Winning the Loser's Game

In 1975, Charlie Ellis, a consultant to institutional investors, famously observed that investing was a loser's game. Like amateur tennis, he wrote in the classic Winning the Loser's Game, investing is an activity in which the victor often prevails because he makes fewer mistakes than his rival does. It is no different for financial advisors and your clients. You make too many mistakes. You follow

In 1975, Charlie Ellis, a consultant to institutional investors, famously observed that investing was “a loser's game.” Like amateur tennis, he wrote in the classic Winning the Loser's Game, investing is an activity in which the victor often prevails because he makes fewer mistakes than his rival does. It is no different for financial advisors and your clients. You make too many mistakes. You follow fads. You chase the hot manager. You take recent events and naively use them to anticipate future results. You are blinded by short-term successes. And you overreact to good news and bad news alike.

That's no way to make money.

Ellis' book, of course, was really advocating indexing (then still in its infancy), but those of us who do believe that active management has a role to play in our clients' financial lives must remember the cardinal rule of investing: Past is not prelude. None of the institutional investment consultants we interviewed uses past performance as an indicator of future success — at least, not directly. Why not? Simply put, because studies show that past performance — like it says in all the prospectuses — is, in fact, an exceptionally poor indicator of future equity returns. According to a study by the Frank Russell Company, of the 73 U.S. equity managers in the top quartile in 1996 surveyed, only 19 managed a repeat performance the following year, two stayed on top three years and just one lasted four years. After five years, none remained in the top quartile.

With odds like that, you might want to sidle into the camp of Ellis and his indexing ilk. And you wouldn't be wrong to join 'em — picking managers is a painstaking, research-intensive, specialized skill that many advisors simply don't possess. “Advisors who don't have or can't develop the skill [of picking winning managers] should just hold index funds and not play the active game,” says Barton Waring, managing director at Barclays Global Investments, who advises his institutional clients on the appropriate mix of active and passive investments.

Despite the popularity of exchange traded funds and index funds, most investors are not satisfied with indexing at the outset, because they like the game so much. They must be made to understand that by using active management they will be taking on an additional layer of risk (on top of systemic or market risk) in their portfolios in hopes of beating the market.

Picking “Winners”

Here's how some top institutional and retail advisors try to select successful managers.

The veteran advisors we spoke to always start off with a client by sitting him down to determine, as specifically as possible, how much risk he is willing to tolerate — to find the edges of the oft-mentioned “efficient frontier.” If an efficient portfolio is the portfolio that takes the least possible amount of risk for the greatest possible expected return (and Modern Portfolio Theory posits that it is) then it follows that risk is a currency that must be spent judiciously. Never should it be spent unless the client has a reasonable expectation of being compensated commensurately.

Once the client's appetite for risk has been determined, a detailed quantitative analysis of the manager's portfolio and performance must be executed. The first goal here is to separate a manager's return into its component parts: systemic return — the return attributable to that of the market (beta) or to the general volatility of the portfolio relative to its benchmark — and alpha return. Alpha essentially is return in excess of the market risk or of the general risk level of the portfolio; this sort of return, therefore, presumably is attributable to the skill of the manager. Look at it another way: After all other variables have been isolated, alpha attempts to answer the question, “What will this manager return without any help from Mr. Market?” The goal, of course, is to discern those managers with real, lasting skills from those who are merely lucky.

For help with confirming consistency, enter the Information Ratio. The Information Ratio is a lot like the Sharpe Ratio in that it's a measure of return per unit of risk. The difference is that while the Sharpe Ratio measures total returns against total risk (measured by the portfolio's standard deviation), the Information Ratio measures the manager's historic alphas against the volatility of his alphas. The more consistently a manager has been able to deliver a given level of alpha, the higher his Information Ratio. So, if two managers in a given asset class have equal alphas, but one has a higher Information Ratio, then he is the superior manager on a risk-adjusted basis. In general, you'll want the Sharpe Ratio to be higher than one, indicating that the manager is being paid for the risk he is incurring. (Tip: The Sharpe Ratio is, ironically enough, a blunt instrument. Because its denominator includes deviations in both directions, the Sharpe Ratio perversely punishes managers for volatility on the upside. To avoid this, some analysts use the Sortino Ratio, which is just like the Sharpe Ratio except that only downside volatility — either absolute or below a certain target return — is used in the risk computation.)

