The bet — final 10-year returns (2008–2017)
S&P 500 index fund vs. Protégé's five funds-of-funds. All figures from the 2017 Annual Report.
Buffett 2017 — "Performance comes, performance goes. Fees never falter."
Warren Buffett's 2017 letter closes the ten-year bet (S&P +125.8% vs. best fund-of-funds +87.7%), redefines investment risk for long-horizon investors, presents the four-decline leverage warning, and quietly names Ajit Jain and Greg Abel as Vice Chairmen — with float crossing $100B for the first time.
Published: February 24, 2018 — Warren Buffett, Berkshire Hathaway 2017 Annual Report
The headline number in Berkshire's 2017 letter is $65.3 billion — the gain in net worth over the year, which lifted per-share book value by 23%. Buffett was quick to deflate it. Only $36 billion came from Berkshire's operations. The remaining $29 billion arrived in December 2017, delivered not by any decision Buffett made but by Congress, when the Tax Cuts and Jobs Act rewrote the deferred tax liabilities on Berkshire's balance sheet. "A large portion of our gain did not come from anything we accomplished at Berkshire," he wrote. 1
That kind of candor — volunteering the asterisk on your own headline — is one of the things that has kept Buffett's annual letters worth reading for five decades. The 2017 letter is no exception, and it arrives with several passages that will still be quoted long after the year's particulars are forgotten.
Three of them stand out. First: the ten-year bet concluded, and Buffett delivered his final verdict on fees in four words that deserve to be printed on every fund prospectus. Second: he offered his most systematic argument that long-term investors routinely misclassify risk — treating bonds as safe and stocks as dangerous when the opposite is true over investment horizons of a decade or more. Third: he named, formally and without ceremony, the two executives who will run Berkshire after he and Charlie Munger are gone. 1
The bet closes: ten years, one verdict
In December 2007, Buffett proposed a public wager: a low-cost Vanguard S&P 500 index fund would outperform any selection of at least five hedge funds over the following decade. Ted Seides of Protégé Partners accepted. Each side put up $318,250 in zero-coupon U.S. Treasury bonds, with the $1 million face value going to charity. The stake would later be converted into 11,200 Berkshire B shares — a decision that ultimately sent $2,222,279 to Girls Inc. of Omaha instead of the original $1 million. 1
Protégé chose five funds-of-funds, which together held interests in more than 200 underlying hedge funds. The idea was that this diversification would smooth out individual manager volatility while preserving access to the best talent on Wall Street. The managers involved, Buffett was careful to note, were "honest and intelligent." The structure was the problem.
The ten-year results:
正在加载统计卡片…
The S&P 500 returned 125.8% cumulatively — 8.5% annualized. The best of the five funds-of-funds, Fund C, reached 87.7% over ten years. The worst, Fund D, returned 2.8% total before liquidating. The four others fell between 21.7% and 42.3%. 1
All five funds-of-funds beat the index in 2008. In every one of the nine following years, the group as a whole trailed. The fees kept compounding regardless of which direction performance went. Buffett's summary, in its entirety: "Performance comes, performance goes. Fees never falter." 1
The arithmetic behind this result is not complex. Active managers in aggregate own the market — so in aggregate, before costs, they must earn the market return. After the underlying hedge fund layer (typically 2% of assets plus 20% of profits) and the fund-of-funds overlay (roughly 1% of assets plus 10% of profits), the aggregate active investor is left with roughly the market return minus several hundred basis points per year. The math does not require any manager to be dishonest or incompetent. It only requires that fees be real.
The bet produced a second lesson that Buffett found equally important: "Stick with big, 'easy' decisions and eschew activity." 1 Over ten years, the 200-plus hedge fund managers made tens of thousands of buy and sell decisions. Buffett and Protégé Partners made exactly one consequential choice — in November 2012, they sold the zero-coupon bonds (yielding 0.88%) and bought Berkshire B shares. The index fund's manager made no decisions at all. Activity was inversely correlated with outcome.
Risk: the definition that changes the answer
The 2017 letter contains Buffett's clearest written statement of how he defines investment risk — and why he thinks the conventional definition produces the wrong conclusions for long-term investors.
