The conventional wisdom is that ETFs and other index-tracking vehicles are designed for beta (market exposure) and that active managers pursue alpha (value added through skill). But what does this actually tell us?

Do our well-used Greek letters help us make sense of the investment landscape, or do they actually hamper us in managing money? As James Montier, formerly of Societe Generale’s asset allocation team, now with Boston-based fund manager GMO, pointed out in an article published in 2007, as soon as you use the terms “alpha” and “beta,” you are invoking the spirit of the capital asset pricing model.

And, unfortunately, CAPM doesn’t actually work in practice.

Why not? Apart from some questionable assumptions about frictionless trading and investors having identical goals, the key problem with CAPM is that it assumes that stock returns are normally distributed. In other words, the theory requires that stock prices follow a random walk, with the price movement in one period entirely independent from that in all previous periods, and having no bearing on the future, either. This “Brownian motion” assumption produces the famous bell curve of statistics when one measures the percentage gains and losses over many time intervals.

However, while the bell curve accurately maps many phenomena in nature—people’s heights and weights, for example—many studies have now shown that it is inaccurate when describing financial market movements. Stock prices, which reflect the collective mood of millions of people, move according to far wilder trajectories, and there is plenty of evidence that markets have “memory”—which shows up in the serial correlation of returns. Even volatility tends to occur in “clusters.”

By viewing the world through the lenses of “alpha” and “beta,” you are automatically assuming a linear relationship between risk and return, with risk measured only according to a bell curve framework. Take away the framework, and the terms have no real meaning at all. So why do we persist with them? Part of the reason for the continuing popularity of the Greek letters is that we use them as a kind of shorthand—beta for passive, systematic, indexed; and alpha for active, more subjective, discretionary. And, of course, in the fund management world, while beta has meant (relatively) low fees, the claim of being able to source “alpha” has been the road to riches for fund providers.

But, just as it’s possible (and indeed, statistically likely) for the self-proclaimed alpha-seeker to deliver below-market performance, it’s just as likely that well-thought-out, entirely model-driven systematic approaches can generate superior returns over time.

At this week’s Terrapinn ETF conference in London, Chris Sutton of Watson Wyatt predicted a change in trend in the investment management industry, away from the artificial separation of alpha and beta and towards a more holistic approach. Sutton also argued that the ETF world needs to move from a period of product proliferation to one of increased innovation in index design.

I agree wholeheartedly with this, and would like to see a great deal more systematic strategies offered to investors in ETF format. As ever, the challenge will be to make these investment approaches understandable and transparent, without losing their competitive edge. But there are surely great opportunities out there for those who can hit the correct fund design. Isn’t it time to send at least two of our friendly “Greeks” back to the language classroom?

Last 5 posts by IndexUniverse Staff

• Will UNL Beat UNG? - November 19th, 2009
• A Real Hedge Fund ETF? - November 17th, 2009
• ProShares’ Positive Tax Surprise? - November 17th, 2009
• Better Than Cash? - November 17th, 2009
• Stoxx Indexes Bought; Company Valued At $900M - November 13th, 2009