Any time I suggest to a potential investor that mutual funds are not likely to be their best approach to obtaining wealth because a stock picker is not likely to beat the market I hear something like, “ my brother/sister/cousin/buddy/advisor, etc. beats the market all the time, its easy.” I am often tempted to say something like, “if it was easy, everybody would do it, and if everyone did it, then it would no longer beat the market”. But that reaction just irritates people because you sound like a condescending, smart-ass. However, if you ever push a bit deeper into the topic you often hear something like “Warren Buffet does it all the time, so I am sure that my guy/advisor/money manager can do the same.” So let me get this out of the way right now. The fact that Warren Buffett appears to beat the market says absolutely nothing about you (or “your guy”) because – you ain’t Warren Buffett.
But let’s be a bit more analytical for a moment. Any argument that advocates stock-picking has to build on the idea that the stock market is not efficient. If a market is efficient, the expected return of your stock picking will be no better than what the market equilibrium suggests. Most people erroneously think that this means that the Efficient Market Hypothesis (EMH) claims that all portfolios should have the same return, when its easy to find observations that show that they don’t. When I use the term “erroneously” here, I am referring to a classis misunderstanding. An efficient market will not have all portfolios yield the same return, and it does not suggest that all portfolios have the same expected return.
What’s “Efficient” mean again?
Some history is helpful here. Many researchers noticed over 100 years ago that stock prices seem to move in a random fashion. For several decades people compared different stochastic processes to stock price data looking for the best fit. In fact, for many years, economists argued that stock prices were so random that they were not informative at all. This started to change in 1965 when Paul Samuelson brought forward the idea that in a well-functioning and competitive market, we would expect prices to change as investors’ expectations adapt to new information. This theoretical explanation was called the “Fair game model” and started to bring meaning to the randomness of stock prices. In theory, stock prices should be a reflection of the expectation of future profitability and the riskiness of the related cash flows. Any time new information comes to the surface, this can alter our understanding of either future flows, or the variability of those flows (or both). Consequently, stock prices should, and do change with new information. The apparent randomness of prices is simply a reflection of the fact that new information arrives at random points in time. Samuelson won the Nobel Prize in Economics in 1970 based in large part upon these ideas.
This work was followed by a famous paper from Eugene Fama in 1970 entitled “Efficient Capital Markets: A Review of Theory and Empirical Work” (Journal of Finance 25:2: 383-417). While this work did review prior research, its main contribution was that it explained how an efficient market would have to work. For a market to be efficient it would have to be followed by a large number of decision makers, with free movement of capital, and the ability to trade with virtually 0 cost, in real time. The idea is that if the market is pricing an asset wrongly, enough people will see it and buy the under-priced asset, driving its price up, or sell the over-priced asset pushing its price down. As long as this can be done freely, quickly, and publicly, the market will be relatively efficient. Fama was recognized with the Nobel Prize in Economics in 2013, largely due to his development of the Efficient Market Hypothesis.
However, this hypothesis is largely misunderstood. A more nuanced explanation is that an efficient market will price an instrument in a fashion consistent with its expected flows AND the riskiness of those flows. A more risky portfolio will have a higher expected return. This must be true because, if it were not, no one would hold the more risky portfolio. Stated differently, a greater expected return must come at the price of enduring a higher variance. Market efficiency does not suggest that everything is the same. It says that the expected return will be consistent with the underlying risk involved.
Perhaps the best explanation of the next step in this analysis was given recently by Ben Felix on his podcast when he stated,
One of the challenges with the Efficient Market Hypothesis is that it cannot be definitively proved or disproved. This was acknowledged by Fama as the “Joint Hypothesis Theorem”. Any attempt to test market efficiency is really a test of two distinct hypotheses. It is jointly a test of the Efficient Market Hypothesis and a test of the model that you have of market equilibrium.
In other words, not everything should have the same price, but the expected return of a portfolio should be consistent with the way that the broader market prices the riskiness of the portfolio involved. To say that Warren Buffett’s returns prove that markets are inefficient, it’s not enough to show that his returns are high. You would also have to prove that they are higher than market equilibrium would allow. In other words, we have to see returns that a complete model of market equilibrium cannot explain.
A Model of Market What????
The first model of market equilibrium that finance students typically see is known as the Capital Asset Pricing Model (CAPM). Some people call this the “Beta” model. It suggests that a portfolio’s expected return should look like:
rt – (rt)f = α + β * MKTt
Starting from the far left, we have rt , which is portfolio return, and (rt)f is the risk-free rate of return. We call the difference between these two terms, the excess return. This effectively what you are being paid for your exposure to greater risk.
Starting from the far right, MKTt is the excess return of the overall equity market. Many people call this the “equity risk premium” and over recent history this value has averaged about 8% because the average return on an S&P 500 index fund is around 8% higher (on average) than the risk-free rate. β is a measurement of risk. For example, a β value of 2 suggests that the portfolio is twice as risky as a broader portfolio covering the entire market. Putting up with a higher value of β earns a higher return. This leaves α, which is your return above what the equilibrium predicts. Saying that your favorite stock picker is adding value is the equivalent of saying that for him/her α is statistically significant and above 0 over multiple periods.
