The Altercation over Indexation

CAPM / Alpha Theory 26 Jul 2007

Continuing the proud tradition of the “Rumble in the Jungle” and the “Thrilla in Manila“, Financial Planning Magazine hosted what it called the “Fundamental Indexation Smack Down” last month between the inventor of fundamental indexation Rob Arnott (the Patent King of Pasadena) and Gus Sauter, CIO at Vanguard.  The entire 15 rounds was just released in a 60-minute webcast – and there’s no annoying $30 pay-per-view charge. (see entire video here, audio here)

It’s actually a pretty interesting 60 minutes.  (Thankfully, even more exciting than the 60 minute SEC meeting we watched earlier this week.)  However, if you’re pressed for time, you will find that both Arnott and Sauter make their key arguments in the first 25 minutes (but you will miss Arnott’s taunt in Round 7 when he pointedly asked Sauter, “Why are you so scared of this?” – not quite Jack Nicholson’s “You can’t handle the truth!”, but a high point nonetheless).  Whether you make it to the end or not, you will find it’s a great way to get up to speed on the arguments for and against fundamental indexation without having to reading mind-numbing academic papers.

It becomes apparent early on that what separates these two prize fighters is their confidence (or lack thereof) that the market is able to efficiently price securities.  Arnott says no:

“If we accept the notion that price is only the market’s best guess at fair valueand if we cap-weight our portfolios, we will assuredly be above fair value weighting on every company that is above its fair value and below fair value weighting on every company that is below fair value.  That creates a return drag – not relative to the market – but relative to the opportunity set.

“The fundamental index is an elegant and very simple way to break that link between over- and under- valuation and the weight in the portfolio.

“Why do we have to index by market cap any way?  If we choose a selection of (business metrics), we can get a nice composite gauge of the company’s footprint in the economy.”

Arnott’s essential argument is that greed and fear distort prices away from “fair value” (using a company’s “economic footprint” as a proxy).  As an advocate of cap-weighting, Sauter believes that prices represent fair value by definition – that all available information is baked into security prices.  The two basically reached an impasse on this somewhat philosophical point.

Late in Round 4, Sauter delivered a flurry to the solar plexus with charges that Arnott was basing his “fair value” on backward looking account data, not future prospects.  Again, Arnott countered by acknowledging that such “accounting” data would indeed produce errors (vs. fair value), but that those errors were randomly distributed.  Cap-weighting, continued Arnott, produced asymmetric errors since it always overweighted high multiple stocks – some of which don’t deserve to be overweighted.  In Arnott’s words:

“We don’t know which companies are undervalue and which are overvalued.  But we know that cap weighting by its very nature will have most of your money in over valued companies.”

It’s all relative, said Sauter as he landed a left hook to Arnott’s mid-section.  When compared to cap weighting, Fundamental indexation amounts to a value bias with a small amount of small-cap bias.  These are simply Fama/French factors that everyone already knows produce out performance.  So why use your fancy-shmancy index?

Clearly prepared for this argument, Arnott took a somewhat conciliatory tone.  He admitted that fundamental indexation always had some amount of value-tilt.  But he said that bias varied widely depending on the prevailing market regime.  In times when value outperformed growth, fundamental indexation out performed cap-weighted indexation handily.  But in times when growth outperformed value, the opposite was not necessarily true.  In fact, he argued, fundamental indexation often topped cap-weighted by a modest amount in such a “headwind” for fundamental indexation.  Making the same point about cap sizes, he said that the index currently had large-cap bias – a fact he said “comes as a surprise to traditional indexers” since they generally perceive fundamental indexation as having a small cap tilt.

Sauter was quick to launch a counter attack with a one-two combination of a jab and right hook aimed at the high correlation between fundamental indices and mid-cap (cap-weighted) indices, telling Arnott through his mouth guard, “It’s remarkable how closely it tracks!”

Somewhat frustrated and bobbing a little by this point in the fracas, Arnott replied, “That’s a smokescreen.  Fundamental indexation is not mid-cap value!”

Sauter explained that he did not mean to say it was literally made up of mid-cap value companies, but that the return stream was highly correlated with mid-cap value.

One of Arnott’s best punches was one aimed at the exaggerated market weights resulting from large cap growth names such as Cisco in 2000 (to this we’d add: Nortel, which at one point represented nearly a third of the TSX, and Nokia which represented nearly three-quarters of the HEX at its peak).

“What is it about cap weighting that says it make sense to own 4% CISCO at 130x earnings at the top of the bubble. When 2 years before it was .4% of the index at 30x earnings.”

When asked by the moderator if fundamental indexation was an active strategy, Arnott replied that it was all relative.  He recounted how William Sharpe told him fundamental indexation was a very interesting “active strategy” when he first learned about it.  According to Arnott, Sharpe had a very “cap-weight-centric” view of the world, but Arnott actually viewed cap-weighting as an active strategy favouring high multiple names.

Sensing an opportunity, Sauter challenged Arnott on fees, asking why Arnott was charging “premium fees” if his strategy was apparently just passive.  Arnott responded that fundamental indexation adds value and that the fees were closer to passive fee levels than they were to active fee levels.

When Sauter was asked whether the S&P 500 was active or passive, he explained that the S&P 500 wasn’t purely passive, since a committee was responsible for security selection, but that he wished it were.  Arnott piled on, saying that only 400 names in the S&P 500 were actually among the 500 largest public companies.

And that led to the “fundamental” question in this debate: What is the definition of an index?

Sauter argued that an index was meant to show “how investors had fared” in the markets.  So cap-weighting was the best way to show how much money the average investor was making.

Arnott believes this “original” objective of indexes makes a lot of sense.  But, indices are not well suited for passive investment strategies.  This was a development that came along well after the advent of the indices themselves.

Both seemed to agree that an index needs to be “formulaic, replicatable, transparent and low turnover”.  But Arnott also clearly emphasizes the relative performance of fundamental indexation.  He said his fundamental index has achieved new highs in 155 months since January 2000 (compared to only a handful for the S&P 500). 

Which begs the question, is Arnott trying to provide an index, or is he trying to actually make money?  (Students of hedge fund indices will recognize this dichotomy in the way hedge fund “indices” often market themselves based on performance.)

After sparring for a full 15 rounds, Financial Planning mercifully called it a no-decision and promised fans a re-match at some future date.  Early rumours are that Don King will moderate the next bout and will call it either “The Provocation Over Correlation“, “The Brawl over Vol” or “A Fightin’ Session on Fund Regression“.

Addendum: If this were a tag team affair, Sauter would surely have liked to have Wilshire Associates in his corner.  That firm just published a report attacking several of the claims made in favour of fundamental indexation.

Be Sociable, Share!

3 Comments

  1. Andy Herdman
    July 29, 2007 at 11:08 pm

    For another suggested explanation of what generates the apparent edge in fundamental indexing, see the article “It’s The Volatility, Stupid” by Ron Hylton at http://www.indexuniverse.com
    http://www.indexuniverse.com/index.php?option=com_content&view=article&Itemid=34&issue=110&id=2631


Leave A Reply

← Hedge Funds as Shareholder Activists from 1994-2005 Sell-side not a bunch of snake-oil salesmen after all →