Barclays first sought regulatory approval for the security in the summer of 2003. Nearly four years later, the shares finally started trading. What took so long? “Because this is the first ETF offering exposure to the high- yield market, there was a greater level of scrutiny than there might be for other products,” says Matthew Tucker, head of investment solutions at Barclays Global Investors.
The ETF is the first junk bond index fund of any kind in the U.S. market. Even among the dozens of mutual funds plying the lower-grade debt market, no publicly traded portfolio formally tracks an index. As such, the new iBoxx ETF fills a long-running hole for the indexing of a major asset class.
As indexing migrates into more exotic corners of the capital markets, the subject of passive management is debated anew. Hedge fund and managed futures strategies, for instance, are now being indexed by some ambitious shops.
If passive management can be applied to alternative investing, surely junk bonds can be indexed. In theory, yes, but in practice, it’s a bit messy.
The new iBoxx ETF is effectively an experiment that can only be judged over time, along with other efforts to bring indexing to areas of finance that have formerly been the domain of active management. Clearly, the market for high-yield bonds poses one of the bigger challenges to date for ETFs. Bonds rated less-than investment grade - below the BBB credit rank but not in default - are still debt securities, of course. But this is where any similarities with Treasuries, and investment-grade corporates, end.
The distinction is a source for both additional return and risk. Finding a happy medium is an untested challenge for junk bond indexing. There are two main incentives to try. One, high-yield offers higher expected returns over the long haul relative to investment-grade bonds. Two, the asset class tends to move independently of higher-grade debt and other asset classes, thereby offering a diversification bonus.
Consider the low correlation that junk bonds have relative to the investment-grade world. For the five years through this past March 31, the Credit Suisse High Yield Index posted a correlation of only 0.18 with the Lehman Brothers Aggregate Bond Index,a broad measure of investment-grade U.S. bonds. (1.0 is perfect positive correlation, 0.0 is no correlation.) Junk bonds also offer diversification benefits relative to stocks. During the five years through March, CS High Yield’s correlation with the S&P 500 was a modest 0.51.
But wait, there’s more. The case for a strategic holding of high yield bonds becomes even stronger when you compare the asset class’s correlations to other asset classes, including REITs, commodities and foreign stocks (see table below). What’s more, junk’s performance has been competitive.
History, in short, suggests that high-yield bonds are worth owning. Unfortunately, the history is drawn from an index that - while widely cited and respected - isn’t investable per se. That should give thoughtful investors pause because high-yield bonds are relatively illiquid. Although some debt in this realm trades with a degree of frequency, much of it lies dormant for days, weeks or longer.
The point is that CS High Yield and other benchmarks that are familiar in the institutional world are hard to trade when it comes to satisfying the minute-by-minute, tick-by-tick liquidity that’s standard for running ETFs. The mechanics of the listed funds include their so-called creation and redemption feature.
New shares are created and liquidated upon demand. That serves to immunize an ETF from the closed-end fund disease: Deviating too far from the underlying net asset value. Meanwhile, maintaining ETF liquidity, a key selling point, means that the market makers must be able to hedge themselves quickly and efficiently when new money flows in or out. The contingency demands a liquid index underlying the ETF.
Considering the hurdle, what can investors expect from the first high-yield bond index fund? A bit less than junk bond indices might suggest. Consider how the iBoxx $ High Yield Corporate Index (the benchmark for the new ETF) compares with Credit Suisse High Yield.
For the eight years through this past March,for instance, the iBoxx benchmark’s 4.7 percent annualized total return trails CS High Yield’s 7.4 percent by more than a little. The disparity narrows for the last three years through March, but iBoxx still lags CS High Yield by over 100 basis points a year.
Why the difference? In essence, that’s the price for moving from a non-investable index to one that is investable. “One of the real challenges with the high-yield market is that a lot of the indices aren’t investable,” says Tucker. “They often contain1,000, maybe 1,500 securities. At any given time, a few hundred of those bonds might actually trade in the market. But the rest really aren’t available to investors.”
CS High Yield is a broad measure of the high-yield universe,holding more than 1,000 securities that encompass a diverse mix across the spectrum of lower-grade credit risk. Some of the bonds in CS High Yield may even be close to default. In sharp contrast, the iBoxx index holds only 50 bonds - 50 securities that are among the most liquid junk bonds, which generally means relatively higher-rated securities.