Capture the Market

Again, be advised: Though alpha is a tremendous tool, MPT statistics have little predictive value on their own. You'll have to assess whether the alpha delivered by the manager in the past is sustainable in the future. You'll also want to know the manager's upside and downside capture ratios. Simply put, capture ratio indicates how much of a bull (or bear) market a manager captures. An upside capture ratio of 1.25 means the manager's upside historically has beat the market by 25 percent during bull markets. A downside capture ratio of 0.75 means the manager falls only 75 percent as much as the market during declines.

Experts advise caution when using this metric. “You use the capture ratio in different ways at different times,” says Margaret Starner, a CFP and Raymond James-affiliated advisor in Coral Gables, Fla. “During bull markets like the one in the late 1990s, the deep value manager's not going to capture anything.” The bottom line: Don't just look for high upside capture ratios — make sure you take into account how any given manager is going to fit into your client's portfolio.

The next steps are to compare a manager or fund with an appropriate benchmark and identify how closely correlated he or it is with an accepted bogey. R-squared — a number obtained from a detailed comparison of investment returns from two or more portfolios — quantifies how much of a portfolio's volatility is explained by the movement of its benchmark (an R-squared of 100 percent indicates a perfect correlation) and its beta, a metric that quantifies how volatile an investment has been compared to its benchmark. (A beta of 1.3 means the portfolio is 30 percent more volatile than its benchmark.)

Look for any incongruities between the manager's stated methodology and objective and his actual performance. For instance, Phil Kosmala, a former SEC compliance officer who now is manager of investment research at consulting company DiMeo Schneider in Chicago, recalls one instance in which an analysis of one high-alpha manager revealed some incongruities vis-à-vis his benchmark. In essence, there was no way the manager could be doing as well as he was given the market in which he toiled. Kosmala took a closer look. It turned out that the manager was juicing up returns — and his alpha — by flipping IPOs in a hot but temporary market. “Not sustainable,” summed up Kosmala. The manager was removed from DiMeo Schneider's recommended list.

Style Box

Since even alpha can be manipulated in the short term, you should understand that you have to use this metric carefully and in conjunction with other statistics. If R-squared to the benchmark is low (again, below 0.80), then alpha is not a useful comparison. Alpha is only useful within a tightly defined style box comparing managers with very similar market cap weightings and performance objectives. Presumably, your firm's due diligence department has already considered such issues and examined the asset manager's transparency to verify his stated use of leverage and long/short ratios, among other metrics, in confirming his integrity. “I'm looking for a manager with a style I can pinpoint,” says Matthew Lum, an advisor with Sentinel Capital Management in Santa Barbara, Calif.

Others, however, argue that money managers don't like being pigeonholed, particularly hedge fund managers who, unlike their mutual fund colleagues, aren't necessarily bound to any particular strategy. “What you're really looking for from a hedge fund manager is absolute returns,” says Jamie Dinan, senior managing director of York Capital Management in New York. “You need to ask yourself, if the market goes up 10 percent will I make money, and if the market goes down 10 percent will I not lose money?”

True enough. But, as Barclay's Waring points out, when active managers do this, their alphas become much less predictable, and active risk becomes more significant within a portfolio. Critics of alpha say it's better to focus on low fees and low standard deviation of returns, which are harder to fake and have historically been better predictors.

