Arnott's Fundamental Indexing Finally Shines

By Matthew Hougan

Well done, Jim: Rob Arnott invites you to his board meeting, and you call him out for under-performance. (See Jim Wiandt's blog about the Research Affiliates board meeting here.) Remind me not to invite you over to dinner anytime soon.

I'm kidding, of course: Rob can take it, and there's no getting around the terrible performance of the RAFI indexes in 2008. The indexes made a big bet on financials, and we all know how that turned out.

Recently, however, that relative performance has reversed. Over the past month, the fundamentally weighted PowerShares FTSE RAFI US 1000 ETF (PRF) is up more than 20%, while the cap-weighted iShares S&P 500 ETF (IVV) is up just 11%. That's a huge performance gap for a single month.

What's driving the outperformance? If you approach it from a sector level, you can see that PRF is making a few concentrated bets.

Specifically, PRF has made huge relative bets on financials and consumer discretionary, while underweighting technology, health care, consumer staples and energy.

So what's working?

The people behind the Select Sector SPDRs have created a great tool on their website that lets you plot the performance of all nine Select Sector SPDR ETFs vs. the S&P 500 over various time periods. It's available here.

If you plot those sectors (note: Select Sector combines telecom and utilities), only three have outperformed the S&P over the past month: financials, which beat the index by about 10%; industrials, which beat the index by about 3%; and consumer discretionary, which beat the index by about 2.5%.

PRF has been overweight all three.

Meanwhile, the worst-performing sectors have been, in order: utilities, health care and consumer staples.

Essentially, PRF has been in all the right places.

Is It Just Value?

A common critique of the fundamental indexes is that they are just value indexes in disguise. In this case, however, that does not appear to be the case.

Over the same time frame that PRF is up 20%, the iShares S&P 500 Value ETF (IVE) is up just 13%. That beats the S&P 500's 11% return, but it doesn't come close to PRF's 20%.

Again, sector bets are driving returns.

PRF has big relative weightings in financials, energy and consumer discretionary stocks; IVE is making big bets on utilities, health care and telecom.

Does this mean you should rush into PRF? Not in the least. That question depends in large part on your outlook for the financial sector.

What it does say, however, is that PRF is capturing an interesting trend in the market right now; that the fundamental vs. non-fundamental divide is providing more information on value than growth vs. value. And that, to me, is interesting indeed.

Original post

Comments

[Stephen Yu:]

Yes, interesting indeed and thanks for providing this data. I'm a fan of Arnott's, and I believe that fundamental indexing does have some theoretical advantages over cap-weighted indexing over the long-run. However, as the author says, one month's outperformance is not enough data point to fully support fundamental indexing. Longer term observations are needed. Also, as time goes by and more data becomes available, fundamental indexing can be enhanced by modifying the parameters used in security selection and weighting to increase its advantages over the cap-weighted alternatives.

Apr 30 10:25 PM

[Alan Young:]

It's nice to see, for a change, that there is value in fundamentals. Yet this report seems to conceal more than it reveals about RAFI indexing. For example:
Why are the sectors so lopsided? Is this an ongoing feature of the index, or just a circumstance of the market? If it's ongoing, the advantage will evaporate when the market favors other sectors; but if the index rotates sector preference as market conditions change, what are the odds that it will continue to favor the sectors that are outperforming? And ultimately, what variables determine when this indexing system is going to outperform and when it is going to underperform?

In addition to PRF, there are RAFI index funds for separate sectors, which can be evaluated in a totally different way from the "sector bias" study above.

Apr 30 10:46 PM