Arnott: Does My Beta Produce Alpha?

CAPM / Alpha Theory 08 Dec 2006

Your humble scribe is reporting to you right now from the cabin of an Embraer 190 having just left LaGuardia Airport in New York City. I’ve spent the day having my brain overloaded with alpha from some of the industry’s bigwigs at Institutional Investor’s “Second Annual Alpha Generation Forum”.  Today, I cover Rob Arnott’s keynote presentation while over the next couple of days, I hope to provide you with other observations & notes.

According to popular mythology, Rob Arnott was a precocious kid in elementary school.  It is said that in the second grade, he calculated the curvature of the earth.  As a 3rd grader, he gave his class weekly lectures on the (then) nine planets.  And on his tenth birthday, he tried to invest the $7.50 his grandmother gave him.

Today, he is a prolific commentator who continues to display an energetic intellect and a healthy disrespect for conventional thinking.  Arnott is well known as an advocate of fundamental indexing and is a veritable thorn in the side of the traditional market indexers upon which we all rely for “beta”.

So Alpha Male was somewhat surprised to see Arnott on the agenda for an alpha generation event. But as he began his remarks, it became apparent that his unorthodox definition of beta has important implications for the alpha world.  And as a result, Arnott argues that alpha and beta are actually inseparable.

Traditional Equity Indices: Goosing Alpha?

For those not familiar with Arnott’s work, he eschews common cap-weight equity indexes (e.g. the S&P500) in favour of fundamental indices that weight companies based on business fundamentals such as sales, book value, and cash flow, rather than simply on market cap.

Market cap-weighted indices, he says, essentially over-weight over-valued stocks and under-weight under-valued stocks. He points to overvalued Internet stocks of the late 1990s as an example of how a cap-weighted index is not actually a good representation of the broader equity market.

His firm, Research Affiliates, has developed its own fundamental index called RAFI (Research Affiliates Fundamental Index). With a flurry of charts, he showed the audience that RAFI outperformed the S&P500 by 25% in a recent 10 year period.  It also outperformed in all sub-sectors and even beat cap-weighted in index in all foreign markets in which it was tested (see article in the Journal of Indexes).

After laying this groundwork, Arnott asked an interesting question – one that brings together the worlds of beta and alpha –  Is RAFI adding alpha?  After all, RAFI seems to consistently beat the market on a risk adjusted basis. But on the other hand, it is purely passive in its construction.

During Q&A, an audience member asked Arnott the same question from a slightly different angle: Isn’t RAFI just another active (fundamental) strategy?  Arnott shot back that RAFI is arguably more passive than a cap-weighted index. After all, overweighting over-valued (higher PE) stocks is itself an active investment decision.

So if Arnott is right and RAFI is a truly passive strategy, then we’ve been comparing ourselves against the wrong benchmark all these years.  And if so, alphas would need to be recalibrated across the board and the active management industry would look a lot less attractive.

At the end of the day, Arnott says we shouldn’t waste too much time on this largely theoretical question.  In his opinion, alpha/beta bifurcation is somewhat of a red herring and we should just accept returns for what they are.  (North American readers may be reminded of that old Reese’s Peanut Butter Cups commercial where a person eating chocolate bumps into a person eating peanut butter and they both declare that the (fortuitous) combination “tastes great!”.  But in this case, Arnott might re-write it: “First guy: Hey, you got alpha in my beta!  Second guy: No, you got beta in my alpha!  Together: Mmmm, performs great!”) But I digress…

Dividends: Modigliani & Miller turned on its head

Not content to just take on the historical record of the entire active management industry, Arnott also questions the commonly-held assumption that dividend policy is irrelevant. It is assumed that a high dividend payout ratio is a signal that profitable internal projects are in short supply.  But according to Arnott’s research, dividend pay-out ratio is actually positively correlated with future income growth.  Management, he says, will channel precious capital only to high-profit initiatives, leaving lower return projects for another time.  While he agrees with Modigliani & Miller under its strict assumptions (no taxes, symmetric information, efficient markets etc.), he holds that these assumptions are simply unrealistic.

Therefore, dividends are not just relevant, They are far and away, the most import element of returns according to Arnott.

Cap-weighted Indices: The Finance Industry’s Pluto?

In conclusion, Arnott accuses the finance industry of being too quick to hold onto theory in the face of conflicting empirical data. In the field of physics, he explains, scientists rejoice when conflicting empirical data is uncovered since it gives them an opportunity to advance their theories and create new ones.  But in finance, Arnott says we tend to question conflicting data sets and conclude that they are either flawed or simply too small.

Which brings us full circle back to Arnott’s grade three lectures on the solar system.  The ability for science to question long-held beliefs was tested earlier this year when Pluto was demoted after 75 years from a full-fledged planet to a “dwarf”.   …If only change was so easy in the world of finance.

– Alpha Male

Be Sociable, Share!

One Comment

Leave A Reply

← Passive Management Rules! No Wait, Active Management Rules! Alpha as a Net Zero Sum Game: How Serious a Constraint? →