once costs were under control, Reichardt never let up. He constantly searched for ways to improve profitability.
One measure of a bank’s operating efficiency is the ratio of its operating (i.e., noninterest) expense to net interest income.37 Wells Fargo’s operating efficiencies were 20 to 30 percent better than First Interstate or Bank of America. Reichardt manages Wells Fargo like an entrepreneur. “We try to run this company like a business,” he said. “Two and two is four. It’s not seven or eight.”38
When Buffett was buying Wells Fargo in 1990, the bank ended that year with the highest percentage of commercial real estate loans of any major bank in the country. Wells Fargo’s $14.5 billion in commercial loans was five times its equity. Because California’s recession was worsening, analysts figured that a large portion of the bank’s commercial loans would sour. It was this that caused Wells Fargo’s share price to decline in 1990 and 1991.
In the wake of the Federal Savings and Loan Insurance Corporation (FSLIC) debacle, bank examiners rigorously reviewed Wells Fargo’s loan portfolio. They pressured the bank to set aside $1.3 billion in reserves for bad loans in 1991 and another $1.2 billion the following year. Because reserves were set aside every quarter, investors began to feel squeamish with each subsequent announcement. Instead of taking one large charge for loan reserves, the bank strung out the charges over a period of two years. Investors began to wonder whether the bank would ever reach the end of its problem loans.
After Berkshire announced its ownership of Wells Fargo in 1990, the stock price climbed briefly, reaching $98 in early 1991 and providing Berkshire with a $200 million profit. But then, in June 1991, the bank announced another charge to reserves and the stock price fell 13 points in two days to $74. Although the stock price recovered slightly in the fourth quarter of 1991, it became clear that Wells Fargo would have to take yet another charge against earnings for additions to its loan loss reserves. At year-end, the stock closed at $58 per share. After a roller coaster ride, Berkshire’s investment was breaking even. “I underestimated the severity of both the California recession and the real estate troubles of the company,” Buffett confessed.39
In 1990, Wells Fargo earned $711 million, an 18 percent increase over 1989. The next year, because of loan loss reserves, the bank earned $21 million. A year later, earnings increased slightly to $283 million—still less than half the earnings just two years earlier. Not surprisingly, there is an inverse relationship between a bank’s earnings and its loan loss provisions. But if you remove Wells Fargo’s loan loss provisions from the income statement, you uncover a company with dynamic earnings power. Since 1983, the bank’s net interest income had grown at an 11.3 percent rate, and its noninterest income (investment fees, trust income, deposit charges) had grown at a 15.3 percent rate. If you exclude the unusual loan loss provisions in 1991 and 1992, the bank would have had approximately $1 billion in earnings power.
The value of a bank is the function of its net worth plus its projected earnings as a going concern. When Berkshire Hathaway began purchasing Wells Fargo in 1990, the company, in the previous year, had earned $600 million. The average yield on the 30-year U.S. government bond in 1990 was approximately 8.5 percent. To remain conservative, we can discount Wells Fargo’s 1989 earnings of $600 million by 9 percent and value the bank at $6.6 billion. If the bank never earned another dime over $600 million in annual earnings during the next 30 years, it was worth at least $6.6 billion. When Buffett purchased Wells Fargo in 1990, he paid $58 per share for its stock. With 52 million shares outstanding, this was equivalent to buying the company for $3 billion, a 55 percent discount to its value.
Of course, the debate over Wells Fargo centered on whether the company, after taking into consideration all of its loan problems, even had earnings power. The short sellers said it did not; Buffett said it did. He knew that ownership of Wells Fargo was not riskless. This is how he rationalized his purchase: “California banks face the specific risk of a major earthquake, which might wreak enough havoc on borrowers to in turn destroy the banks lending to them,” Buffett said. “A second risk is systemic—the possibility of a business contraction or financial panic so severe that it would endanger almost every highly-leveraged institution, no matter how intelligently run.”40 Now, the possibility of these two events occurring, in Buffett’s judgment, was low. But there still remained one viable risk, he said. “The market’s major fear of the moment is that West Coast real estate values will tumble because of overbuilding and deliver huge losses to banks that financed the expansion. Because it is a leading real estate lender, Wells Fargo is thought to be particularly vulnerable.”41
Buffett knew that Wells Fargo earned $1 billion pretax annually after expensing an average $300 million for loan losses. He figured if 10 percent of the bank’s $48 billion in loans (not just commercial loans, but all of the bank’s loans) were problem loans in 1991 and produced losses, including interest, averaging 30 percent of the principal, Wells Fargo would break even. He calculated that this was unlikely. Even if Wells Fargo earned no money for a year, that would not be distressing. “At Berkshire,” Buffett said, “we would love to acquire businesses or invest in capital projects that produce no return for a year, but that could then be expected to earn 20% on growing equity.”42 The attraction of Wells Fargo intensified when Buffett was able to purchase shares at a 50 percent discount to value.