He starts from first principles:
"Investing is an activity in which consumption today is foregone in an attempt to allow greater consumption at a later date. 'Risk' is the possibility that this objective won't be attained." 1
By this definition, the risk relevant to a long-term investor is not short-term price volatility. It is the probability that the investment fails to preserve and grow purchasing power over the investor's actual time horizon. Measured this way, the conventional hierarchy inverts.
Buffett illustrated it with a thought experiment set in November 2012, when ten-year U.S. Treasury bonds yielded roughly 0.88%. An investor who locked into that yield for a decade was guaranteed to earn less than 1% annually in nominal terms and, if inflation ran at even 2%, was guaranteed to lose purchasing power. The S&P 500 at the same moment yielded about 2.5% in dividends — roughly three times the bond yield — and those dividends were, in Buffett's words, "almost certain to grow." 1
His conclusion:
"It is a terrible mistake for investors with long-term horizons — among them, pension funds, college endowments and savings-minded individuals — to measure their investment 'risk' by their portfolio's ratio of bonds to stocks. Often, high-grade bonds in an investment portfolio increase its risk." 1
The passage targets the conventional wisdom embedded in most financial planning — that a "conservative" portfolio means a high bond allocation, and that adding bonds reduces risk. Buffett's argument is that this is true only if you define risk as short-term price volatility. If you define risk as the probability of failing to maintain purchasing power over a twenty- or thirty-year horizon, a high bond allocation at low yields is the risky choice.
This does not mean stocks are always preferable. The argument is horizon-dependent: in the short run, stocks can lose half their value while bonds remain stable. Over decades, a diversified equity portfolio purchased at reasonable valuations has, historically, grown purchasing power substantially. The risk equation changes as the time horizon lengthens.
The practical implication Buffett drew was for institutions with long investment horizons — pension funds and endowments — that routinely hold large bond allocations as a matter of policy. For those investors, he argued, the risk management is backward.
Four declines and the leverage warning
One of the most useful tables in the 2017 letter is also one of the least commented-on: a history of Berkshire's four largest stock price declines.
| Period | High | Low | Decline |
|---|---|---|---|
| Mar 1973 – Jan 1975 | $93 | $38 | –59.1% |
| Oct 2 – Oct 27, 1987 | $4,250 | $2,675 | –37.1% |
| Jun 19, 1998 – Mar 10, 2000 | $80,900 | $41,300 | –48.9% |
| Sep 19, 2008 – Mar 5, 2009 | $147,000 | $72,400 | –50.7% |
All four episodes drawn from the 2017 Annual Report. 1
Buffett wrote: "This table offers the strongest argument I can muster against ever using borrowed money to own stocks." 1 Four times in fifty years, the stock of one of the most conservatively managed businesses in the United States fell by more than a third, with two episodes exceeding 50%. Each time, the business itself continued to compound. Each time, a leveraged holder would have faced either forced liquidation or margin calls at exactly the wrong moment.
The accompanying warning is characteristically unambiguous: "The light can at any time go from green to red without pausing at yellow." 1 There is no warning period. There is no time to exit before a decline becomes a crisis. The only protection against forced selling is having nothing to force.
Buffett then quoted six lines from Kipling's If — including the famous "keep your head when all about you are losing theirs" — and added: "It's not that Charles Darwin would have expected us to. It's that the rules of the game require it." The point was not sentimental. An investor without leverage who survives a 50% decline with no debt due and no capital withdrawn is positioned to buy — precisely when leveraged investors are compelled to sell.
This is the structural argument beneath the washtub metaphor from the 2016 letter. Leverage does not just threaten to destroy individual portfolios. It creates the temporary mispricings that patient, unlevered capital can exploit. The two facts are connected.
Abel and Jain: succession named
Buffett is 87. Charlie Munger was 94 when the 2017 letter was published. The succession question — who runs Berkshire after them — has shadowed the shareholder letters for years without being answered explicitly. The 2017 letter answered it.
