If you build better indices, the possibility of enhancing asset allocation strategies naturally follows. Why, then-in a world that's minting a new generation of benchmarks at a record clip-isn't there more discussion of the opportunities for improving portfolio design? Whatever the answer, it's not for lack of choice.
The market is flush with new indices and related ETFs and index mutual funds that claim an edge over their predecessors. Some date the new world of indexing to 2003, with the launch of the first equal-weighted S&P 500 ETF: Rydex S&P 500 Equal Weight (NYSE: RSP). True, Morgan Stanley had been running an equal-weight S&P index mutual fund since 1987 (Morgan Stanley Equal Weighted S&P 500: VADBX). But for many, the Rydex ETF signaled a fresh start for rethinking and reinventing index design and, perhaps, asset allocation, too.
Certainly since 2003 there has been an explosion of new benchmarks with grand ambitions. That includes efforts to combine passive indexing with alternatives to the traditional methods of weighting securities by assigning stocks a share in an index according to their market value. Otherwise known as market-cap weighting, this was-and is-the old standby used for the S&P 500, Russell 3000, etc. Now, although there is a growing list of new ideas in indexing, expectations may be highest for what is known as fundamentally weighted indices. Broadly speaking, this group uses earnings, dividends and other "fundamentals" for weighting stocks.
Using one or more of the fundamental measures for designing equity indices represents a "huge paradigm shift" for indexing, wrote Wharton professor Jeremy Siegel in a widely quoted Wall Street Journal op-ed in 2006. And while the prospects for the new indices continue to be hotly debated, that hasn't stopped several high-profile academics and consultants from throwing their intellectual weight on the side of the fundamental indices. Nobel Laureate Harry Markowitz of portfolio-optimization fame has written in favor of the idea. So has Jack Treynor, one of the originators of the 40-year-old capital asset pricing model (CAPM), which is the basis for using cap-weighting as the default choice for index design.
But that was then. There's a new kid in town, and it's time to upgrade, Treynor, Markowitz and others advise. Investors who agree can choose from a growing list of products tied to equity indices that distinguish themselves by shunning market-cap weighting. Extending the fundamental indexing concept to other asset classes is reportedly also under study. As a result, revising asset allocation strategies may be the next big thing in indexing.
Meanwhile, old habits still die hard in 21st century finance. Cap-weighted equity indices remain the overwhelming preference for strategists who favor betas in money management. A half century of financial economics isn't easily overturned-although that doesn't stop some from trying.
And no one is trying harder than Robert Arnott, chairman of Research Affiliates in Pasadena, Calif. Three years ago, Arnott and two co-authors laid out their case for an equity index that weights stocks based on book value, sales, cash flow and dividends ("Fundamental Indexation," Financial Analysts Journal, March/April 2005). This foursome, the paper asserts, goes a long way in correcting the "pricing errors" embedded in cap-weighted indices. As evidence, the paper cites a back test that shows the fundamentally weighted index for U.S. stocks outperformed a comparable cap-weighted benchmark by an annualized 197 basis points for the 43 years through 2004, and with similar volatility to boot.
Reportedly, the source of the superior performance in the Fundamental Index concept comes from sidestepping cap-weighting's bias for holding overvalued growth stocks at the expense of undervalued value stocks. "Mathematically, cap weighting assuredly gives additional weight to stocks that are currently overpriced relative to their (unknowable) discounted future cash flows (the true fair value) and reduces weights in stocks that are currently trading below that true fair value," Arnott and his associates write. (A brief digression: Research Affiliates has claimed its Fundamental Index label is a registered trademark, thus the firm's preference for capitalizing the name.)
Like most new ideas in finance, this one has spawned its share of dissent and debate. Much of the criticism is less about the results than about labels-namely, that fundamental weighting is really value investing by a different name. Recent issues of the Financial Analysts Journal highlight some of the more pointed critiques, which in turn have elicited rebuttal from the defenders. (See "Fundamentally Flawed Indexing" in November/December 2007 FAJ; and "Why Fundamental Indexation Might-or Might Not-Work," as well as letters to the editor in March/April 2008 FAJ.)
The jury may still be out on this debate, but as a business, the fundamental benchmarking idea is rolling along. A mere two-and-a-half years old, the first fundamentally weighted index ETF (PowerShares FTSE RAFI US 1000, NYSE: PRF) had net assets at a tidy $850 million as of this past March, according to Morningstar Principia. Globally, roughly $20 billion in ETFs, mutual funds and other accounts track Research Affiliates' indices, which now come in a rainbow of broad market, industry, domestic and foreign equity indices offered through several product vendors. There's also an expanding list of ETFs based on other fundamentally oriented indices, including a suite of funds weighted by dividends and earnings via Wisdom Tree, which claims the aforementioned Jeremy Siegel as senior investment strategy advisor. And the new players keep coming, such as RevenueShares Investor Services, which earlier this year launched a trio of ETFs that weight stocks by revenues.