Since β is the only term here that is portfolio-specific, we call this a single factor model. Using linear regression, considering historical data, we find that when we set α equal to 0, this model explains roughly half of the variance in returns that we see in practice. In other words, this model is better than nothing, but it’s still woefully incomplete and leaves lots of room for α to be either positive or negative. Proponents of the argument that Buffett proves the market is inefficient point to the fact that the CAPM suggests that his returns greatly exceed what equilibrium suggests – in other words Buffett’s α is positive and large.
More Complete Models
This conclusion is not justified, because the CAPM model is incomplete. Clearly Buffett did a number of brilliant things, but two really stand out here. First, he put as much money as he could in what Forrest Gump called “some little fruit company” which happened to be Apple. This accounts for roughly half of Buffett’s net worth today. If you put every penny you could get your hands on into Apple stock when Steve Jobs left the company in 1985 and held that stock until today, you would have beaten the market by a wide margin over most of that period, and it would appear that you have a high value of α.
Notice that I said “most” of that period – not all of it. When Jobs left Apple in 1985, its stock sold at $16.25 per share. By July of 1997, right after Jobs returned, Apple stock had fallen to $3.19 per share. (Yes, you read that right, about 3 dollars!!) Most people who had Apple stock in 1985 sold it long before that point, but Buffett did not. NOT selling his holdings in Apple, may have been Buffett’s greatest move ever. After we account for dividends and stock splits, we find that $1 million invested in Apple in 1997 would be worth roughly $1.4 Billion today. (Yes, I said Billion!)
But here is the more interesting part of the story. Buffett came up with his investment approach long before he started buying Apple stock. He didn’t buy it because he thought it was cool. He bought it because it fit with his broader approach. At the start of Buffett’s career he invested in companies that had high book to price ratios. This means the total of all assets as listed on the balance sheet, was relatively high compared to the company’s stock price. We call these Value stocks. Early in Buffett’s career he also focused on relatively small companies. (Apple was still relatively small in 1997.) Academics eventually figured out that these were both really good ideas. Our old friend Eugene Fama comes into the discussion again. In 1995 Fama and Kenneth French published a related paper that argued that if we add factors that reflect firm size, and book to price ratios as a measure of value, the resulting model explains much more of the observed variance. (Fama, Eugene F., and Kenneth R. French. “Size and book‐to‐market factors in earnings and returns.” The journal of finance 50.1 (1995): 131-155.) This 3-factor model also explained more (but still not most) of what Buffett accomplished. Finally, in 2014 Fama and French developed a 5-factor model that added terms to reflect profitability, and reinvestment. This model seems to explain over 90% of the variance in portfolio returns.
But wait, it gets better. More recently, three other researchers (Andrea Frazzini, David Kabiller, and Lasse Heje Pedersen) published a paper in the Financial Analyst Journal (74.4 (2018) 35-55) in which they argue that Buffett also focused on firms that had unusually low Beta values, and were of high “quality”. Quality, in this model refers to firms that are highly profitable, growing, and pay attractive dividends. In addition, Buffett used leverage (borrowed money) to increase his exposure to these risk factors beyond what you could do without using borrowed funds. It turns out that this rather idiosyncratic combination of factors explains what Buffett did, in the sense that the resulting model of market equilibrium produces an estimate of Buffett’s α that is not statistically significant and different from 0.
But my Guy is Special!!!
Of course, its relatively easy to explain what happened in hindsight, but here is the kicker. Buffett figured all of this out about 40 years before the rest of us. This reminds me of a famous line from the German philosopher, Arthur Schopenhauer that “Talent hits a target no one else can hit. Genius hits a target no one else can see.” Even if you convince me that your stock picker has talent, it’s going to take a lot more than that to convince me that he is a genius, especially if he is willing to sell you that skill for far less than it is worth.
The fact that an explanation now exists, does nothing to detract from the fact that Buffett is a genius. Quite the contrary. Whenever someone tells me that he is going to beat the market because he saw Buffett (or Peter Lynch, or George Soros or a few other people) do it is like telling me that you think you can dunk while jumping from the free throw line because you saw Michael Jordan do it, or that you are going to hit 50 3-point shots in a row because you saw Steph Curry do it. That may sound reasonable to you, but I am here to tell you with all the care and kindness that I can muster – you ain’t Michael Jordan, you ain’t Steph Curry and you ain’t Warren Buffett. In addition, I am willing to bet my house that your favorite stock picker isn’t either. In addition, remember that Buffett wasn’t selling this skill to you for a 2% fee. He was using it to build his own firm and holding onto the gains.
One last point. When you share this info with your advisor/broker, etc. remember what Upton Sinclair said in 1934; “It is difficult to get a man to understand something, when his salary depends on his not understanding it.” (This quote may have been adapted from an earlier statement from William Jennings Bryan who wrote in 1893, “It is useless to argue with a man whose opinion is based upon a personal or pecuniary interest.”) In other words, if a man’s pay is based on his ability to sell you on the idea that his favorite stock picker will beat the market, there is probably nothing you or I can tell him that will change his mind. My advice in such cases is simple. Wish them all the best and run like hell!!