“Banking doesn’t have to be a bad business, but it often is,” Buffett said, adding that “bankers don’t have to do stupid things, but they often do.”43 He describes a high-risk loan as any loan made by a stupid banker. When Buffett purchased Wells Fargo, he bet that Reichardt was not a stupid banker. “It’s all a bet on management,” said Charlie Munger. “We think they will fix the problems faster and better than other people.”44 Berkshire’s bet paid off. By the end of 1993, Wells Fargo’s price per share had reached $137.
“I find that a long-term familiarity with a company and its products is often helpful in evaluating it,” Buffett said.45 With the exception of selling bottles of Coca-Cola for a nickel, delivering copies of the Washington Post, and recommending that his father’s clients buy shares of GEICO, Buffett has had a longer history with American Express than any other company Berkshire owns. You may recall that in the mid-1960s, the Buffett Limited Partnership invested 40 percent of its assets in American Express shortly after the company’s losses in the salad oil scandal. Thirty years later, Berkshire accumulated 10 percent of American Express shares for $1.4 billion.
Although the company has weathered a cycle of changes, American Express is essentially the same business it was when Buffett first purchased the company in his Partnership. There were three divisions. Travel Related Services (TRS), which issues the American Express charge card and American Express Travelers Cheques, contributed about 72 percent of American Express’s sales. American Express Financial Advisors (formerly IDS Financial Services), a financial planning, insurance, and investment product division, contributed 22 percent of sales. The American Express Bank contributed a modest 5 percent of sales. The bank had long been the local representative for the American Express card with a network of 87 offices in 37 countries worldwide.
American Express Travel Related Services continues to be a predictable provider of profits. The division had always generated substantial owner earnings and easily funded the company’s growth. But when a company generates more cash than it requires for operations, it often becomes a test of management to allocate this capital responsibly. Some managers pass this test by investing only that capital that is required and returning the balance to the company’s owners either by increasing the dividend or by repurchasing shares. Other managers, unable to resist the institutional imperative, constantly find ways to spend cash and expand the corporate empire. Unfortunately, this was the fate of American Express for several years under the leadership of James Robinson.
Robinson’s plan was to use TRS’s excess cash to acquire related businesses and thus build American Express into a financial services powerhouse. IDS proved to be a profitable purchase. However, Robinson’s purchase of Shearson-Lehman was disappointing. Not only was Shearson unable to fund itself, but it also required increasing amounts of TRS’s excess cash for its own operation. Over time, Robinson invested $4 billion in Shearson. It was this financial drain that prompted Robinson to contact Buffett. Berkshire purchased $300 million in preferred shares. Although Buffett was willing to invest in American Express at that time via preferred shares, it was not until rationality finally surfaced at the company that he felt confident to become a common stock shareholder.
It is no secret that the company’s crown jewel is the famed American Express Card. What seemed to be lacking at American Express was a management team that recognized and appreciated the economics of this business. Fortunately, this realization occurred in 1992, when Robinson unceremoniously resigned and Harvey Golub became chief executive. Golub, striking a familiar tone with Buffett, began to use terms such as franchise and brand value when he referred to the American Express Card. Golub’s immediate task was to strengthen the brand awareness of TRS and shore up the capital structure at Shearson-Lehman to ready it for sale.
Over the next two years, Golub began the process of liquidating American Express’s underperforming assets and restoring profitability and high returns on equity. In 1992, Golub initiated a public offering for First Data Corporation (the company’s information data service division), which netted American Express over $1 billion. The following year, the company sold The Boston Company, its money management division, to Mellon Bank for $1.5 billion. Soon after, Shearson-Lehman was separated into two businesses. Shearson’s retail accounts were sold and Lehman Brothers was spun off to American Express shareholders via a tax-free distribution, but not before Golub had to pump a final $1 billion into Lehman.
By 1994, American Express was beginning to show signs of its old profitable self. The resources of the company were now firmly behind TRS. The goal of management was to build the American Express Card into the “world’s most respected service brand.” Every communication from the company emphasized the franchise value of the name American Express. Even IDS Financial Services was renamed American Express Financial Advisors.
Now that everything was in place, Golub set the financial targets for the company: to increase earnings per share by 12 to 15 percent annually and to achieve an 18 to 20 percent return on equity. Then American Express, in September 1994, issued a statement that demonstrated clearly the rationality of the company’s new management. Subject to market conditions, the board of directors authorized management to repurchase 20 million shares of its common stock. That was music to Buffett’s ears.