Meanwhile, Arnott says that Research Affiliates is considering an expansion of the firm's indexing methodology to bonds. Farther down the road, REITs and even commodities are possibilities, he tells Wealth Manager.
Looking down that road, one can imagine building multi-asset class portfolios exclusively with fundamentally weighted indices. In turn, that suggests moving asset allocation to the next level, so to speak. In fact, the next level has already arrived for global equity portions of asset allocation.
If the new indices live up to the billing, their use in asset allocation holds out the possibility of boosting stability and visibility in the overall portfolio relative to building portfolios with cap-weighted indices. The reasoning is tied to the basic proposition of fundamentally weighted indices and their claim of offering a smoother ride and higher returns than cap-weighting. If this enhancement proves durable, more of the rebalancing work that's now typical-and perhaps necessary-with cap-weighted indices will come pre-packaged in fundamentally weighted benchmarks.
The implied rebalancing bonus of fundamentally weighted indices may come in handy, given the evolving state of investment theory. As discussed in the May 2008 issue of Wealth Manager ("Back to the Future-Again," p. 66), the finance literature now supports the case for a relatively dynamic asset allocation. That's based on the accumulating evidence in the academic literature that expected returns are predictable to a higher degree than previously recognized. Markets go to extremes at times, which means that prospective risk premia fluctuate. Higher dividend yields, for instance, imply higher expected returns, and vice versa. The point is that allocations to the various asset classes should rise and fall in accordance with the current fundamental outlook.
The source of the higher predictability is hotly debated. Some say it's a function of market flaws or investor irrationality. Others speak of a revised efficient market hypothesis that offers a fluctuating risk premium driven by economic fundamentals. In either case, the implication for investment strategy is clear: Asset allocation should be dynamic to capitalize on the return predictability. In contrast, an older view of modern portfolio theory and EMH suggests keeping asset allocation static, based on the assumption that markets are completely unpredictable, and so expected return is constant.
Not so, financial economics now counsels. Markets are not completely predictable, of course. But expected returns are partially visible, we're told, which implies that asset allocation should be partially active. Certainly that's true for asset allocation strategies using cap-weighted indices, which are susceptible to the extremes of market conditions and investor sentiment.
But what if we're using fundamentally weighted benchmarks? These indices are designed for a smoother ride with a comparable and perhaps higher return relative to cap-weighted benchmarks. Certainly in the case of Research Affiliates' indices the intent is to provide a more efficient sampling of the economy's footprint via a diversified basket of stocks. All things equal, running asset allocation strategies with fundamentally weighted ETFs and mutual funds implies a less-active rebalancing program compared with cap-weighted products.
Arnott and two Research Affiliates associates have considered the strategic possibilities in The Fundamental Index: A Better Way To Invest (Wiley, 2008). As the graph on the previous page-from the book-illustrates, Fundamental Index strategies are said to offer three basic opportunities to "reshape return expectations" when the benchmarks replace cap-weighted indices by:
1) lowering risk without lowering expected return, or
2) raising expected return without raising risk, or
3) raising risk levels by increasing equity allocation but without increasing downside risk.
Assuming Fundamental Indexes deliver as promised, all of the above are possible because of the enhanced stability in capturing the equity market's expected return. In short, a strategic use of Fundamental Indexes implies a lesser degree of active asset allocation compared to the use of cap-weighted indices, while remaining true to current academic thinking.
It's a bit ironic that the Fundamental Index ideal revives the case for more stability in asset allocation. Recall that relatively stable, unchanging asset allocation strategies apply only under the old view of market efficiency-i.e., market prices equate with fair value. Fundamental Indexing rejects the old view of market efficiency. Yet this solution leads one back to an asset allocation approach that's closer in spirit to a world defined by a classic definition of market efficiency. In some respects, what's old is new again.
Ironic, perhaps, but not necessarily surprising. We're told that Fundamental Indexing effectively seeks to reinstate some of the pricing efficiency in the capital markets that is lost with cap-weighted indices. If true, it should be no wonder that asset allocation via Fundamental Indexing offers the potential for bringing us closer to a less-dynamic strategy than would be appropriate with cap-weighted indices.
So far, so good. But the question remains: Will fundamentally weighted indices live up to their marketing? Yes, the analysis by Arnott and others is compelling. At the same time, the history of empirical research is littered with studies that, in varying degrees, performed better on paper than in the real world. Will fundamental indexing fare any better? Unfortunately, it's still too early to say.
Just as the first experiments in indexing required time to prove their worth in the real world, so too must fundamental indexing go through economic-cycle stress testing, with actual dollars on the line. In fact, the testing is unfolding as we speak. Stay tuned.
James Picerno (jpicerno@highlinemedia.com) is senior writer at Wealth Manager.
