One of the most vocal critics of EMH over the years has been the legendary Warren Buffett (Trades, Portfolio). The famed Oracle of Omaha has been bashing the EMH since it was first popularized by Eugene Fama in the 1960s. Recently, a new academic paper analyzing Buffett’s career returns proves once and for all that the EMH does not hold in the real world.
Now, that’s not to say everyone can beat the market (a mathematical impossibility) or that the inefficiencies existent during Buffett’s tenure will continue into the future, but the proof of Buffett’s claims that value investing is counter to the EMH is in the numbers.
I’ve generally avoided the technical side of the investment management business on LBS, but today I’m going to geek-out a bit and analyze an academic paper that sheds light on how Buffett made his billions and became the world’s greatest investor in the process.
In their paper Buffett’s Alpha, Andrea Frazzini, David Kabiller, and Lasse Heje Pedersen analyzed Buffett’s returns from 1976–2011 and decomposed them to identify the primary factors driving Buffett’s significant alpha. For regular followers of Buffett, the results should come as no surprise, but let’s dig in and discuss the three primary sources of Buffett’s alpha …
In summary, the authors regressed Buffett’s returns against factor exposures commonly known to influence returns. The three factors that showed as significant are forehead-slappingly obvious.
- Value: Buffett has a tendency to favor low price-to-book value stocks.
- Safe: He bets against beta in that he favors low beta stocks.
- Quality: Finally, he favors quality companies (profitable, growing) over junky companies.
As a student of Ben Graham and proponent of value investing, none of us should be amazed by the fact that Buffett has favored cheap stocks. The authors use the standard academic measure of value in the form of the price-to-book value ratio (PB). This compares the market value of the equity to the balance sheetvalue of the equity. In some industries (e.g., financials and homebuilders), this can be a decent measure of value. However, in most industries, the use of PB is fraught with difficulties.
Graham’s favored measure of value was a normalized price-to-earnings ratio (PE), but he admitted that finding an exact intrinsic value is impossible. Instead he advocated investing only when the disconnect between price and value was large enough to account for the uncertainty in intrinsic value. It would be interesting to substitute other value metrics for PB, see
While we don’t know the exact measures Buffett uses when calculating the intrinsic value of a company, we do know he has always advocated buying stocks cheaply. Given the data covers such a long period of time, I believe it’s safe to accept that Buffett did indeed maintain a portfolio tilt toward cheap stocks.
Of course, this in no way will satisfy the EMH folks because the outperformance of value has been known for decades. The EMH proponents call this the “value premium,” letting us know that cheap stocks are riskier than expensive stocks and, therefore, require a higher return to compensate for that extra risk.
Now, go back and read that sentence again only, this time, reverse the argument to see what you are left with according to the “value premium”:
The most expensive stocks are the safest stocks to invest in.
So, buying Dutch tulips at the height of Tulipmania was the safest investment decision. Buying tech stocks with non-existent (due to a divide-by-zero problem) or massive PEs during the TMT bubble was the safest investment decision to make. Buying single family homes during the housing bubble was the safest decision to make.
Phew. Good thing these academics aren’t investment professionals. You’d go bankrupt every cycle!
I could go on and on with examples of how cheap stocks are actually safer than expensive stocks, but I don’t need to because the next two factors take the cake when it comes to refuting EMH …
In addition to having a cheap stock bias, Buffett’s returns have a positive tilt toward the betting against beta (BAB) factor. In showing this, the authors of Buffett’s Alpha become a little self-serving in that they reference their own work elsewhere, but they’ve nonetheless analyzed Buffett’s returns in a unique and novel way. Given that the results fit perfectly with the anecdotal information we have about Buffett’s investment process, I believe the results are worthy of our attention.
The BAB factor exploits one of the real-world flaws in the CAPM equation. In the original equation, all investors invest in the single portfolio with the highest return per unit of risk (Sharpe ratio) and leverage or de-leverage this portfolio to meet their return and risk preferences. This does make sense in a world without constraints.
The leverage part is, of course, the real-world constraint that cannot be ignored. The vast majority of investment funds are simply not allowed to use leverage, and it’s very difficult for retail investors to access leverage. You can probably thank ERISA for that one, but that’s another post for another day. Meanwhile, this constraint, which is intended to be for the benefit and safety of the ultimate investor, creates a market inefficiency and ends up hurting the investor in the long run. This is because funds tend to overweight risky securities in order to hit their required returns in the absence of leverage.