During the summer of 1994, Buffett converted Berkshire’s preferred issue into American Express common stock. Soon thereafter, he began to acquire even more shares of common stock. By year-end, Berkshire owned 27 million shares at an average price of $25 per share. With the completion of the stock purchase plan it had announced in the fall of 1994, the following spring American Express announced it would repurchase an additional 40 million shares, representing 8 percent of the total stock outstanding.
Clearly, American Express was a changed company. After jettisoning Shearson-Lehman with its massive capital needs, American Express had a powerful ability to generate excess cash. For the first time, the company had more capital and more shares than needed. Buffett, appreciating the economic changes that were under way at American Express, dramatically increased Berkshire’s position in the company. By March 1995, he had added another 20 million shares, bringing Berkshire’s ownership of American Express to slightly less than 10 percent.
Since 1990, the noncash charges, depreciation, and amortization have roughly equaled American Express’s acquisition of land, buildings, and equipment. When depreciation and amortization charges approximate capital spending, owner earnings equal net income. However, because of the company’s erratic history, it is difficult to ascertain the growth rate of American Express’s owner earnings. Under these circumstances, it is best to use a very conservative growth projection.
By the end of 1994, reflecting the results of American Express’s net of the subsidiaries sold in 1993, the company’s owner earnings were approximately $1.4 billion. Golub’s goal, you will remember, was to grow earnings at a 12 to 15 percent rate going forward. Using a 10 percent growth in earnings for the next 10 years followed by a 5 percent residual growth thereafter (which is decidedly below management’s forecast) and discounting the earnings by 10 percent (which is a conservative discount factor considering the 30-year U.S. Treasury bond was yielding 8 percent), the intrinsic value of American Express was $43.4 billion, or $87 per share. If the company was able to grow its earnings at 12 percent, the intrinsic value of American Express was closer to $50 billion, or $100 per share. At the more conservative valuation, Buffett was purchasing American Express at a 70 percent discount to intrinsic value—a significant margin of safety.
When Buffett announced during a CNBC interview in October 2011 that Berkshire Hathaway had been purchasing shares of IBM, I am sure more than a few Berkshire shareholders were scratching their heads. After all, this was the man who had repeatedly confessed that he was not interested in buying technology companies. “I could spend all my time thinking about technology for the next year and still not be the 100th, 1,000th or even the 10,000th smartest guy in the country in analyzing those businesses,” he once said.46
What prevented Buffett from buying technology companies was not that he didn’t understand these companies; he understood them all too well. What had always troubled him was the difficulty of predicting their future cash flows. The constant disruption and innovation that is inherent in the industry made the life spans of technology franchises very short. Buffett could see a future that included Coca-Cola, Wells Fargo, American Express, Johnson & Johnson, Procter & Gamble, Kraft Foods, and Wal-Mart, and think confidently. A future that included Microsoft, Cisco, Oracle, Intel, and—so we thought—IBM was just too unpredictable.
But by year-end 2011, Berkshire Hathaway had purchased 63.9 million shares of IBM, about 5.4 percent of the company. It was a bold $10.8 billion purchase, the single biggest purchase of an individual stock Buffett had ever made.
When Buffett introduced the IBM purchase to Berkshire shareholders in the 2011 annual report, many might have thought they would be getting a crash course on the competitive advantages of IBM’s advance information processing technology. But what they got instead was a tutorial on the value of common stock repurchases and how to think intelligently about this corporate strategy over the long term.
Buffett began, “All business observers know, CEOs Lou Gerstner and Sam Palmisano did a super job in moving IBM from near-bankruptcy twenty years ago to its prominence today. Their operational accomplishments were truly extraordinary.”47 It is hard to imagine that 20 years ago, the hundred-year-old IBM was near collapse. But in 1992, the company lost $5 billion, the most money any U.S. company had ever lost in a single year. The next year Lou Gerstner was brought in to turn the company around. In his book Who Says Elephants Can’t Dance? (HarperCollins, 2002), Gerstner outlined his strategies, including selling low-margin hardware technology assets and moving more into software and services. Later, when Sam Palmisano became CEO in 2002, he sold off the personal computer business and kept IBM growing for the next 10 years by focusing its business on services, the Internet, and software.
