Finding Alpha with Active Managers
Robeco Investment Management
By Jay Feeney
May 22, 2012
Summary
Many investors are convinced that alpha has disappeared from U.S. equity markets and prefer to use passive investment tools such as exchange traded funds (ETFs) to broadly gain exposure to these markets. The problem with this approach is that it gives up any chance of outperformance
and forces an investor to settle for benchmark returns minus fees. It also ignores the fact that alpha potential does exist. Although many active managers have not done a good job in capturing alpha, there are many who have outperformed over time, producing very sizeable excess returns.
The Efficient Market Hypothesis is flawed
Before discussing the potential for alpha let’s consider the Efficient Market Hypothesis (EMH). There is a widespread belief that the EMH still holds true, particularly when looking at U.S. markets, but it rests on several questionable assumptions. Primarily, it assumes that people are rational decision-makers who use all new information to update their viewpoint and make optimal choices. The rationality assumption is profoundly flawed and there is ample empirical evidence that humans are prone to irrational choices, particularly in extreme market situations of rapidly falling or rising prices. For example, in basic economics we learn that demand curves are downward sloping, meaning lower prices increase quantities demanded. Yet there are many instances where the demand curve for stocks appears to be upward sloping, as investors are often drawn to a market by higher and higher prices.
There are several major implications of accepting the EMH. One is that in order to outperform, investors need to have an advantage with respect to information gathering or forecasting. To outperform the market you have to know something that no one else does if you are to take a position ahead of the market. This belief has had a major influence on how most institutional investors structure and staff their research process; the vast majority of investors make information gathering and forecasting the cornerstone of their decision making process. They hire larger numbers of analysts with impressive academic and professional credentials in a quest to know more and more about less and less. They not only engage in an unwinnable research arms race for information as they fight for superior insight on each and every stock in their portfolio, but then compound the mistake by trying to forecast an unknowable future.
We would argue that the real key to outperformance is building a research process that is analytical, not informational. In other words, an investment process designed to filter information more effectively, rather than forecast more precisely, should lead to better portfolio design. Technological and regulatory changes in the investment industry have commoditized information and levelled the playing field across managers large and small. Herbert Simon, a Nobel Prize winner, summed it up when he said that "information consumes the attention of the recipient. A wealth of information creates a poverty of attention." So in our view the key is to distill and transform information into useful knowledge.
Another implication of the EMH is its emphasis on alpha versus beta, and the ill-conceived notion that the two can be cleanly separated. This in turn has spawned the growth of ETFs in order to gain beta exposure, with the addition of "alpha only" managers. The underlying presumption here is that alpha is more or less a commodity that can be fabricated on a research assembly line. In reality, there is no such thing as "consistent alpha," just as there is no such concept as a perfectly consistent golf swing. Though alpha potential clearly exists it is episodic, with the greatest opportunities available by leaning against the wind of popular sentiment, when market stress is at an extreme.
The EMH has also influenced modern risk management – for the worse. As a consequence of this hypothesis as well as the Capital Asset Pricing Model (CAPM), risk assessments emphasize statistical measures of return variation, which in fact offer no insights about “real risk” - the risk of losing capital (drawdown risk). Uncertainty and risk are quite different in nature. Uncertainty is unavoidable because the future is unknowable and the value of a security is wholly dependent on future events. In contrast, real risk (the risk of loss) is avoidable, because it stems from three sources that can be analyzed here and now: paying too high a price, earnings disappointments and bankruptcy risk. These sources are obscured by top-down portfolio volatility measures (such as mean variance optimizers) which confuse uncertainty with risk.
Falling stock prices and consequent increasing volatility imply rising risk in a conventional framework; in fact the risk of loss declines as prices drop, since the return potential and margin of safety both rise and the downside decreases, the lower the price paid. One of the important lessons of the recent financial crisis is that risk management is best implemented via a bottom-up approach and there is no substitute for diversification and an effective sell discipline (the ability to admit that you might be wrong).
Uncertainty never dies.
Unfortunately, the investment world is characterized by “true uncertainty,” a situation where you cannot put a probability on anything, although most investors try to do so. The simple truth is that we do not face a distribution of probabilities but a distribution of possibilities. The real world is not like roulette where there are a limited number of known outcomes. It is similar to poker, and in that game the outcome depends on the behavior of the other players; it is therefore unpredictable. Further complications arise because choices evolve from previous choices. In other words, the system continuously adapts such that ultimately it is impossible to forecast. Yet most investors try to make forecasts anyway. The future is unknowable and the only certainty is that things will change, positively or negatively.
One of the biggest problems we face as an industry is that most investors are trying to predict the future. John Kenneth Galbraith said: "There are two types of forecasters – those that don’t know and those that don’t know they don’t know. The [latter] are the dangerous ones, the people who think they know what the future will behold." In investment management there are many who believe they have all the answers but this is simply not possible. Uncertainty always exists. It is what an investor does in the face of true uncertainty that matters.
The good news: potential alpha is everywhere
Where do we look for alpha in an uncertain world? Consider that the aggregate performance of an index only shows part of its true performance.
For example the return of the S&P 500 in 2011 was 2.1% overall, but 230 of the 500 companies posted a better return than the index, returning
20.8% on average. So alpha existed in the S&P
500 that year.
