What's the Best Diversifier for Equity Risk?
Simple is best, according to our annual review of asset-class correlations.
In a bull market that's more than 10 years old, it's only reasonable to be thinking hard about risk controls. And the best way to diversify equity risk is to make sure you have asset classes in your portfolio that will hold their ground, or possibly even gain a bit, when stocks fall.
Our most recent run of asset-class correlations data suggests that you probably don't need to venture into arcane, expensive alternatives to build a diversified portfolio. Plain-vanilla, high-quality bonds have been among the best diversifiers for equities over the past decade, eclipsing every other asset class on that front.
Of course, a relationship that has held in the past--high-quality bonds zigging while stocks are zagging--may not hold up in the future. After all, correlations data are based on past performance, and a performance pattern that held true in one market environment may not prevail into the next. In the financial crisis, for example, categories that had performed well in previous equity-market downturns, including municipal bonds, corporate bonds, and commodities, lost value amid a global flight to quality and liquidity. Treasury bonds managed to gain value over that stretch, but little else did.
Yet, historic correlations, while imperfect, are the most reliable gauge we have when aiming to build diversified portfolios. Moreover, even as categories like municipal bonds and corporate bonds became more correlated with stocks and lost value during the financial crisis, their losses were substantially smaller than equities. Thus, it's worth considering not just the direction that an asset class moves in, but how much it loses or gains over the period examined.
By the Numbers
To examine whether correlations among asset classes have changed since I last examined the data in 2018, I turned to Morningstar Direct's software, which is geared toward institutional investors. I examined correlations among an assortment of equity, bond, and alternative investments over various time periods. My first choice was to use indexes to examine correlations, because they're a pure play on a given asset class. However, I used mutual fund categories and even a single fund (SPDR Gold Shares (GLD)) when I couldn't find an index with a sufficiently long track record.
To measure correlations among assets, Morningstar uses a statistic called "correlation coefficient." A correlation coefficient of 1 indicates that two assets are perfectly correlated--hence the statement that "all correlations went to 1" during the financial crisis. At the other extreme, a correlation coefficient of negative 1 indicates a perfect inverse relationship; if one asset goes up, the other goes down. Finally, a correlation coefficient of 0 indicates no correlation at all.
In Morningstar Direct, you can assemble what are called "correlation matrixes" that depict various assets' interrelationships. Here's a look at the one-, three-, five-, 10-, and 15-year correlation data.
The charts can be difficult to make sense of, but the basic idea is that the numbers along the rows in the horizontal axis correspond to the investment type with the same number along the columns in the vertical axis. To see the correlation between two assets, focus on the square on the grid where they intersect.
On the 10-year chart, for example, you can see that the S&P 500 and Russell 2000 indexes have had a close correlation of 0.88--about what you'd expect given that they're both equity indexes. But correlations between those two equity indexes and the Bloomberg Barclays U.S. Aggregate Bond Index drop into negative territory, and correlations between the equity indexes and Bloomberg Barclays U.S. Treasury 20+ Year Bond Index are even lower.
Separating Signal From Noise
Indeed, the negative correlation between high-quality U.S. bonds and equities was among the most reliable patterns in my survey of correlations. Over every time frame examined, the Bloomberg Barclays U.S. Aggregate Bond Index had a negative correlation with stocks, and the correlation between Treasury bonds and stocks was lower still.
At the same time, it's worth noting that the correlation between high-quality U.S. bonds and stocks has indeed been weakening in recent years; the 10-year correlations are much lower than the three- and five-year correlations, for example. That may owe in part to the fact that some periods of equity-market weakness recently, such as stocks' slump in early 2018, owed to interest-rate jitters that hurt both stock and bond prices. It's also worth pointing out that even as bonds have dramatically underperformed stocks over the past decade, bond returns haven't been a disaster by any stretch. Thus, investors shouldn't expect miracles from the asset class in an equity-market shock, especially one fueled by fears over rising interest rates.
In previous data runs, I had focused on long-term Treasury bonds to assess Treasury/equity correlations, but in this series I added the Bloomberg Barclays Treasury 5-10 Year Index. Somewhat surprising (to me) was that the short- and intermediate-term index provided nearly as much diversification as the long-term index. That's good news, because long-term Treasuries can be really volatile; short- and intermediate-term bonds, while not invulnerable to interest-rate changes, tend to be more placid and easier to own. Not surprisingly, high-yield bonds were a poor diversifier for equity exposure. Foreign bonds, as represented by the Bloomberg Barclays Global Aggregate ex-USD Index, delivered diversification that was roughly in line with what was provided by the U.S. Aggregate Index. The foreign index's correlation with U.S. bonds was high, however.
As you might expect if you're even a casual observer of global markets, all of the equity indexes, both U.S. and non-U.S., tended to have fairly high correlations with one another over all trailing periods. In other words, own small caps and foreign stocks for their return potential, but not to diversify your U.S. large caps.
I also included a sampler of different alternatives categories to help assess their efficacy as diversifiers: gold, managed futures, market neutral, and commodities. Over the past decade, their ability to diversify equity exposure isn't especially compelling. They've generally behaved less in sync with the S&P 500 than, say, the Russell 2000 or MSCI EAFE indexes, but that's not saying a lot. The least equity-correlated alternative asset type of those in my survey was SPDR Gold Shares, which invests in gold bullion. Indeed, its correlation with equities plummeted over the past year, even as bonds and equities became more correlated with one another.
In addition to examining recent trailing-period correlations, I also went back in time to examine correlations from 1999-2009 relative to what they are today. The earlier time period is instructive in that it captures much of the "lost decade" in equities, a period marked by two big bear markets for stocks. By looking at correlations during this stretch, we can see which categories came through in the clutch. (Some of the assets, such as SPDR Gold Shares, weren't around for the whole period and therefore aren't included in my grid.)
As discussed above, bonds, especially U.S. Treasuries, were negatively correlated with stocks during this stretch, albeit not to the same extent as in the 2009-19 period. And the market-neutral category exhibited an even lower correlation with equities than bonds did. (The correlation hasn't been as impressively low recently.) Such funds tend to exhibit a volatility profile that's more in line with bonds than stocks, thanks to strategies that make equally weighted bets on stocks the portfolio manager thinks will rise in value and those that will fall in value.
Introducing Morningstar's New Podcast: The Long View
Expand your investing horizons and look to the long term. Join hosts Christine Benz and Jeff Ptak each week on The Long View for wide-ranging conversations with leaders in investing, advice, and personal finance. Subscribe to and rate the podcast today, and read more about the most recent episode here.
Christine Benz does not own (actual or beneficial) shares in any of the securities mentioned above. Find out about Morningstar’s editorial policies.