“Their financial management was equally brilliant,” Buffett continued. “I can think of no major company that has had better financial management, a skill that has materially increased the gains enjoyed by IBM shareholders. The company has used debt wisely, made value-adding acquisitions almost exclusively for cash and aggressively repurchased its own stock.”48
In 1993, IBM had 2.3 billion shares outstanding. Ten years later, when Gerstner retired and Palmisano became CEO, the company had 1.7 billion shares outstanding. Over the 10 years, Gerstner bought back 26 percent of the shares outstanding while raising the dividends by 136 percent. This lesson was not lost on Palmisano. During his 10-year reign as CEO, IBM reduced shares outstanding from 1.7 billion to 1.1 billion, a 36 percent reduction. Combined, Gerstner and Palmisano repurchased over half of the shares outstanding. And if that was not enough, Palmisano, in the decade he ran the company, increased the dividend from $0.59 to $3.30, a 460 percent increase.
Of the four technology horsemen—IBM, Microsoft, Intel, and Cisco—only one company has seen its recent share price surpass its 1999 high—the peak of the tech bubble. And that is IBM. At year-end 1999, IBM was selling at $112 per share and by the end of 2012 it was trading for $191. Compare this to Cisco ($54 to $19), Intel ($42 to $20), and Microsoft ($52 to $27). It was not that IBM grew the company faster than the others, but rather that its share price went up because it grew the per-share value faster. Between 1999 and 2012, Microsoft reduced shares by 19 percent and Intel and Cisco both reduced shares outstanding by 23 percent, but IBM reduced its shares outstanding by 36 percent.
Remember Buffett’s notion: that after he begins buying shares in a company, he likes the stock market to delay its recognition, for that gives him the opportunity to buy more shares at bargain prices. The same is true for a company that is in the midst of a share repurchase program. “When Berkshire buys stock in a company that is repurchasing shares, we hope for two events: First, we have the normal hope that earnings of the business will increase at a good clip for a long time to come; and second, we also hope that the stock underperforms in the market for a long time as well.” As Buffett explains, IBM will likely spend $50 billion over the next five years to repurchase stock. He then asks, “What should a long-term shareholder, such as Berkshire, cheer for during that period? We should wish for IBM’s stock price to languish throughout the five years.”49
In a world that is obsessed with short-term performance, wishing for a stock to underperform the market for a long time sounds backward, to say the least. But if one is truly a long-term investor, such thinking is actually quite rational. Buffett walks us through the math. “If IBM’s stock price averages say $200, the company will acquire 250 million shares for its $50 billion. There would consequently be 910 million shares outstanding and we would own about 7% of the company. If the stock conversely sells for an average of $300 during the five-year period, IBM will acquire only 167 million shares. That would leave about 990 million shares outstanding after five years, of which we would own 6.5%.”50 The difference to Berkshire is significant. At the lower stock price, Berkshire would increase its share of earnings by $100 million, which five years down the road might mean an increase in value of $1.5 billion.
Buffett confessed that he came late to the IBM party. Like Coca-Cola in 1988 and Burlington Northern Santa-Fe in 2006, he had been reading the annual reports for 50 years before his epiphany. It arrived, he said, one Saturday in March 2011. Quoting Thoreau, Buffett says, “It’s not what you look at that matters; it’s what you see.” Buffett admitted to CNBC that he had been “hit between the eyes” by the competitive advantages IBM possesses in finding and keeping clients.51
The information technology (IT) services industry is a dynamic and global industry within the technology sector, and no one is bigger in this industry than IBM.52 Information technology is an $800 billion plus market that covers a broad spectrum of services broken down into four different buckets: consulting, systems integration, IT outsourcing, and business process outsourcing. The first two, combined, contribute 52 percent of IBM’s revenues; 32 percent comes from IT outsourcing; and 16 percent from business process outsourcing. In the consulting and systems integration space, IBM is the number-one global provider—38 percent bigger than the next competitor, Accenture. In the IT outsourcing space, IBM is also the number-one global provider—78 percent larger than the next competitor, Hewlett-Packard. In business process outsourcing, IBM is the seventh-largest provider, behind Teleperformance, Atento, Convergys, Sitel, Aegis, and Genpact.
Information technology services are considered to be a growth-defensive industry within the technology sector. Whereas technology sectors like hardware and semiconductors are more cyclical in nature, the services sector benefits from relatively stable growth prospects. The IT industry is more resilient because its revenues are both recurring and tied to the nondiscretionary budgets of larger corporations and governments. So important are IT services that consulting, systems integration, and IT outsourcing are thought to have “moatlike” qualities. According to Grady Burkett, associate director of technology at Morningstar, the intangible assets like reputation, track record, and client relationships are the sources of a moat in consulting and systems integration. In IT outsourcing, switching costs and scale advantages create their own moat, ensuring that once IBM lands a customer, that customer is likely to remain a loyal client for many years to come. Only one area, the relatively small business process outsourcing, is not protected by either the intangible assets or the cost of switching.
According to Gartner, the world’s leading information technology research and advisory company, the total market for IT services is expected to grow at a compounded annual rate of 4.6 percent, from an estimated $844 billion in 2011 to $1.05 trillion in 2016.