"Early in 2018, Berkshire's board elected Ajit Jain and Greg Abel as directors of Berkshire and also designated each as Vice Chairman. Ajit is now responsible for insurance operations, and Greg oversees the rest of our businesses. Charlie and I will focus on investments and capital allocation." 1
The tone was quiet to the point of understatement. After two years in which Greg Abel's name had not appeared in the annual letter at all, his appointment as Vice Chairman of all non-insurance operations was noted in a single paragraph. Ajit Jain — who has run Berkshire's insurance operations for decades and whom Buffett has called irreplaceable since at least the 2000s — received the same brief treatment.
Buffett's endorsement, by his standards, was lavish: "You and I are lucky to have Ajit and Greg working for us. Each has been with Berkshire for decades, and Berkshire's blood flows through their veins. The character of each man matches his talents. And that says it all." 1
The division of responsibility matters. Jain manages the insurance complex — the engine that generates the float. As of year-end 2017, Berkshire's float had reached $114.5 billion, crossing $100 billion for the first time. 1 The underwriting streak of 14 consecutive profitable years ended in 2017 when hurricanes Harvey, Irma, and Maria produced a $3.2 billion pre-tax underwriting loss — but Berkshire absorbed roughly 3% of total industry hurricane losses while carrying only a fraction of the net worth erosion that other reinsurers suffered. 1 That resilience is the institution Jain was formally handed.
正在加载图表…
Berkshire insurance float, 1970–2017. From $39 million in 1970 to $114.5 billion at year-end 2017. 1
Abel's domain is the non-insurance operating businesses — BNSF, Berkshire Hathaway Energy, Clayton Homes, IMC, Lubrizol, Marmon, Precision Castparts, and several dozen others — which together generated $20 billion in pre-tax income in 2017. 1 The operational structure has been in place for years; the 2017 letter made the governance structure match it in name.
What this letter means for investors today
The 2017 letter was filed against a year of all-time highs in U.S. equity markets, an acquisition landscape Buffett described as one where "price seemed almost irrelevant to an army of optimistic purchasers," a cash pile of $116 billion he called "far beyond the level Charlie and I wish Berkshire to have," and the formal conclusion of the most publicly visible investment experiment in the industry's recent history. 1
The passages worth returning to are the ones that don't depend on market conditions.
On fees. The bet's final tally is not surprising — Buffett had been signaling the outcome since the 2014 letter. What is worth marking is the mechanism. The fund-of-funds structure did not underperform because the managers were careless. It underperformed because the costs were certain and the outperformance was not. "Performance comes, performance goes. Fees never falter" is the most compressed version of an argument Buffett has been making for thirty years. The question for any investor holding an actively managed fund is not whether the manager is skilled — it is whether the skill premium is large enough, and consistent enough, to overcome the annual cost that compounds regardless.
On risk. The bond-vs.-stock risk argument becomes most relevant when interest rates are low and investment horizons are long. Buffett's definition — risk as the probability of failing to grow purchasing power — is the one that matters for a pension fund with forty-year liabilities or an individual saving for a retirement two decades away. A portfolio designed to minimize volatility can maximize the other kind of risk.
On leverage. The four-decline table is a useful pressure test. If any of those four episodes — Berkshire falling 59% between 1973 and 1975, or 51% in six months in 2008 — would have forced a sale, the position was too large or too leveraged. The question is not whether a drawdown can be endured emotionally. It is whether it can be survived structurally.
On succession. The Abel/Jain announcement in the 2017 letter resolved the explicit uncertainty that had surrounded Berkshire for years. What it did not resolve is what comes after — how the culture survives the generation after them, how the capital allocation discipline propagates without Buffett's specific judgment, and whether the conglomerate structure continues to attract the kind of owner-operator businesses that built it. Those are open questions the letter did not address and probably cannot.
正在加载内容卡片…
The 2017 letter was published on February 24, 2018 — the year's longest, at roughly 30 pages, covering fifty-three years of compounding, a bet that ended a decade of public accounting for fees, two Vice Chairman designations made with one quiet paragraph, and a risk framework that inverts the standard financial planning playbook. The cumulative return since 1964: Berkshire's market value, +2,404,748%. The S&P 500 over the same period: +15,508%. Buffett noted the compounding calmly, then spent the rest of the letter explaining why the number is less important than the conditions that made it possible. 1
Cover image: AI-generated illustration
围绕这条内容继续补充观点或上下文。