Passive Asset Allocation Strategies Are Still Tough To Beat

Brett Arends of SmartMoney skewers the simple stock/bond balanced fund strategy, and rightly so. There’s no reason to rely on a basic equity/fixed income mix in a world where a wider array of asset classes are available through low-cost ETFs.

Recognizing that the average investor has a rich menu of betas at his disposal brings us to the critical issues for portfolio management: a) choosing asset classes; and b) managing the mix through time. There’s plenty to say on these topics (see my book Dynamic Asset Allocation, for instance), although it’s useful to start by considering a good benchmark, which is the inspiration on these pages for the Global Market Index. This index is a passive allocation to all the major asset classes and it comes in three primary flavors:
1) unmanaged, market-value-weighted (GMI)
2) a year-end rebalanced version of the market-valued weighted index (GMI-R)
3) an equally weighted allocation to the major asset classes that’s rebalanced back to equal weights at the end of every year (GMI-E)
One reason why passively owning everything is a strong benchmark is that it makes no assumptions about what’s hot or what’s not. As such, it’s a fully transparent strategy that requires no skills or talent. The straight GMI, in fact, requires no decisions at all once it’s initially launched. Market-value weighting takes care of itself. But if you’re skeptical of the underlying theory for such a system—a.k.a. the capital asset pricing model—you might consider a model-free allocation to the major asset classes: GMI-E. In order to keep the weights equal, periodic rebalancing is necessary; in this case, I arbitrarily chose the end of each calendar year. As for the rebalanced version of the market-value weighted index (GMI-R), it’s something of a hybrid of the other two.
How have these indices performed? Rather well, at least in comparison with a broad sampling of actively managed asset allocation mutual funds. In my previous update in January, GMI was a strong competitor. In fact, GMI has continued to outperform nearly 90% of roughly 1,200 funds for the 10 years through April 30, 2012 (based on funds with at least 10 years of history through the end of last month via Morningstar Principia software). The rebalanced version of GMI fared even better, and the equal-weighted strategy did better still, as you can see in the chart below.

Here’s the 10-year annualized total return figures for the chart above:
Does GMI’s strong relative performance mean that everyone should abandon their asset allocation strategy and go passive? No, probably not, although you should think twice before straying too far from the GMI allocations. In any case, there are other factors to consider for building portfolios, such as your investment horizon and how your finances and career differ from the average investor’s profile.
Meantime, the simple GMI strategies tell us once again that mindless diversification across asset classes and basic rebalancing captures quite a lot of what’s otherwise billed as enlightened investing techniques. You can pay a lot more for investment advice, or spend a lot more time analyzing expected return and risk. Assuming you’ll do a lot better, however, is probably asking too much for most of us.

6 thoughts on “Passive Asset Allocation Strategies Are Still Tough To Beat

  1. JP

    It is if Israelson’s strategy is a) holding a broad mix of asset classes and b) maintaining equal weights in those asset classes. We can debate the list of asset classes to hold, including how to define them. We can also debate how often to rebalance the mix. But it all boils down to broad asset allocation and rebalancing. Those are the two critical factors. Beyond that, decision details are more dependent on forecasts.

  2. saunderscc

    The list of major asset classes includes the following number of indices: 1 REIT, 1 Commodity, 1 Cash, 3 Equity, and 7 Bond, for a total of 13 positions.
    Equally weighted, the equity positions represent ~23% of the equal-weight portfolio mix.
    If you add to the equity allocation both the commodity and the REIT exposures, this still only represents ~38% of the equal-weight portfolio mix.
    The GMI “broadly diversified” equal-weight passive strategy has all the makings of a “hot dot” in so far as fixed income investments have broadly outperformed equity investments during the new millennium and they absolutely dominate the portfolio mix.
    The numbers are attractive, but without any further review may set investors up for a nasty surprise.
    Just saying.

  3. JP

    For the record, cash isn’t included in the three main GMI indices used in the chart. However, you are correct: the way the major asset classes are defined translates into a bond-heavy allocation for the equal-weighted mix. By contrast, the market-value weighted portfolio is equity heavy. That said, some of the bonds defined here are equity like: high yield and emerging market bonds, in particular. Nonetheless, your point is well taken. There’s plenty of room for debate about deciding if the past will remain prologue and how to initially allocate a portfolio. Of course, that’s one more reason to rebalance the mix. Assuming a portfolio is broadly defined, I’m highly confident that the result will remain competitive with a broad sampling of the actively managed competition via periodic rebalancing over time.

  4. Henk

    Why only the last 10 years? If you want to make the case you should go back to the 1930’s and show how this has worked against some of the well known equity and bond indices. This reminds me very much of a back tested period. And I have never seen a bad backtesting.

  5. JP

    Why only 10 years? Well, it’s a round number and a decade, while hardly definitive, isn’t chopped liver either. The problem is that the sample of actively managed asset allocation funds with, say, 20 or 30 years fades quickly. You could, of course, look at, say, simple balanced funds, but even that’s a small pool of funds. By contrast, the last 10 years offers 1,000 plus multi-asset class funds with verifiable records. If nothing else, it’s a starting point. That and 50 years of financial economics research drops a few clues, IMHO.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>