The move away from hardware technology to consulting and software begun by Lou Gerstner and accelerated by Sam Palmisano transformed IBM away from the low-margin, commoditized part of the technology industry to the higher-margin, moat-protected business of consulting, systems integration, and IT outsourcing. When Gerstner righted the ship in 1994, the return on equity at IBM was 14 percent. When he retired in 2002, return on equity had increased to 35 percent. Palmisano kept the ball rolling by moving return on equity even higher; it had reached 62 percent by the time he retired in 2012.
Part of the increase in return on equity can be attributed to the dramatic reduction in equity shares outstanding. But the more significant reason was the decision to leave behind low-margin businesses while dramatically growing the higher-margin businesses of consulting and outsourcing. In 2002, net profit margins at IBM were 8.5 percent. Ten years later, net margins had almost doubled to 15.6 percent.
Over 10 years (2002–2011), IBM generated a net profit for shareholders of $108 billion. It paid $20 billion to shareholders in the form of dividends, leaving it with $88 billion to run its business, which included capital reinvestments, acquisitions, and share repurchases. During this same period, the market capitalization of IBM gained $80 billion. This is not quite the one dollar of market value for each dollar retained that Buffett would prefer to see in his companies, but considering that the past 10 years had been a dreadful period for the performance of large-capitalization stocks, it is still respectable.
In 2010, IBM generated a net profit for shareholders of $14.8 billion. That year it spent $4.2 billion in capital expenditures, more than offset by $4.8 billion in depreciation and amortization charges. The net result was $15.4 billion in owner earnings. What is a business that in a year generates $15.4 billion in cash worth? According to John Burr Williams (and Warren Buffett), the value is the future cash flows of the business discounted back to the present value. The future cash flows will be determined by the growth of the company, and the discount rate that Buffett uses is the long-term U.S. government bond yield, his definition of a risk-free rate. Remember, Buffett does not use an equity risk premium in his calculations. Instead, he adjusts for riskiness by the margin of safety represented by the price he is willing to pay.
Using this theory, we can make our own calculation of IBM’s value. Using a two-stage dividend discount model, I assumed IBM would grow its cash earnings at 7 percent for 10 years and 5 percent after that. I then discounted these cash flows by 10 percent—substantially higher than the 2 percent of the 10-year U.S. Treasury note. The higher discount rate simply builds in a greater margin of safety. Based on these calculations, IBM is worth $326 per share, much higher than the average $169 price Buffett paid. If we adjust the growth rate for the next 10 years down to 5 percent, closer to Gartner’s estimate of the IT services industry’s growth, the value is $279 per share, still $100 per share higher than what Buffett paid.
Another way to look at the valuation question is to ask what growth rates are embedded in a $169 share price. To be worth $169 per share, IBM would have to grow its owner earnings at 2 percent in perpetuity. Readers might quibble at a $326 or $279 estimate of fair value for a share of IBM. But just as many, I suspect, would argue that IBM will grow faster than 2 percent per year over the next decade. The fair value answer, I am sure, lies somewhere in between these two estimates, reminding us again of one of Buffett’s favorite maxims: “I would rather be approximately right than precisely wrong.”
In many ways, the IBM purchase reminds me of Buffett’s purchase of the Coca-Cola Company. At the time, many critics were puzzled. Buffett had bought the stock near its all-time high (just like IBM). Many believed Coca-Cola was a boring, slow-growing company whose best days were behind it (just like IBM). When Buffett purchased Coca-Cola, the company’s share price was 15 times earnings and 12 times cash flow, a 30 percent and 50 percent premium to the market average. When we ran the dividend discount model on Coca-Cola’s owner earnings using various growth rates, we found the company was selling for a significant discount to fair value despite the premium price-to-earnings and price-to-cash-flow ratios. Assuming an unbelievably low 5 percent growth rate for Coca-Cola, the dividend discount model said Coca-Cola was worth $20.7 billion—far higher than its current market value of $15.1 billion.
Over the next 10 years, Coca-Cola’s share price went up tenfold versus the Standard & Poor’s 500 index, which increased threefold. A caution flag: I am certainly not saying that IBM can go up tenfold over the next 10 years, only that Wall Street’s method of using accounting ratios to determine value may capture the here and now but does a woefully poor job of calculating sustainable long-term growth. Or, put differently, more often than not, sustainable long-term growth is mispriced by the market.
Make no mistake—the greatest impact on IBM’s future success will be the company’s future earnings. Toni Sacconaghi, technology analyst at Sanford C. Bernstein, calls the company “fortress IBM, a company whose profit performance seems all but impervious to industry cycles.” Toni has gone so far as to call IBM, the largest global supplier of information technology to corporations and governments, “boringly predictable.”53 Boringly predictable was what they said about Coca-Cola in 1989. Boringly predictable is just the kind of company Buffett likes best.