Alpha also exists over longer periods. The table on the next page shows the Index return for the
Russell 1000®, Russell Midcap® and Russell 2000® Indices, along with an equal weighted portfolio
of the best performing stocks in the Index, and an
equal weighted portfolio of the worst performing stocks for that year. It is clear from this data that there is a tremendous amount of alpha potential in the market, if you can correctly discriminate between winners and losers. Yet most of the asset management industry does not correctly discriminate and the average manager fails to beat the benchmark.
While alpha potential is most prevalent in the small and mid cap segments of the universe, there is alpha potential right across the market, as noted by the significant spread between winners and losers in all market cap segments.
Then why don’t most managers beat the market?
If alpha potential is everywhere, the ability to outperform exists. So why is it a widely held belief that the average manager cannot outperform the index? Let’s examine the type of stocks that outperform, and then consider how managers look for those characteristics in the market.
There is substantial empirical evidence that outperforming stocks have an underlying set of common features, a "genetic signature." They tend to be inexpensive, of high quality, and they have positive momentum. Managers who consistently seek this genetic signature should succeed over the long term.
Most managers lose to the benchmark for three key reasons. First, they make forecasting the centerpiece of their investment process; they spend a great deal of time gathering information and trying to forecast better than everybody else. As mentioned earlier, it is difficult to consistently outperform the market in this way. Knowing how Boeing constructs its 787 or how Apple builds the iPad offers no insight into whether these companies make good investments. Yet most industry analysts feel the need to delve into such minutiae. Second, they ignore the genetic signature of a good investment – its basic characteristics - by getting lost in the details. Looking for this signature does not imply finding every piece of information about a company. It does mean focusing on the information that allows an investor to determine whether a stock has the genetic code.
The third reason why managers fail to meet the benchmark is that they have behavioral biases; the battle faced by managers and analysts every day is how to counteract those biases. Worst among these is "confirmation bias," where investors spend most of their time gathering data that confirm their forecasts while ignoring data that contradict their theories. Another is the narrative or good- story bias where investors develop detailed and precise descriptions of their investment theory, and as they provide more detail it becomes more believable - but actually it is less likely according to the laws of probability. Adding additional detail to a story makes
it less probable, even though humans gravitate to detailed stories because they seem more credible and plausible. A manager has to be aware of these biases and make an effort to counteract them in a consistent way.
Assessing managers’ ability to generate alpha
Finding information and trying to forecast better is not the way to consistent success; rather it is filtering information and looking for the genetic signature that counts. With that in mind, how do we judge a manager’s ability to find alpha? The chart below illustrates the Four P’s: Philosophy, Process, People, and Performance.
There are several successful investment philosophies that can be used to beat the benchmark. In general, a manager needs to be able to explain his/her philosophy in very plain, simple language, preferably in no more than a few sentences.
When it comes to process, managers should be able to answer two questions very clearly and consistently: how do they identify good investments and why are particular names of stocks or bonds in the portfolio? The responses will determine how a manager finds the genetic code in those names.
The investment process can be top-down, bottom-up, quantitative, and/or fundamental. While there is not much evidence that one process is superior to the other, some evidence does suggest that bottom-up investing – security selection stock by stock – has an advantage over a top-down macro-thematic process, simply because the signal at the company level is more informative than the noise at the macro level. In today's environment, most people focus on the macroeconomic headline risks and ignore the favourable microeconomics of companies, particularly in the U.S. equity market. Empirical data show that microeconomic considerations
will ultimately dominate over macroeconomic sentiment. Quantitative and fundamental techniques both have strengths and weaknesses. Ideally a manager can combine the two methods to find the genetic code in a consistent way.
The culture of an investment firm is a key factor in determining its success. One of the challenges any firm faces is the temptation to be what it is not, leading to a loss of focus on its core investing capabilities. For example, a fixed-income firm might want an equity capability; an equity-only firm might want to add fixed income or a traditional firm might seek alternative investments. A manager should have a culture that concentrates on the investment side of the business, consistently aiming to generate returns for client portfolios, enabling the organization to remain focused on this goal over the long term. This culture should pervade the entire organization in order to drive long-run success.
In conclusion, successful active managers:
-Focus on what they know and understand. They do not waste time on future predictions.
-Concentrate on micro, not macro factors.
-Tend to be very clinical and process-driven.
-Are willing to tolerate periods of underperformance. Compound interest is maximised over the long run by avoiding large losses, not by beating the benchmark every single day.
-Constantly seek the genetic signature of a good investment to tilt the odds in their favor.
-Possess a stable team, a sensible philosophy, and a clear, consistent decision process.
Jon.Davis@robecoinvest.com
(213) 687-1667
Robeco Investment Management Disclosure
Robeco Investment Management, Inc. ("RIM"), an investment adviser registered with the SEC under the Investment Advisers Act of 1940. The views expressed in this commentary reflect those of RIM as of the date of this commentary. Any such views are subject to change at any time based on market and other conditions and RIM disclaims any responsibility to update such views. Past performance is not an indication of future results. Discussions of market returns and trends are not intended to be a forecast of future events or returns.
Indices
The S&P 500 Index is an unmanaged index of the common stocks of 500 widely held U.S. companies. The Russell® Value Indices typically measure the performance of universes of stocks displaying low price-to-book ratios and low forecasted growth values. The Russell 2500™, and 2000® Indices measure performance of the 2,500 and 2,000 smallest companies in the Russell 3000® Index respectively. The Russell Midcap® Index measures the performance of the 800 smallest companies in the Russell 1000® Index.
(c) Robeco Investment Management
