Rumors of buy-and-hold's death are premature but persistent. If you Google the phrase, you'll find a long list of stories and books through the years that proclaim that buying and holding is an investment idea that's passed to the afterworld of defunct strategies. But a funny thing happened on the way to the funeral: the corpse survived. In fact, he's doing quite well, to judge by the performance numbers.
Yet the obituaries keep coming. The latest arrives via MIT professor Andrew Lo, who tells Money Magazine:
Buy-and-hold doesn't work anymore. The volatility is too significant. Almost any asset can suddenly become much more risky. Buying into a mutual fund and holding it for 10 years is no longer going to deliver the same kind of expected return that we saw over the course of the last seven decades, simply because of the nature of financial markets and how complex it's gotten.
The conceit here is that the future will be unusually harsh on betas, but alpha will be unaffected and perhaps even shine brighter. The interview with Lo doesn't convincingly explain why this scenario is likely to unfold. Meanwhile, history gives us quite a bit of evidence to remain skeptical that the future will be much different when it comes to the relationship between beta and alpha.
This much is clear: buying and holding a portfolio that passively owns all the major asset classes, each initially weighted by market values, has generated competitive returns in recent history. For example, consider how the Global Market Index (GMI) has fared for the 10 years through the end of 2011. This unmanaged market-value weighted index of the major asset classes earned a 6.0% annualized total return for the past decade. Should we be impressed by that result? Judge for yourself by considering the performance histories of 1,201 actively managed, multi-asset class mutual funds with at least 10 years of history, as illustrated in the chart below.
GMI's track record rises to the 89th percentile in this group for the past decade. In other words, a buy-and-hold strategy that's comprised of all the major asset classes earned strong above-average results relative to the actively managed competition. Keep in mind that you can replicate GMI with ETFs for under 50 basis points, or about half as much as what many actively managed funds charge. In fact, quite a few actively managed multi-asset class funds even charge three times more than the investable version of GMI.
There's reason to wonder if GMI will deliver equally strong performance results over the next 10 years. In fact, I expect that its relative performance will fade a bit. That said, there's no reason to think that GMI won't deliver average to moderately above-average results in the years ahead. To expect otherwise requires some rather radical assumptions.
One of the reasons that GMI has earned such a handsome relative return vs. active managers is that the future's uncertain. Predicting winners and losers, whether it's individual securities or asset classes, is tough—really tough. We know this to be true because mediocrity is widespread in actively managed funds. The chart above is just the tip of this empirical iceberg. Maybe this will change in the years ahead, but making a persuasive argument along these lines relies heavily on predictions, starting with the idea that more active managers will do a better job at overcoming the uncertainty that otherwise harasses the rest of the crowd. Good luck with that.
If the future will suffer higher volatility, as Lo predicts, it's not clear how this will translate into dramatically stronger performance records for active managers. Predicting is tough enough in periods of relative calm. Will forecasting get any easier if markets are headed for rougher seas? Unlikely. Actually, you can just as easily make the opposite case: greater volatility will enhance the relative allure of a buy-and-hold strategy that holds everything. If prices are going to bounce around a lot more, and markets are becoming increasingly complex, as Lo says, it's not obvious that the job of active management is destined to get easier.
But let's think about what's required to improve a manager's relative ranking vis-à-vis GMI. In other words, what would we have to do to earn a substantially higher return vs. a buy-and-hold strategy built on holding all the major asset classes? One possibility is rethinking the asset allocation. An extreme example: choose the asset class (or group of asset classes) with the highest expected return and shun (or short) those that look poised for trouble.
Another possibility: trade in and out of asset classes with a fair amount of frequency in an effort to time the markets. Yet another strategy is picking the most-promising securities within the asset classes while avoiding those that appear to have relatively dark futures. The problem, of course, is that few investors can take advantage of these opportunities in relative terms because of a simple fact: alpha sums to zero. For every investor who earns above-average performance, there is someone who lags the benchmark. In other words, the winners are financed by the losers. Meanwhile, the market average—the benchmark—is forever in the middle. After adjusting for expenses, the impact of trading, and taxes, average results tend to become above-average results through time. That's a key reason why GMI has earned competitive returns in the past, and it's a foundational argument for thinking that GMI will continue to earn above-average results in the future.
But how, then, does one square that view with all the research that tells us that markets aren't perfectly efficient? Doesn't that open the door to earning benchmark-beating returns? Yes, although quite a bit of these opportunities have little if anything to do with assuming that buy-and-hold investing is dead.
A quick example is everyone's favorite poster boy for earning alpha: Warren Buffett. But think about how Buffett has earned alpha through the decades. It's not from trading. Rather, he's a buy-and-hold man, right down the line, albeit with a twist. Buffett is a buy-and-hold investor with a preference for what financial economists call return anomalies. In particular, Buffett favors the value anomaly--companies trading at a discount to market value, but he's succeeded by emphasizing that anomaly in a buy-and-hold framework.
In fact, there are lots of documented return anomalies, as outlined by Antti Ilmanen in his magnificent book Expected Returns: An Investor's Guide to Harvesting Market Rewards. What's in short supply, however, is clear, verifiable evidence that more investors are exploiting these anomalies on a sustained basis. Why? Alpha still sums to zero, even if you're conversant in the finer points of market anomalies.
To the extent that you can earn market-beating returns, shunning buy and hold isn't an absolute requirement. True, some anomalies require trading, but many don't. Meanwhile, history counsels that we should be cautious in thinking that earning alpha is going to be any easier in the future than it has been in the past. Buy and hold isn't dead. As usual, however, it continues to be underestimated.