Is Buffett Really a Good Investor?

Most people will be familiar with Warren Buffett, possibly the greatest investor of all time. But what few people know is that there is another side to this story: a group of people who argue that, as unbelievable as it may sound, his success was due to luck. We are not talking about idiots with no prestige, but Nobel Prize winners, top-level economists, and respected people.

This argument is not as stupid as it sounds.

They are not saying that Buffett is dumb and simply wins by chance without knowing what he is doing; it is something more interesting, and worth paying attention to.

The Statistical Anomaly

In the 1960s and 1970s, the idea that the market incorporates available information became consolidated. Fama formulated the efficient market hypothesis, and the idea that beating the market is statistically improbable started to become popular. Through that lens, Buffett was an anomaly the model could not explain: if nobody can beat the market except by chance, what do you do with a guy who beats it for decades?

Warren Buffett’s returns have traditionally been ignored or discredited in academia, even by some of the most influential and reputable finance professors, because his track record is classified as a statistical outlier.

The discussion begins with Michael Jensen, who dismissed Warren Buffett’s returns in a debate held in 1984 at Columbia University:¹

If I look at a group of analysts with no talent, all of whom do nothing but flip coins, I expect to find some who have flipped two heads in a row, and even some who have flipped ten heads in a row.

Years later, Merton Miller summarized that same intuition even more directly:²

If there are 10,000 people looking at stocks and trying to pick winners, one in 10,000 is going to get, just by chance, a great success, and that is all that is happening. It is a game, a chance operation, and people think they are doing something deliberate, but in reality they are not.

Another Nobel Prize winner, William Sharpe, described Buffett as a “3-sigma event” and said he was “a statistical aberration so far outside the norm that it requires no further attention.”

Burton Malkiel also repeated something similar:³

In any activity in which a large number of people participate, although the average will probably prevail, the unexpected is bound to happen. The tiny number of truly good managers we find in the investment management industry is not at all inconsistent with the laws of chance.

With these arguments, and others like them, Warren Buffett’s extraordinary returns are attributed entirely to chance. And again, although this may seem insane, it is not that far-fetched.

The argument becomes more convincing when you hear people like Mauboussin explain a counterintuitive idea: in competitive markets, it is very difficult to distinguish real skill from a statistical tail.

Paradoxically, the more skilled and evenly matched the players in a game are, the more the result depends on luck. One would expect that, at the highest competitive levels, skill would decide the outcome, but since all the players are extremely capable and their skill barely differs, luck often ends up dictating the result.

Knowing this, is it so crazy to think that in the most competitive market in the world, the stock market, Buffett’s returns could be due to luck? Is he really that far ahead of everyone else?

Orangutans Flipping Coins

The idea is simple: in an experiment with millions of investors, someone has to come out on top. Just as if millions of people flipped coins, some would string together 20 heads in a row, something extremely unlikely. That does not prove skill; it proves the sample space was huge and someone had to win.

Luckily for all of us, Buffett’s response to these claims has gone down in history:⁴

To begin with, if (a) you had taken 225 million orangutans distributed roughly like the population of the United States; if (b) after 20 days there were 215 winners left; and if (c) you found that 40 came from a particular zoo in Omaha, you would be pretty sure you had found something.

You would probably go ask the zookeeper what he feeds them, whether they do special exercises, what books they read, and who knows what else. In other words, if you find a truly extraordinary concentration of success, you may want to see whether you can identify a concentration of unusual characteristics that could be causal factors.

The analogy is brilliant. In a world of millions of coin flippers, of course someone will have an absurd streak. But if a disproportionate number of the winners come from the same place, with the same principles and the same intellectual school, perhaps the right question stops being “was he lucky?” and becomes “what did they have in common?”

That intellectual school is, of course, the Graham & Dodd school.

