The original “fundamental indexing” ETF - RAFI US 1000 ETF (PRF) - recently celebrated its fifth birthday and there was reason to celebrate. The fund, which was launched in December 2005, posted a tidy premium in its first five years vs. its competitors: The likes of the S&P 500 and Russell 1000. Fundamental indexing, as practiced by Rob Arnott’s benchmarking designs at Research Affiliates, seems to be working.
For the five years through the end of last year, PRF earned a 4.3% annualized total return. That’s nearly double the performance of the S&P 500 (2.3%) and Russell 1000 (2.6%) over that period. The question, of course, is whether there’s reason to think that PRF’s premium is durable - or just a random draw that happened to come up a winner?
There’s a case for arguing that there’s something more than luck at work here, but not necessarily for the reasons that receive most of the attention in the press with fundamental indexing. Research Affiliates Fundamental Index (RAFI) is a process for weighting securities based on their economic footprint, which the firm defines in a specific way. The RAFI benchmarks contrast with market capitalization weighting, the conventional approach for indexing that holds securities in accordance with their market value. Analysts have been debating for several years if these indices deliver superior risk-adjusted returns relative to the traditional cap-weighted benchmarks. PRF's results suggest that they do.
Why? A better index, is one response. The RAFI design exploits some widely known weaknesses inherent in capitalization-weighted indices. If you build a better index, you'll reap the rewards. Another explanation is that RAFI has a value bias. Since value tends to beat the market over time, RAFI is simply piggybacking on this risk premium. Those advantages have merit, but they're only a partial answer.
The third factor, and arguably the main one, is rebalancing. "Rebalancing is the heart and soul of Fundamental Indexing—it’s what makes it work," Arnott told me earlier this month for a Q&A in The Beta Investment Report.
Even so, quite a bit of the wider discussion over the years about what makes RAFI tick has either focused on its alternative weighting design or on the fact that the value factor offers a tailwind. Relevant topics, to be sure, although rebalancing seems to have been marginalized in the conversation.
It remains to be seen, of course, if the years ahead will continue to treat PRF and its sibling benchmarks kindly relative to market-cap indexing. Meantime, the results to date speak clearly for the original RAFI fund. To be fair, PRF’s success hit a rough patch for a time. During the crushing losses of 2008, the ETF’s 40% retreat exceeded the cap-weighted Russell 1000’s 37.6% loss for the calendar year. But FTSE RAFI US 1000 has since regained its edge, as the chart below shows.
click to enlarge image
Should investors expect the RAFI edge to endure? Yes, assuming that you also expect that a rebalancing bonus will prevail. History suggests no less, at least over the long run. Why? It all boils down to the persistence with reversion to the mean in asset pricing.
That’s hardly a new or radical idea. The literature on rebalancing may be slim compared to other topics in finance, but what has been published generally supports the case for thinking that there’s a premium linked with routinely trimming the influence of winners and raising the allocation in losers. Yes, this is hazardous work when it comes to a handful of individual securities. But the risk diminishes considerably when systematically applied across a broadly defined asset class.
Rebalancing’s virtues are no less conspicuous in managing asset allocation. Arnott agrees, and he speaks from experience, having penned a number of studies on asset allocation over the years. He’s also a veteran of managing multi-asset class portfolios. In addition to overseeing all things related to RAFI, he supervises several asset allocation products, including the PIMCO All Asset Fund (PASDX). And, yes, the track record is strong, relative to multi-asset class funds generally.
How much excess return can you expect from a simple rebalancing strategy when applied to a broad mix of asset classes? Roughly 100 basis points over time, Arnott says. That
more or less echoes results in a number of studies. It’s also in the same neighborhood of the premium dispatched by the naïve year-end rebalanced version of our proprietary Global Market Index, an unmanaged, market-value weighted benchmark of all the major asset classes.
Is rebalancing guaranteed to generate excess returns? No, but no other strategy rises to that ideal either. Let’s also point out that rebalancing will pinch you at times. In addition, some analysts warn that rebalancing is less about minting alpha and more about lowering risk without giving up much, if any, expected return. In the end, much depends on how you define and implement rebalancing, how a portfolio’s structured, and the time period under scrutiny for analyzing historical results.
Don’t confuse the benefits of rebalancing with a free lunch. In order to participate in this bounty, you have to be comfortable with a contrarian mindset. Easy to say, tough to do. Expected return fluctuates, sometimes radically. That’s obvious in hindsight, of course, and it inspires promises to exploit the next bout of volatility. The challenge is quite a bit harder when markets actually dive or soar, at which point acting on the assumption that the ex ante rebalancing bonus will persevere is subject to a fair amount of second guessing.
There’s always fresh doubt in real time. But the crowd’s reluctance to exploit mean reversion in a timely and efficient manner, if at all, is a big part of the reason for why there’s excess-return gold in these hills, as one recent study reminds.
Meantime, quite a few of the rules-based strategies that are being securitized through ETFs these days owe some degree (perhaps a large degree) of their expected alpha to rebalancing of one form or another, either at the individual security or asset allocation level. That point isn’t always acknowledged, although Arnott’s not shy about discussing the leading source of RAFI’s alpha. Rebalancing "is a factor that is relentlessly rewarded for the long-term investor."
That's not surprising once you step back and consider the broader linkage between macro and financial markets. There's a growing body empirical research that connects the economic cycle with risk premia. Some recent examples include "Gains from Active Bond Portfolio Management Strategies" and "Business Cycle Variation in the Risk-Return Trade-Off."
The message in these and many other studies is that expected return fluctuates and that investors should be willing to adjust asset allocation at least modestly to take advantage of these fluctuations. Buy and hold can work, of course, although it may take too long for most investors. In addition, buy-and-hold strategies may not be as safe in the long run as once thought.
"Conventional wisdom views stock returns as less volatile over longer investment horizons," professors Lubos Pastor and Robert Stambaugh write in a recent study, "Are Stocks Really Less Volatile in the Long Run?" But hazards to your investment portfolio may actually rise over longer horizons, they demonstrate.
A bit of common sense helps illustrates the point as it relates to the challenges of trying to look far into the future. Imagine an investor in December 1988 who dutifully invested with a buy-and-hold strategy for 20 years. He did all the right things, but there was one fatal flaw: he needed the money exactly 20 years later, in December 2008, when financial markets were collapsing.
That's a simple and perhaps impractical example, of course, but it highlights the dangers of expecting the long run to always bail you out. The longer out your investment horizon, the greater the so-called parameter risk. Or, as some call it, the Black Swan risk.
Whatever you call it, that risk is out there. A robust risk-management strategy is needed to tame the darker side of uncertainty, along with diversifying within and across asset classes. For many investors (and even some sophisticated ETF products), a simple rebalancing strategy will do the job.