We have also learned that financial management is a very important secondary factor in determining a company’s success. The legacy left by Gerstner and Palmisano will no doubt have a big influence on Ginni Rometty, IBM’s new CEO. Already, IBM’s Financial Model and Business Perspective (a five-year plan) includes $50 billion for future share repurchases. At IBM’s current share repurchase rate, the company may be down to less than 100 million shares outstanding by 2030. Of course, no one can say for sure that IBM will maintain its current share repurchase rate. But that doesn’t stop Buffett from dreaming. He told shareholders. “If repurchases ever reduce the IBM shares outstanding to 63.9 million, I will abandon my famed frugality and give Berkshire employees a paid holiday.”54
On February 14, 2013, Berkshire Hathaway and 3G Capital purchased H.J. Heinz Company for $23 billion. At $72.50 per share, the deal was a 20 percent premium to Heinz’s share price the day before.
It was easy to see how Heinz fit the Berkshire Hathaway mold. One of the best-known food companies in the world, it has global recognition that approaches that of Coca-Cola and IBM. Heinz’s deep red ketchup bottles can be found in millions of homes alongside Ore-Ida french fries and Lea & Perrins Worcestershire sauce. In 2012, the company reported $11.6 billion in revenue with a majority of sales coming from Europe and the rapidly expanding emerging markets. “It’s our kind of company,” said Buffett.55
Eighteen years before the pharmacist John Pemberton invented the formula for Coca-Cola, Henry J. Heinz was packing foodstuffs in Sharpsburg, Pennsylvania. The company first began selling horseradish sauce in 1869 but turned to tomato ketchup in 1876. Henry Heinz bought out his two partners in 1888 and renamed the company H.J. Heinz Company. The company’s famous slogan “57 varieties” was introduced in 1896. It is said Henry Heinz was riding an elevated train in New York City one day when he spotted a shoe store that claimed it had “21 styles.” Heinz picked the number “57” at random but selected the number “7” specifically because of its positive psychological influence. Buffett noted that 1869, when Heinz was started, was the same year his great-grandfather Sidney founded a grocery store.
Heinz is number one in ketchup globally and second in sauces. The future of Heinz depends on not only maintaining its market share leadership but also positioning itself in the rapidly growing emerging markets. Here Heinz is well positioned. The company bought Foodstar of China in 2010 and an 80 percent stake in Brazilian Coniexpress SA Industrias Alimeticias in 2011. Today, emerging markets account for seven of the company’s top 10 markets. Already, the value of Foodstar in China has doubled.
How important are emerging markets to Heinz? Over the past five years, these rapidly growing markets have accounted for over 80 percent of the company’s sales growth. In fiscal 2012, 21 percent of revenue came from emerging markets; estimates for fiscal year 2013 are for close to 25 percent. According to CEO William Johnson, the company’s organic growth rate in emerging markets is among the best in its peer group.
In 2012, Heinz reported $923 million in net income, $342 million in depreciation and amortization charges, and $418 million in capital expenditures, leaving shareholders with $847 million in owner earnings. But in the company’s annual report we note that $163 million was an after-tax charge for severance, asset write-downs, and other implementation costs. Adding back these nonoperational expenses, we can estimate that the company generated about $1 billion in owner earnings.
Using a two-stage dividend discount model, we estimated the company would grow its $1 billion in owner earnings at a 7 percent rate for 10 years, then 5 percent thereafter. Discounting the cash flows at 9 percent (which is the cost of the preferred stock Berkshire contributed to the deal) produces an estimated per-share value of $96.40. At a slower 5 percent constant growth rate in perpetuity, the company’s shares would be worth $82.10. Considering that the company has grown its earnings per share at an 8.4 percent compounded annual growth rate over the past five years, and knowing that the next five years will see a larger portion of its earnings coming from the rapidly growing emerging markets, I would argue these valuation estimates are very conservative.
If the company can grow at 7 percent for 10 years then 5 percent thereafter, Buffett bought the shares at a 25 percent discount to intrinsic value. At the very conservative estimated growth rate of 5 percent, he bought the company at a 12 percent discount. Admittedly, these are not the margin-of-safety discounts we usually see in Buffett’s purchases. But the attractiveness of the Heinz purchase goes beyond the usual discount formula.