Charlie Munger told a similar story in his Herb Kay Memorial Lecture at the University of California, Santa Barbara:

For a long time, there was a Nobel Prize-winning economist who explained the success of Berkshire Hathaway in the following way:

First, he said Berkshire had beaten the market by investing in common stocks thanks to one sigma of luck, because nobody could beat the market except by luck. This hard version of the efficient market theory was then being taught in most economics departments. People were taught that nobody could beat the market.

Then the professor moved to two sigmas, and three sigmas, and four sigmas, and when he finally got to six sigmas of luck, people laughed so much that he stopped doing it.

Then he reversed the explanation 180 degrees. He said: “No, it was still six sigmas, but it was six sigmas of skill.”

Now, I do believe that Warren Buffett has been very lucky. But not from the coin-flipping perspective.

Buffett was born in the right place, at the right time, with the right market. In post-war America, with less efficient markets, less quantitative competition, early access to small caps and special situations, and then the ability to reinvest for decades. That was his luck, which of course does not negate his skill, but it does explain why repeating “another Buffett” today is much harder.

The Quantitative Explanation

The classic paper Buffett’s Alpha⁵ provides a very interesting explanation of Buffett’s returns without resorting to “luck” or to the esoteric idea of “Buffett’s innate ability.” In the version published in the Financial Analysts Journal, Frazzini, Kabiller, and Pedersen estimate that Berkshire achieved a Sharpe ratio of 0.79 and that Buffett used average leverage of approximately 1.7x. In addition, after controlling for factors such as Betting-Against-Beta and Quality-Minus-Junk, the alpha is no longer statistically significant.

In other words, his success can be explained by factor exposure, leverage, and a very special corporate structure.

I particularly like this paper because it acts as a bridge between the academic world and the world of value investors, something that rarely happens. While it is not perfect, it helps us understand that Buffett’s genius is compatible with academic theory, which is also not perfect, and with the school of value investing.

What Factors Explain Buffett’s Success?

1. He bought cheap, safe, high-quality stocks.

   It was not just “value” in the academic sense. Berkshire had a lot of exposure to value, low beta, and quality factors: profitable, stable companies, with low risk, reasonable growth, and healthy payout policies. In other words, something quite similar to what value investors usually mean by “value,” which is consistent with his investment philosophy. In addition, when the authors control for BAB and QMJ, much of the “alpha” disappears.

   In other words, Buffett’s returns are explained by those variables without needing to resort to more intangible and esoteric explanations.

2. He used cheap and stable leverage.

   The paper estimates that Berkshire had a Sharpe ratio of 0.79: very good, yes, but not spectacular. What turns that into a gigantic fortune is applying leverage to it for decades. The paper estimates average leverage of around 1.7 times. The interesting part is that Berkshire did not borrow like a normal hedge fund. Berkshire had insurance float: insurance premiums collected in advance and paid later in the form of claims. The paper estimates that around 35-36% of its liabilities came from float, with an average cost below the T-Bill rate.

   This significantly increases returns, and low-cost financing is a brutal advantage.

3. He had the reputation.

   Another advantage, this time qualitative, is that Berkshire, thanks to the reputation built after decades of strong performance, achieved something very valuable: it did not have to sell at the worst moment.

   Many people can have a good strategy but fail to survive the drawdowns. Berkshire lost 44% between June 1998 and February 2000 while the market rose 32%. Many managers would have lost clients, funding, or directly their jobs. Buffett, thanks to his prestige, was able to endure it.

In reality, Buffett’s great skill seems to have been detecting earlier and better than other investors what many people now know: that buying “quality, value, and low beta” stocks with ridiculously cheap financing, and holding them for decades, was a great investment strategy.

At this point, value investors and academics can finally hug each other and agree on something. Buffett’s success is demonstrable according to both “schools,” and both find common ground where they can have a conversation.

Berkshire Dominated for 40 Years

Perhaps, after reading Charlie’s quote earlier, some of you may have thought he was exaggerating. It is hard to understand how incredible and consistent Buffett’s returns have been, so it is no surprise that academics had a hard time accepting that they were due to something other than luck.