By focusing on higher-rated bonds with higher liquidity, the iBoxx index may sacrifice some expected return relative to the non-investable indices that claim a broader universe. Add to that the 50-basis-point expense ratio charged by Barclays iBoxx $ High Yield Corporate Bond ETF.
The price tag associated with going investable is hardly a surprise. Lower risk tends to translate into lower returns. It’s true for equities and investment-grade bonds, and it prevails in high yield securities, too.
But there’s no mystery as to why: “Triple-C bonds default at a higher rate the single-Bs; single-Bs default at a higher rate than double-Bs,” says Martin Fridson, president of FridsonVision LLC, a New York research shop focused on high-yield bonds.
In exchange for the higher default rate, investors demand higher returns. A diversified portfolio of high-yield bonds that holds more lower-rated securities has a higher expected return, albeit with higher risk. Modern portfolio theory teaches no less, and recent history seems to confirm it.
For example, for the year through the end of this past March, the BB-rated bonds in the Citigroup High Yield Index posted a 9.3 percent return. The CCC-rated bonds in the same index rose by more than twice as much: 19.4 percent over the same 12 months. Of course, when the market turns, the opposite will likely be true as well - namely, bigger losses for bigger risks.
Barclays didn’t have much of a choice other than to use an index like the iBoxx benchmark, which targets the more liquid issues. Barclays’ Tucker notes that the iBoxx $ High Yield Corporate Index has a “built-in liquidity screen.”
The Frankfurt, Germany-based International Index Co. (NYSE:IIC) maintains the iBoxx index. At the end of each month, IIC re-balances, favoring the 50 most liquid bonds, each with a roughly 2 percent weighting.
That raises another question: Are 50 bonds enough to maintain a diversified, representative portfolio in a varied universe of debt with thousands of issues? Barclays thinks so. Tucker says that the index holds securities from different industries to avoid concentration in any one corner of the economy. On top of that, the most liquid securities are chosen to represent each industry.
The bottom line: The iBoxx is designed to be a liquid measure that reflects the broader high-yield universe, with weights in industries that roughly approximate the weightings in the overall marketplace.
Still, some observers are skeptical. Sonya Morris, editor of Morningstar’s ETF Investor newsletter, says the new high-yield ETF “isn’t very diversified.” She also opines that the relative inefficiency in the junk bond market gives active managers an edge over passive strategies.
In fact, of the 117 high-yield mutual funds with track records of five years or more through the end of this past March, the range of annualized five-year total returns ranged from a high of 16.2 percent down to 4.7 percent. The iBoxx index fares near the low end of that range, posting a 6.3 percent annualized return.
On the other hand, there’s no doubt that the iBoxx index has captured a large chunk of the market profile for high-yield bonds, albeit with lesser performance. The correlation of monthly returns between iBoxx and CS High Yield is a high 0.89, based on monthly total returns from January 1999, through this past March.
While investors may wonder what’s held in the CS High Yield and other benchmarks, Barclays should be applauded for bringing a high degree of transparency to its junk bond ETF. All 50 bonds, including weights, yields and other information, are published at the end of each trading day on iShares.com.
Whatever the caveats, wealth managers are intrigued by the first high-yield index fund. “I wouldn’t be afraid of using it,” says Matt Forester, a portfolio manager at Cumberland Investors, a Vineland, N.J. firm that manages money for high-net-worth clients.
Herb Morgan, chief investment officer of ETFAccount.com, is watching the new fund as well, although he’s in no rush to buy, noting that he first wants to see trading depth. “I really hope the liquidity comes [with the ETF] because high yield’s a great asset class.”
The biggest obstacle may be the asset class itself, which has been in a bull market for several years. As a result, some are wary of junk’s prospects for the foreseeable future. One reason is the shrinking yield premium. For example, iBoxx’s trailing 12-month yield was 7.5 percent at the end of March, or about 290 basis points over the 10-year Treasury yield. That’s down from a nearly 900-basis-point spread over the 10-year at one point in August, 2002.
In other words, the timing issue related to high-yield bond investing worries many wealth managers. There are some things that even financial engineering, and innovation on Wall Street can’t conquer.