Why isn’t everybody doing it

Jan Mohr 12 Nov 2011

This post is all about how to survive a bar conversation as a banker these days.

I happen to be a banker and my friends happen to know that I am a banker. As the global media has lately been “occupied” by demonstrations of how bankers know sure ways to make money, I have been entangled in bar conversations around the stock market and techniques on how to make money in the stock market. So please let me share with you some content from a typical bar conversation I had recently. (Disclaimer: if you expect this post to become a little bit more quantitative and useful, keep on reading- if you actually think this post deals with how to survive a bar conversation as a banker, you might be well-advised to keep on reading here )

So my typical friend would ask me, with a gentle sub-tone of sarcasm: "as you are a banker and you work with the stock market, you can now tell me how to make money."

When I respond “yes, indeed”, my friend would normally not believe me and would ask for evidence. Luckily, it’s easy to show evidence in a blog (I have a harder time in a bar), so look here, here, here, here, or here We basically talk about a phenomenon of structural return patterns that can be explained by certain factors. In practice these factors come down to buying cheap instead of expensive stocks (value factor), buying stocks on the rise (momentum factor), avoiding stocks that have recently issued new shares (equity offerings factor), and several others. Let me tackle the most prominent of the factors which is the value factor. You have definitely heard about “value managers”, i.e. fund managers who are on the constant hunt for cheap stocks. Taking a look at value managers oftentimes reveals very attractive long-term track records as these guys capture a large chunk of the structural return outperformance of the value effect. In fact, buying cheap stocks and avoiding expensive ones will make you money, just take a look here.

Now my friend would ask me: "why in the world isn’t everybody doing it?" If you are keen to make money but have no interest at all in my elaborate explanation of "why", then feel free to scroll down to the “how” section in the very last paragraph of this article.

The scientists Fama and French got a great deal of attention for empirically testing the value effect (among other effects) and showing that it in fact yields an outperformance. These guys are both disciples of a breed that believes in markets being efficient everywhere and anytime. In their world, excess return can only be achieved by taking excess risk. As value stocks show excess returns they are riskier. QED, right?

Well, both gentlemen define risk as the variance of returns (vulgo: the more your return jumps around the riskier it is). Unfortunately for Fama / French, the returns of value stocks actually do not jump around as much as they would if the authors’ assumptions were right. Check Capaul, Rowley, and Sharpe (1993) for further reference.

My very bar friend might now argue that risk could be something else but the variance of returns. I couldn’t agree more, so let’s take a look at what Lakonishok, Shleifer, and Vishny tested in 1994. They were not satisfied by Fama and French’s risk concept and hence derived a different idea of risk. They say that if value stocks were riskier than growth stocks (i.e. expensive stocks), we could observe periods in which value stocks underperform growth stocks. Moreover, they regard the underperformance of an asset class in times of economic trouble as a good indicator for the embedded risk. In fact, during recessions, investors tend to be more risk-averse and get rid of risky assets. So if they get rid of value stocks during recessions- we prove that all outperformance of value stocks is just compensation for taking the recession risk! However, during the 1968 to 1989 timeframe, the authors could not show any relation between value stock underperformance and the occurrence of recessions. Indeed, they found only a few instances where value stocks underperformed at all.

As my bar friend is not stupid, he argues: "if it is not risk that causes value stocks to perform so well- couldn’t I generate a risk-adjusted outperformance easily?"

"Kinda", according to Lakonishok, Shleifer and Vishny. The argument of the authors is that the value effect is strongly related to several behavioral heuristics many investors apply. For example, human beings tend to have a hard time thinking about mean-reversion concepts. An ongoing trend is often extrapolated rather than assumed to revert to a general mean. The authors actually find very cool evidence regarding expected growth rates for stocks. For example, there is evidence that the companies with the lowest expected future growth actually show higher growth ex-ante than companies with the highest expected future growth. So much for analyst expectations! Moreover, the authors argue that positions in growth stocks are easier to maintain for institutional investors as they have a history of strong fundamental performance. After all: who wants to present an ugly duckling to the investment committee? Go Groupon! Go Green Mountain Coffee Roasters! (The latter trading for a mere 4.5x sales- after dropping 65% over the last three months)

"Good", my friend argues, "I am neither an institutional investor nor do I have to extrapolate every trend into the future- can I now safely make money by buying value stocks?"

"Wait", argues Jeremy Grantham of investment firm GMO. In his April 2010 speech at the Graham and Dodd breakfast he tears apart the value effect and presents to us a very convincing case of why value effect is related to a special risk when you look closely enough. Read the piece here, I am not exaggerating when I say it is one of the five best articles I have ever read on value investing.

Grantham takes a look at proprietary data and tests the endurance of value effect when the economy goes down a lot. Think Great Depression (post-1929) and Great Recession (post-2007). During these times, value stocks performed much, much worse than growth stocks. Please read his paper, the data is mind-boggling. In other words, the outperformance of value stocks is compensation for the near-wipeout investors suffer in a value portfolio when very hard times hit. For all of you who are familiar with the risk-management concept of value-at-risk: think about the outperformance phases for value stocks as the 99%-area of the return distribution. Only 1% of the time, disaster hits and your value portfolio gets impaired. Hence, the excess return of value stocks is somewhat of the discounted expected shortfall value stocks suffer during times of crisis.

Adding to his point, Grantham also mentions how well high-quality stocks perform over time. Contrary to what you would expect, stocks with healthy balance sheets, high returns on capital, and low earnings swings actually outperform the market. Quality is important, especially in times of deep economic trouble but value stocks rarely possess quality attributes.

My friend: "therefore my goal should be to a) buy value stocks to capture the institutional mispricing and b) avoid the large declines during deep recessions. How can I do that?"

I then go on to explain concepts that combine value and quality like the Magic Formula of Joel Greenblatt or F-Score by Joseph Piotroski. Both add to the list of value stocks an additional layer that kicks out fundamentally distressed companies. What you end up with is a mixture of cheap and good. "Does this make sense to you?" I ask my friend.

While my friend nods and orders another round of beer, check out the empirical results from such a cheap/good strategy here.