Luckily, the charts in the paper’s appendix make this quite clear. Buffett is not simply “a little above average,” but dominates all his peers.

In the universe of U.S. equity funds with at least 40 years of history, Berkshire appears completely isolated in the right tail. Nobody comes close.

The same thing happens when the comparison is made against U.S. stocks with more than 40 years of trading history. How could you not think that something strange is going on?

Among U.S. stocks with at least 40 years of history, Berkshire had the highest Sharpe ratio; and among comparable mutual funds, it did too.

So, while the paper provides a theoretical framework for understanding Berkshire’s spectacular performance, this does not dilute how incredible the event was in any way. In fact, I think it helps us understand it even better.

Qualitative Factors

Many of you will be thinking: where does the qualitative part fit in?

Well, you are not the only ones. As I said, the previous paper is not perfect, nor does it pretend to be, so a later paper titled “Buffett’s Alpha: Further Explanations from a Behavioral Value Investing Perspective” adds qualitative layers to Buffett’s performance. The main argument is that the CAPM and its derivatives have little practical application and, of course, their equations fail to capture the qualitative nuances that truly explain Berkshire’s returns.

The factors added by the paper are no less important, but in my opinion they are less interesting. They are things any value investor has heard hundreds of times. Still, I will mention some of them briefly, not all of them, because they should be kept in mind, since investing in practice is very different from investing in theory.

1. Temperament: Buffett not only knew what to buy; he had the psychological ability to hold a strategy for decades, even when it looked stupid or out of fashion. The quantitative strategy explains the “what,” but temperament explains “how you survive long enough to capture it.”

2. Reputation: Here it appears in a different form from the one already mentioned. Berkshire became the “preferred buyer” for owners of good family businesses. Many sellers preferred to sell to Buffett because they trusted that he would not break up the company, destroy the culture, or mistreat the managers. That gave him access to deals other financial buyers did not have.

3. Circle of competence: Buffett and Munger avoid many mistakes simply by not playing games where they do not have an edge. The paper connects this with behavioral biases, especially overconfidence in one’s own abilities. Their advantage was not knowing everything, but knowing where they did not know enough.

4. Berkshire’s organizational culture: Berkshire combines two rare things: capital allocation centralized in Buffett/Munger and highly decentralized operations in the managers of each business. That reduces bureaucracy, attracts independent managers, and allows capital decisions to be made rationally without the typical pressure of running the day-to-day business.

This paper concludes that factor models explain an important part of Buffett’s result, but they do not capture the human and organizational mechanisms that made that result possible.

In other words, the first paper explains the exposures: quality, value, low beta, leverage. This second paper tries to explain the deeper source: discipline, culture, reputation, patience, rationality, and capital allocation.

Conclusion

I felt like writing this article to tell the funny story of the orangutans flipping coins and the arrogance of academia toward results that escape its models. But I also wanted to show that academic research is not incompatible with different schools of investing, such as value investing, as long as one understands what it is about and remains coherent.

Both sides, academia and value investing, would benefit from listening more carefully to what the other has to say; so would individual investors who lean toward either of them. Let this small blog serve as a point of reconciliation between both doctrines, using one of the greatest investors of all time as an excuse.

Footnotes

1. Michael Jensen, cited in Roger Lowenstein, Buffett: The Making of an American Capitalist, Random House, 1995, p. 317. ↩

2. Merton Miller, cited in Trillion Dollar Bet, NOVA, PBS, February 8, 2000. ↩

3. Burton G. Malkiel, A Random Walk Down Wall Street, 1999 edition, p. 181. ↩

4. Warren Buffett, The Superinvestors of Graham-and-Doddsville, Columbia Business School. ↩

5. Andrea Frazzini, David Kabiller, and Lasse Heje Pedersen, Buffett’s Alpha, Financial Analysts Journal, 2018, 74(4), 35-55. DOI: 10.2469/faj.v74.n4.3. ↩