So much emphasis is placed on fundamental analysis of individual stocks and broader market indices in the popular press and in general discussion about investment products that technical analysis can get lost in the wash. What is technical analysis, for those not necessarily familiar with the concept? Without getting too far off the reservation, technical analysts generally use indicators based on stock charts, stock price movements, or other details to make buy and sell decisions. Generally, indicators fall into one of the following categories: trend-following, which includes indicators such as moving averages; overbought/oversold indicators; momentum indicators; or indicators providing information on market “internals” or market “breadth.” Practitioners can use indicators in isolation or in combination with other indicators. Time frames for analysis range from yearly data down to the microsecond. There’s an extensive body of literature out there dealing with the subject.
Fundamental analysis is the “backbone” of the industry’s stock selection and allocation process, and rightfully in our opinion. Many studies have shown that using fundamental indicators and ratios such as Price to Book over longer periods of time can provide a statistically significant advantage. On the other hand, studies on many technical indicators, especially studies focusing on single indicators from a pure performance perspective, have shown inconsistent results at best, leading some to declare technical analysis a “voodoo” type approach to stock selection or market allocation. We think this categorization is unfair and that many critics often miss the forest for the trees when it comes to incorporating technical techniques into a largely fundamental process. Yes, there are plenty of individuals, funds, quant investors, and the like using technicals solely to drive decision making. Many do quite well from a pure performance standpoint. We tend to look at it, and appreciate it, from a slightly different perspective, namely a risk management and discipline perspective. Instead of focusing on the performance attributes of indicators, we think it’s just as important if not more important to think about how incorporating this analysis into a fundamentally driven process can help one maintain objectivity, consistency, and discipline in process, and avoid the pitfalls often associated with our bad behavioral human traits.
Again, in theory, fundamental analysis can generate excess return if approached properly and in a disciplined manner. Unfortunately, humans have always been and continue to be susceptible to human psychological mess-ups that end up undermining good fundamental work. Behavioral finance has become a hot topic in recent years as academics and professionals have come to appreciate the performance problems associated with our psychological shortcomings. What are some of the behavioral short-comings, quickly? We’ll quote directly from a CFA Institute paper on behavioral finance:
• Overconfidence and overoptimism—investors overestimate their ability and the accuracy of the information they have.
• Representativeness—investors assess situations based on superficial characteristics rather than underlying probabilities.
• Conservatism—forecasters cling to prior beliefs in the face of new information.
• Availability bias—investors overstate the probabilities of recently observed or experienced events because the memory is fresh.
• Frame dependence and anchoring—the form of presentation of information can affect the decision made.
• Mental accounting—individuals allocate wealth to separate mental compartments and ignore fungibility and correlation effects.
• Regret aversion—individuals make decisions in a way that allows them to avoid feeling emotional pain in the event of an adverse outcome.*
Bottom line: we’re oftentimes our own worst enemy. We just can’t help ourselves. We buy the hot stock because all the news stories are glowing. We ignore an undervalued stock that has all the hallmarks of a good fundamental buy from a valuation and fundamental stance because the last several articles we’ve read are cruddy. We add to positions at exactly the wrong moment after they’ve run significantly higher and sell positions after they’ve collapsed because we just can’t take the pain anymore. We seek out information that confirms our current ideas and theories and ignore news and information that tells us we may be wrong, a problem that’s been enhanced in the information age.
Incorporating simple technical techniques can help avoid these types of mistakes, both at position entry and position exit. With practice in terms of figuring out what indicators may work best in the broader fundamental portfolio context, rules can be set and used that inject complete objectivity into a process, thus undermining the ill-effects of the psychological biases listed above. Using longer-term moving average, for instance, or longer-term momentum indicators can provide confirmation and trigger points for stocks or indices that are attractively valued, giving us confidence that a stock is under “accumulation.” On the flip side, simple technical techniques can provide early warnings or triggers that a portfolio position is “breaking down.” Ultimately, the holy grail is to allow winners to run and to cut losing positions before they become albatrosses. This sounds great in practice, but is ultimately difficult to implement when relying on human judgment alone. Under some market circumstances, this type of approach to the problem can enhance returns. More importantly, this type of approach can demonstrably reduce risk of massive drawdowns and thus improve the risk-adjusted return of a portfolio over the intermediate and long-term.
Let’s use a very simple example provided by the Big Picture blog earlier this year, which was originally calculated and demonstrated on Mebane Faber’s website.** Faber analyzed Dow Jones Industrial Average returns from 1922 to 2012 using a simple timing model based upon the 10-month moving average. Basically, if the index closes below the 10-month moving average at the end of a month, sell. Buy if it closes above. Meanwhile, during periods when out of the market, place the money in T-bills or the 10-year Treasury. Here are the results:
While annualized returns were essentially the same in this case (very simple model, again), the risk metrics over time are demonstrably better. This strategy produced lower overall standard deviation (over 30% lower), higher Sharpe ratios, higher Sortinos, and a maximum drawdown (during the Great Depression presumably) that was basically half of the buy and hold strategy. In practice, returns for many investors would have probably been much worse over time because of the biases we listed. As demonstrated over the past two decades, many individuals and professionals alike have tended to dump equities at or near long-term market bottoms, and buy wholesale at market peaks, irrespective of the fundamental metrics arrayed ahead of them. How many “new paradigm” metrics were created, for instance in the late 90s to justify buying stocks at 40x earnings?
A final note: incorporating rules into a fundamentally-oriented portfolio can create “whipsaws” in terms of position entry and exit. No system is perfect. This key over time in our opinion, though, is to avoid putting oneself in positions where you face a really bad decisions and heartburn. It’s precisely those moments where massively destructive portfolio decisions are made. How many folks were fully invested in late ’08 and early ’09 who decided they “couldn’t take the pain anymore” and decided to completely liquidate at market lows? Many of these same investors have missed a good portion of a rally over the past four and a half years. Emotion and psychological biases played a huge part. We’re only human and can’t turn our brains off. Combining technical processes with fundamental metrics can help overcome the destructive fight or flight portions of the investing brain.
*Alistair Byrne and Mike Brooks. “Behavioral Finance: Theories and Evidence.” The Research Foundation of CFA Institute Literature Review. Volume 3, No. 1. 2008.
**Barry Ritholtz. “200 Day versus 10 month Moving Averages.” 2/13/13. http://www.ritholtz.com/blog/2013/02/200-day-versus-10-month-moving-averages/
