Invisible Hand

Warren Buffett, Market Efficiency and the Disappearance of Alpha

Invisible Hand

As staunch admirers of Warren Buffett, we at IFA can only marvel at the value he has created for his investors over the course of his illustrious career. Anyone fortunate enough to own Berkshire Hathaway since June 1st, 1979 has seen a remarkable 750-fold increase in wealth as of December 31, 2016, all without having to pay taxes on distributions. Of course, these investors endured a lot of volatility, and we know for a fact that some of them bailed out in the late 1990s when Berkshire lagged the overall market during the dot-com craze.

When explaining to investors that seeking alpha (an additional return above a risk-appropriate benchmark attributable to a fund manager's security selection skills) is a waste of time and potentially quite hazardous to their wealth, IFA's advisors often hear the refrain, "What about Warren Buffett?" Our answer to this question is multi-faceted, and it begins by noting that Warren Buffett is not a mutual fund manager; he is the head of a conglomerate which also happens to own equity positions in several blue chip companies. Much of the value he has added to Berkshire has derived from his extraordinary ability at capital allocation among companies, identifying which of Berkshire's companies should be sources of cash for future acquisitions and which ones should have their earnings re-invested into the business. All this is aside from his skill at managing the CEOs of these companies, extracting the highest possible value from their abilities.

An additional area where Buffett has provided value can be summarized in two words, his Rolodex and his reputation. This assertion was exemplified by the preferred stock deal he negotiated with Goldman Sachs during the depths of the 2008 financial crisis. These 10%-yielding securities would not have been made available to the general public because more than Berkshire's capital, Goldman needed the cachet of having Buffett as an investor. It assured Goldman's other investors (as well as potential future investors) that Goldman was a sound enterprise and was not going the way of Lehman and Bear Stearns. Needless to say, Berkshire has received a far higher return on capital than did the U.S. taxpayers from the capital injected into Goldman under TARP.

Since Berkshire Hathaway is not a mutual fund, it is valued by the market on a continuous basis and the share price of Berkshire reflects the market's estimate of the value added by Buffett's Rolodex, his capital allocation abilities and his company-picking prowess. To say otherwise would imply that the thousands of buyers and sellers of Berkshire Hathaway stock are somehow ignorant and readily taken advantage of. This means that the degree to which Buffett can add value in the future is already incorporated into today's price of Berkshire Hathaway, and it is simply not possible for an investor to purchase Buffett's alpha. In terms of labor economics, Buffett's skill is the scarce resource and the numerous traders will bid up the price of having a stake in Buffett's endeavors to the point where they are indifferent between Berkshire and a similar collection of companies without the benefit of Buffett's management. 

As we like to say at IFA, the benefit of Warren Buffett is baked in the cake. We can see evidence of this assertion in the chart below which shows the annual alpha of Berkshire Hathaway starting from 1980. Splitting the overall time period into two approximately equal pieces illustrates this point. In the second half, the outperformance relative to the Russell 1000 Value Index is statistically insignificant. This can be seen by clicking on the buttons at the top of the chart. The first button reveals a statistically significant alpha for the entire period (t-stat of 2.79), but the second button indicates that almost the entire alpha occurred in the first half (t-stat of 3.08), and the third button (19 Year Period) shows that the alpha of the second half is not even close to statistically significant (t-stat of 0.73).

 

Here is a calculator to determine the t-stat. Don't trust an alpha or average return without one.

A working paper1 from researchers at New York University and AQR Capital Management, “Buffett’s Alpha”, attributes Berkshire’s returns to the use of leverage combined with a focus on cheap, safe, quality stocks. One of Berkshire’s primary sources of capital is the “float” from its insurance operations which refers to the difference between premiums collected and claims paid. Combining this with Berkshire’s ability to issue AAA-rated debt has resulted in a very low cost of capital which Buffett has used to great advantage.   When leverage is combined with a diversified portfolio of stocks with similar characteristics to the companies chosen by Buffett, most of the returns of Berkshire can be replicated. Nevertheless, the authors are quick to point out that explaining Buffett’s performance with the benefit of hindsight does not diminish his outstanding accomplishment of deciding to consistently invest on these principles for the last half century.

This leaves us with one final question: Suppose that the Oracle of Omaha were to decide to switch careers and start his own mutual fund; since mutual funds are valued daily based only on the market value of the current holdings, would this not be a fantastic opportunity to purchase the alpha that was unavailable in Berkshire Hathaway? Not so fast. Two things would happen in this scenario: First, as long as investors were convinced that excess return was there for the taking, they would direct their money into Buffett's mutual fund until the potential for excess return vanished. Second, Buffett would be a fool not to raise his management fee to the point where investors become indifferent between his fund and any other fund. If he wanted to be altruistic, he could certainly find a beneficiary of his choosing rather than offering his services to the general public at an artificially low price. An example of an investment manager who could have made a much larger salary on Wall Street is David Swensen, the Chief Investment Officer of Yale University.

To summarize, the fact that long-term alpha has been identified in only a few individuals such as Buffett and Swensen indicates how rare it is. However, even in those cases where it has been identified, it is not available for purchase by the investing public. To think otherwise is the pinnacle of naiveté or perhaps arrogance in believing that someone who has alpha-generating ability somehow owes it to you. IFA echoes the advice of both Warren Buffett and David Swensen in their repeated endorsements of index funds.

 


1”Buffett’s Alpha”, by Andrea Frazzini, David Kabiller and Lasse Pedersen, August 2012 (http://www.econ.yale.edu/~af227/pdf/Buffett's%20Alpha%20-%20Frazzini,%20Kabiller%20and%20Pedersen.pdf).