Berkshire Hathaway and the Brazilian private equity firm 3G Capital will each own half of Heinz for $4 billion in equity. To that, Berkshire invested $8 billion in redeemable preferred stock yielding 9 percent. The preferred shares also have two other attractive features for Berkshire. First, at some point in the future the preferred will be redeemed at a significant premium; second, the preferred shares come with warrants that will allow Berkshire to buy 5 percent of the company’s common stock for a nominal amount. All in all, Berkshire Hathaway will earn a 6 percent annual return on its blended investment, excluding the value of the warrants, the premium conversion of the preferred shares, and any future growth in the intrinsic value of the company. Even if Heinz loses money, Berkshire Hathaway will be paid its preferred dividend. And even in the remote possibility that Heinz would go bankrupt, Berkshire would be in a position to wipe out other creditors and position itself favorably in a newly and cheaply reorganized Heinz.
It is easy to see how Heinz fits the profile of the type of company Buffett likes to purchase. The business is simple and understandable; it has had a consistent operating history; and, because it has positioned itself to benefit from the growth of emerging markets, the company has favorable long-term prospects. The return on investable capital (including debt) is 17 percent, with a return on shareholders’ equity of 35 percent.
But there are two caveats to the Heinz deal that make this purchase unique. First, the company will carry $6 of debt for every $1 of equity. Interest on the debt along with the 9 percent preferred dividend paid to Berkshire Hathaway will command much of the company’s owner earnings. In a word, the company is now highly leveraged. Second, the company will be run by a new management team at 3G Capital led by Brazil’s richest man, Jorge Paulo Lemann.
In the past, when Buffett bought a company, he preferred to have the existing management team continue to run the business. But in this case, a new management team will be responsible for the future of Heinz.
Buffett first met Lemann when both were on the board of Gillette in the 1990s. Although Lemann and 3G Capital are not well known in the United States, this team has had tremendous success with fast-food restaurants, banks, and brewers. In 2004, in what is considered his watershed deal, Lemann merged his smaller Brazilian beer company AmBev with Belgium’s much larger Interbrew, the maker of Stella Artois and Beck’s. Even though AmBev was smaller, it was Lemann’s associates who took the senior jobs at the combined company. Then in 2008, this newly merged company paid $52 billion for Anheuser-Busch in what is now the largest beer merger ever.
In 2010, 3G Capital bought Burger King for $3.3 billion. Within weeks, the company had fired half the 600 employees at its Miami headquarters, sold off the executive wing, and henceforth required employees to get permission to make color printouts. Since the acquisition, 3G Capital has cut the operating costs by 30 percent. In the meantime, each restaurant introduced new products, including smoothies and snack wraps, and orchestrated a remodeling campaign paid for by the franchise owners. In its fourth quarter 2012 report, Burger King noted a doubled profit and improved cash flows.
In studying the management profile of 3G Capital and the success it has had with each of its deals, I am reminded of another manager whom Buffett considered one of the best in the world, Tom Murphy. The same passion that drove Murphy to slash unnecessary costs and improve the productivity of Cap Cities/ABC businesses is evident in the management team at 3G Capital.
Some have suggested that 3G Capital, being a private equity firm, will look to sell its Heinz investment sooner rather than later. But Lemann insists that 3G Capital intends to hold Heinz for the long term in much the same way it has become a long-term owner of Anheuser-Busch InBev N.V. Just as AmBev served as a platform for the consolidation and future growth of the beer business, so too could Heinz serve as an early platform for the consolidation and future growth of the food industry.
Whether 3G Capital stays or not, Buffett is happy to be a long-term owner. “Heinz will be 3G’s baby,” says Buffett. But of course he adds that “we may increase our ownership if any members of the 3G Group want to sell out later.”56
You may have noticed a standard refrain in these case studies: Warren Buffett is in no hurry to sell, even when the stocks he purchases are doing well. Short-term appreciation doesn’t interest him. Philip Fisher taught him that either the investment you hold is a better investment than cash or it is not. Buffett says that he is “quite content to hold any security indefinitely, so long as the prospective return on equity capital of the underlying business is satisfactory, management is competent and honest, and the market does not overvalue the business.”57 (Recognize the tenets in that statement?) As he reminds shareholders, his favorite holding period is “forever.”
And with that memorable statement, Buffett takes us to the other side of the coin: Having made rational decisions about which stocks to buy, how should we think about managing the portfolio? That is the subject of the next chapter.
Notes
1. Mary Rowland, “Mastermind of a Media Empire,” Working Women, November 11, 1989, 115.
2. The Washington Post Company Annual Report, 1991, 2.
3. Berkshire Hathaway Annual Report, 1992, 5.
4. Berkshire Hathaway Annual Report, 1985, 19.
5. Chalmers M. Roberts, The Washington Post: The First 100 Years (Boston: Houghton Mifflin, 1977), 449.
6. Berkshire Hathaway Annual Report, 1991, 8.
7. Ibid., 9.
8. William Thorndike Jr., The Outsiders: Eight Unconventional CEOs and Their Radically Rational Blueprint for Success (Boston: Harvard Business Review Press, 2012), 9.110.