Buffett, on the other hand, has access to cheap leverage. How? Through the insurance float he talks about in every annual letter he’s ever written. The insurance float consists of insurance premiums paid into his insurance companies. While that money is just sitting there waiting to be paid out in claims, Buffett puts the money to use by investing it. Voilà! Leverage! Throw in the fact that Buffett seeks out simple, low beta businesses (e.g., BNSF and Coke) and you’ve got yourself exposure to BAB.
In another reference to their own prior work, the authors examine Buffett’s returns against a quality-minus-junk (QMJ) factor that pits high-quality stocks against low-quality stocks. Quality is defined by the authors as the following:
- Profitability: Several methods of profitability are used, including earnings, cash flows, accruals, and margins.
- Growth: The trailing five-year growth in the profitability metrics.
- Safety: This infringes on the BAB factor a bit, but it measures price safety (market beta and volatility) and fundamental safety (low leverage, low credit risk, low volatility in profitability).
- Payout: Measures how shareholder-friendly management is. Are they hoarding free cash flow for empire building? Or are they returning it to shareholders?
First, regarding size, we are all aware of the small cap anomaly whereby small cap stocks have outperformed their large cap brethren over time. Now, the EMH folks like to point out that small cap stocks are junkier and thus carry higher risk, which is the reason for the higher return. I cannot disagree with this assertion as a generality. However, the authors took this a step further and overlaid the small-minus-big (SMB) factor on the QMJ factor to control for size. The results are astonishing, so I’ll let the authors speak for themselves on this one:
“While SMB has an insignificant alpha of 13 basis points per month controlling for the other standard factors, this increases to a highly significant alpha of 64 basis points (t-statistic of 6.39). In other words, when comparing stocks of similar quality, smaller stocks significantly outperform larger ones on average, which corresponds to our finding in price space that larger firms are more expensive.”
Whoa! Sixty-four basis points per month! That’s over 700 bps per year just from buying high quality small cap names irrespective of price! Except this is bad news for Buffett. Why? Because he’s done so well and Berkshire has gotten so big, small cap stocks are not in his coverage universe anymore. He has to go after the big, liquid names, which are more expensive. It also explains why Buffett’s outperformance has diminished over time. When he was running the partnership pre-Berkshire, he was able to go after high quality, cheap, small cap stocks. Today that is not the case.
Another takeaway from the authors is that the price of quality changes over time. In fact, the price of quality was at its relative cheapest in 1987 prior to the market crash, in 2000 prior to the TMT bubble bursting, and again leading into the GFC in 2007. A rising tide does lift all boats, but as Buffett himself says, when the tide goes out you get to see who’s been swimming naked.
THE NOT-SO-SECRET SECRET EXPOSED
Well, there you have it. Everything Warren Buffett (Trades, Portfolio) has been saying for the past 50-plus years has been borne out in his portfolio management. The authors’ conclusion is so simple that it is likely to make many readers respond sarcastically, “If it’s so easy … ”
Which brings me to my take-home point: Investing is simple, but it is never easy. Warren Buffett (Trades, Portfolio) identified early on in his career the value investing tenets that would guide his investing career. He then had the audacity to apply those concepts year in and year out regardless of what was happening in the world around him.
Buffett’s worst years were at the height of the TMT bubble when the market saw his way of investing as out of date and not fit for this “new era” of investing. (Side note: Whenever something is said to be in a “new era” or of a “new paradigm,” do yourself a favor and short the ever-loving shit out of those securities—it’ll be like winning a rigged lottery because all those words really mean is “BUBBLE!” and bubbles always burst).
Despite this external pressure, Buffett stuck to his process and made an absolute killing in the tech wreck years that followed. Being a value investor often means you are taking a contrarian view by definition, which can be hard for humans. We are hardwired to take our social cues from the herd, so maintaining an opposite view requires determination and guts.
Do your homework better than the other person, stick to your process, and you too can tilt the odds in your favor. Unless, of course, you still believe EMH holds tightly. In that case, please park your money in a passive vehicle, and then the rest of us will clean up in the aftermath of a 100% passive market.