Freedom from Quant

All quantitative measurements, alas, have limited predictive value since they are all fundamentally backward-looking. Roger Urwin, researcher at consulting firm Watson Wyatt in Washington, argues that qualitative analysis — the study of the four “Ps”: People, Process, Philosophy, Performance — is a better predictor of results. Urwin values permanent advantages, such as a more experienced management team, better expertise in a particular business model, and low turnover (People); access to information, or superior ongoing risk management techniques (Process); and a thought-out, disciplined, proven and well articulated approach to the art and science of investing (Philosophy). Often, these metrics can be ascertained only by visiting the manager, interviewing the staff and asking detailed questions about buys and sells. “Track records can't last forever,” says Lum, “but disciplines can.”

In practice, most due diligence is done by brokers' home offices or delegated to an outside consultant (think Barra RogersCasey, Frank Russell, Callan et cetera). “I talk to all of our managers personally,” says Starner. Smaller managers ($500 million or so in assets under management) are more eager for business and will make their portfolio managers and analysts available to advisors and clients and will even travel to advisor events.

Putting it All Together

Get one thing straight: Risk budgeting is not simply making a choice between utilizing all active management or all passive management. The idea behind risk budgeting is the judicious use of active and passive management strategies in tandem. Many advisors simply use the core-and-explore strategy, that is, they use passive ETS or indexes for core holdings (say, large-cap U.S. equities) to capture the overall market's moves and then, for alpha, go “exploring” with a specialists manager (say, small-cap growth). Either way, according to Waring, the best portfolio balances active risk and active return across several managers. Explore combining managers with low or even negative correlation statistics. Above all, advises Waring, “Maintain a sense of humility.” Identifying quality managers who will consistently add value is an exceedingly difficult task. Does it fit into your efficiency frontier?

“Key Terms”

Alpha: Excess return. Or more specifically, alpha attempts to measure the investment return from specific, non-market risk. Considered a measure of a portfolio manager's skill. The higher the alpha, the better the manager is at selecting stocks. Specifically, measures a manager's excess return over and above that predicted by his benchmark and beta. Answers the question, “What is this manager's return when the market return and the beta are both zero?”

Beta: Measures the volatility of a portfolio relative to a benchmark. A beta of one means portfolio is as volatile as the market. A beta of 1.10 signifies that volatility in price has been 10 percent greater than the market.

Capture ratio: Measures the extent to which the manager's portfolio typically participates in, or captures, bull or bear markets. May be greater than 1.00 for aggressive managers.

Efficient frontier: The Holy Grail of investing. A portfolio that provides the optimum return for a given risk level. An efficient portfolio either will have a higher expected rate of return than any other portfolio of equal risk or will be less risky than any other portfolio with the same expected return.

Information Ratio: Measures the consistency with which a manager delivers his alpha.

Quantitative analysis: The process for determining the strategy of a portfolio manager via measurable data, such as portfolio holdings and performance. Differs from qualitative analysis and relies on historical data.

Qualitative analysis: Attempts to quantify the non-measurable variables affecting asset manager performance. Centers on three of the Four Ps: People, process, philosophy. Attempts to be forward looking.

R-Squared: a measure of how closely a portfolio's performance has replicated its benchmark. A score of 1.00 means the portfolio tracked exactly the benchmark's movement. The more concentrated the fund, the less the portfolio will “act like” broad indexes, such as the S&P 500.

Sharpe Ratio: Attempts to measure risk-adjusted return. Calculated different ways, but often is calculated by taking the annualized return minus a risk-free rate of return divided by the standard deviation of returns. The higher the ratio, the better the fund's risk-adjusted performance. Should be compared to other managers and the benchmark. Attempts to answer question: Does the manager add value? Was he compensated in excess return for the risk assumed?

Sortino Ratio: A permutation on the Sharpe Ratio. Differs from the Sharpe Ratio in that upside volatility is excluded. Measures annualized rate of return for a given level of downside risk.

TAGS: Archive
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.