9. Carol Loomis, “An Accident Report on GEICO,” Fortune, June 1976, 120.
10. Although the 1973–1974 bear market might have contributed to part of GEICO’s earlier fall, its decline in 1975 and 1976 was all of its own making. In 1975, the Standard & Poor’s 500 Index began at 70.23 and ended the year at 90.9. The next year, the stock market was equally strong. In 1976, the stock market rose and interest rates fell. GEICO’s share price decline had nothing to do with the financial markets.
11. Beth Brophy, “After the Fall and Rise,” Forbes, February 2, 1981, 86.
12. Lynn Dodds, “Handling the Naysayers,” Financial World, August 17, 1985, 42.
13. Berkshire Hathaway Letters to Shareholders, 1977–1983, 33.
14. Andrew Kilpatrick, Warren Buffett: The Good Guy of Wall Street (New York: Donald Fine, 1992), 102.
15. Anthony Bianco, “Why Warren Buffett Is Breaking His Own Rules,” BusinessWeek, April 15, 1985, 34.
16. Berkshire Hathaway Annual Report, 1991, 8.
17. Bianco, “Why Warren Buffett Is Breaking His Own Rules.”
18. Dennis Kneale, “Murphy & Burke,” Wall Street Journal, February 2, 1990, 1.
19. Capital Cities/ABC Inc. Annual Report, 1992.
20. “A Star Is Born,” BusinessWeek, April 1, 1985, 77.
21. Anthony Baldo, “CEO of the Year Daniel B. Burke,” Financial World, April 2, 1991, 38.
22. Berkshire Hathaway Annual Report, 1985, 20.
23. Roger Lowenstein, Buffett: The Making of an American Capitalist (New York: Random House, 1995), 323.
24. Berkshire Hathaway Annual Report, 1993, 14.
25. Kilpatrick, Warren Buffett: The Good Guy of Wall Street, 123.
26. Mark Pendergrast, For God, Country and Coca-Cola (New York: Scribners, 1993).
27. Art Harris, “The Man Who Changed the Real Thing,” Washington Post, July 22, 1985, B1.
28. “Strategy of the 1980s,” Coca-Cola Company.
29. Ibid.
30. Berkshire Hathaway Annual Report, 1992, 13.
31. Ibid.
32. John Dorfman, “Wells Fargo Has Bulls and Bears; So Who’s Right?,” Wall Street Journal, November 1, 1990, C1.
33. Ibid.
34. John Liscio, “Trading Points,” Barron’s, October 29, 1990, 51.
35. Berkshire Hathaway Letters to Shareholders, 1977–1983, 15.
36. Berkshire Hathaway Annual Report, 1990, 16.
37. Reid Nagle, “Interpreting the Banking Numbers,” in The Financial Services Industry—Banks, Thrifts, Insurance Companies, and Securities Firms, ed. Alfred C. Morley, 25–41 (Charlottesville, VA: Association of Investment Management and Research, 1991).
38. “CEO Silver Award,” Financial World, April 5, 1988, 92.
39. Gary Hector, “Warren Buffett’s Favorite Banker,” Forbes, October 18, 1993, 46.
40. Berkshire Hathaway Annual Report, 1990, 16.
41. Ibid.
42. Ibid.
43. R. Hutchings Vernon, “Mother of All Annual Meetings,” Barron’s, May 6, 1991.
44. John Taylor, “A Leveraged Bet,” Forbes, April 15, 1991, 42.
45. Berkshire Hathaway Annual Report, 1994, 17.
46. Dominic Rushe, “Warren Buffett Buys $10bn IBM Stake,” The Guardian, November 14, 2011.
47. Berkshire Hathaway Annual Report, 2011, 7.
48. Ibid.
49. Ibid., 6.
50. Ibid., 7.
51. Rushe, “Warren Buffett Buys $10bn IBM Stake.”
52. I am grateful to Grady Buckett, CFA, director of technology at Morningstar, for his tutorial.
53. Steve Lohr, “IBM Delivers Solid Quarterly Profits,” New York Times, July 18, 2012.
54. Berkshire Hathaway Annual Report, 2011, 7.
55. Quote from Warren Buffett on CNBC, February 14, 2013.
56. Michael de La Merced and Andrew Ross Sorkin, “Berkshire and 3G Capital in a $23 Billion Deal for Heinz,” New York Times, February 19, 2013.
57. Berkshire Hathaway Annual Report, 1987, 15.