The Error-Proof Portfolio: Do These Popular Investments Add Diversification?

For some investors, one of the most disappointing aspects of the financial crisis was how a number of investments perceived as diversifiers did not deliver on their expected benefits.

Despite being widely touted for their diversification benefits in the years leading up to the crisis, commodities funds generally followed the fortunes of the equity market, with many losing more than 60% from peak to trough. Real estate? Forget about it--the typical fund also lost more than 60% from early 2007 through early 2009. International funds posted losses that were in line with, and in some cases greater than, what U.S. stocks experienced. High-quality bonds, gold bullion, and shorting stocks were among the only ways to generate a positive return during the crisis.

In a previous article, I took a look at the diversification benefits offered by various noncore asset classes--including commodities, gold bullion, high-yield bonds, and Treasury Inflation-Protected Securities. I assessed their overlap factor--that is, whether they tend to naturally occur in other core investments that are mainstays for most investors. I also took a look at their diversification factor, or their correlation with plain-vanilla stock and bond funds.

At the suggestion of a Morningstar.com reader (thank you, Thomas!), I ran the same exercise on a few other categories, focusing on some of the ones that investors have been gravitating to during the past few years: master limited partnerships, bank loans, and long-short equity funds.

In assessing diversification benefits, I subjected each asset type to a two-part test: 1) overlap factor--that is the extent to which the asset is likely to appear in diversified funds investors already own--and 2) diversification benefit, as measured by each investment's correlation coefficient relative to other widely held asset types. (We plan to incorporate correlation data into Morningstar.com in the future, but right now it's only available in Morningstar's Direct software.)

In testing the correlations, I used a combination of indexes and fund categories, generally opting for the one with the longest observable track record. Note that as with any exercise that relies on past performance, what has been correlated--or not--in the past may not hold up in the future.

Master Limited Partnerships
Overlap Factor: As interest in MLPs has skyrocketed in recent years, institutional ownership has also picked up and now accounts for roughly 25% of the sector's market capitalization. A number of exchanged-traded funds and notes focus on MLPs, and the investments have increasingly found their way into pension funds and other institutional vehicles. Nonetheless, MLPs still tend not to pop up in diversified equity mutual funds, primarily because of tax reasons. Most funds are not taxed as corporations but pass through gains to investors, who in turn owe taxes on them. But Morningstar analyst Mike Rawson explains that once a fund owns more than 25% of MLPs, the fund itself loses its tax-exempt status. Morningstar analyst Abby Woodham says that these tax issues make it unlikely that MLPs would find their way into broad market indexes, and she also notes that the S&P 500 specifically disallows limited partnerships. (She says that a handful of pipeline names, such as  Kinder Morgan Inc. (KMI), do appear in the index because they're not structured as limited partnerships.) The fact that most equity mutual funds avoid MLPs means that an investor could add a position in individual MLPs or in an MLP exchange-traded fund without creating substantial overlap.

Diversification Factor: MLPs also tend not to be highly correlated with other core categories, especially bonds. During the past decade, the Alerian MLP Index has a correlation of 0.50 with a broad U.S. stock market index and a slightly negative correlation (-0.07) with the Barclays Aggregate Bond Index. (A correlation of 1 means that two investments move in lockstep, whereas a correlation of -1 means that two investments move in opposite directions.) Curiously, the Barclays Aggregate Index's correlation with MLPs' during the past decade was even lower than its correlation with the total U.S. market. MLPs' correlation of 0.69 with energy stocks is slightly higher than its correlation with the broad U.S. equity market but lower relative to commodities-tracking investments. On a hunch, I also tested MLPs' correlation with a handful of broadly diversified dividend-paying stock ETFs and was surprised to see it was even lower than MLPs' correlation with the U.S. equity market.

Investors have gravitated to MLPs during the past five years, primarily because of their rich yields in a low-yield environment, but it appears the securities have some diversifying properties, as well. That said, it would be a mistake to expect strong defensive properties during equity market downturns. In 2008, for example, many MLPs had losses in line with the broad equity market.

Bank Loans
Overlap Factor: Like MLPs, bank loans have, to date, had limited penetration in broadly diversified funds. The average short- and intermediate-term bond fund has but a tiny take in the sector, and the typical high-yield and multisector bond fund has only slightly more. (A handful of nontraditional bond funds have more meaningful positions in the sector, says Morningstar senior fund analyst Eric Jacobson.) That means that the legions of investors buying these funds lately aren't likely to have their holdings overlap with their existing high-quality bond funds.

Diversification Factor: Bank-loan funds' correlation with high-quality bond funds is also pretty low. That's perhaps not surprising when you consider that bank loans are typically beneficiaries when rates rise (their yields tick up to keep pace with LIBOR), whereas high-quality bond funds get hurt. During the past decade, bank-loan funds have exhibited a slightly negative correlation with the Barclays Aggregate Index and an even lower correlation (-0.35) with long-term Treasuries. The 10-year correlation with short-term bonds is higher (0.57) and higher still for equities (0.61) and high-yield bonds (0.87).

Thus, even though bank-loan investments may help mitigate the pain in a rising-rate environment, investors expecting these funds to provide ballast in an equity market shock might not get it here. In 2008, the typical bank-loan fund lost 30% of its value, though higher-quality offerings such as  Fidelity Floating Rate High Income (FFRHX) held up substantially better.

Long-Short Equity
Overlap Factor: Most investors already have long equity exposure in their portfolios, but they're unlikely to have any short equity exposure unless they actively seek it out via a long-short or bear market fund. (Some tactical asset-allocation funds also employ shorts--for example,  PIMCO All Asset All Authority (PAUAX).) Although most plain-vanilla equity mutual funds are allowed by charter to employ shorting strategies--essentially, to bet against stocks--in practice very few do. Most have prospectus objectives that limit the extent to which they can short. Further stymieing shorting activity is that for fund managers, there's enormous career risk in getting caught betting against stocks in an up market.

At the same time, it's worth noting that long-short equity funds vary widely in the extent to which they take long positions and employ shorts. Morningstar's Phil Guziec notes that some long-short equity funds keep their long equity exposure quite low, at about 30%, while others have up to 80% net long exposure. That makes it tough to generalize about the category's overlap factor.

Diversification Factor: Perhaps not surprisingly, given that some long-short equity funds have such high net equity exposure, the long-short equity category looks underwhelming when it comes to diversifying an equity portfolio. The correlation with the broad U.S. stock market and international equities was 0.93 during the past decade. Not surprisingly, the correlation with the U.S. bond market is much lower and even negative relative to long-term Treasuries. But U.S. stocks also have a negative correlation with long Treasuries. Morningstar has a handful of highly rated funds in this category, including  Gateway (GATEX) and  Wasatch Long/Short Investor ; their specific strategies and managers, rather than their diversifying properties, are the biggest factor driving those recommendations.

See More Articles by Christine Benz