Credit: What’s in a name?
Confusing terminology aside, ‘short-duration high-yield' looks like a compelling opportunity for low-volatility yield pick-up. Martin Steward assesses the risks, and underlines the importance of defining objectives
A rose by any other name would smell as sweet, said Shakespeare's Juliet.
Undaunted, the marketing industry has spent 400 years trying to prove her wrong - and has struck again with the recent rash of ‘short-duration, high-yield bond' funds.
The consensus is that rising interest rates indicate a recovering economy - which is precisely when high-yield outperforms. High-yield is credit risk; interest-rate risk contributes little, if anything, to its returns. Moreover, it is strange that marketing felt the need to sweeten short-maturity high-yield with the ‘D' word in the first place. It already smells pretty sweet in its own right.
Compared with the broader high-yield market, when you get within 2-3 years of a bond's maturity, the credit spread is much less volatile and returns come almost entirely from ‘coupon clipping'. That leaves you with a lower yield - at the moment, around 5-6% versus the broader market's 7-8%. But it does not appear to leave you with lower total returns. Bonds with 1-3 year spread duration have outperformed the broader market back to the mid-1990s. Focus on higher-quality BBs and investors have benefited from about 15 extra basis points of Sharpe ratio - perhaps because most defaults occur early in a high-yield bond's lifetime: historically default rates on single-Bs peak at about 6% at nine years to maturity and fall to 3% by one year to maturity.
But there are two key tail risks that investors may be picking up as the price for that high Sharpe ratio - extension risk, and default-pricing risk.
The high-yield universe with a 1-2 year maturity is small - about 12% of the Barclays Global High Yield index. By the time bonds get to within a few years of their legal maturity, issuers have often chosen to ‘call' them. As Phil Milburn, a high yield portfolio manager at Kames Capital, puts it: "Companies tend to refinance when they can, rather than when they have to."
There are a couple of ways around this. The first is simply to expand the maturities allowed in your strategy. Roman Gaiser, head of high yield at Pictet Asset Management (which has called its new fund ‘Short-Term High Yield'), notes that his 0-4 year euro-denominated universe exhibits very similar risk and return characteristics to the 1-3 year benchmarks, but enlarges his universe to 72 names and €55bn of bonds. Similarly, for the global market Neuberger Berman and AlianceBernstein define their new funds' universes as 0-5 years: "If you shorten much more than that, then you do have to start worrying about concentration because you rapidly narrow your universe," says AlianceBernstein's Rudolph-Shabinsky.
The second way is to expand legal maturities even further by including ‘yield-to-call' paper. As long as the market feels that an issuer is likely to refinance when the call date on a bond arrives, that bond will trade as if the call date were the legal maturity date: a 2019 bond with a 2014 first-call date, for example, might trade at a yield congruent with a three-year maturity.
By including yield-to-call paper, Covode doubles his universe to $450bn. But he can hold no more than 30% in these positions and warns of the "tail risk" of extensions. If the market sours on high yield in general, or certain issuers specifically, those issuers may postpone refinancing or be unable to refinance at all: bonds will suddenly re-price to a higher yield congruent with a later call date - or legal maturity.
How to manage this risk? Well, better credits are less likely to extend. And Tom Huggins, a portfolio manager specialising in shorter-dated high-yield at Eaton Vance suggests relying on issuers that had to issue at double-digit yields in 2008 and 2009: their current cost of financing at 6-7% is too good for them to pass-up.
"Where it gets a bit dicey is when shorter-dated portfolios start buying bonds issued over the last two years," he says. They offered 7-8% coupons and trade to a first call in 2015-16 - but will 7-8% seem expensive to their issuers in 2016? "That's not paper that I'd feel comfortable putting into a dedicated short-duration product."
Some managers have ingenious ways to get around extension risk. With some issuers that only have long-dated bonds outstanding, both Pictet and AllianceBernstein will, instead, buy a short-dated government bond and sell a short-dated CDS to replicate a short-dated credit with no call structure. Gaiser says that these synthetic credits add another 25 names to his 72-name universe and enable him to limit yield-to-call positions to 10%.
Another way to limit extension risk is to expand your universe by relaxing credit-quality standards - CCCs add perhaps 10-15% more paper. Indeed, practitioners seem to split along this trade-off between extension and credit risk. Covode, who can carry 30% in yield-to-call, has hardly anything in CCCs. "This kind of strategy needs to be bullet-proof," he says. "Most CCCs that allow maturities to creep up are not actively waiting for a better rate environment - they simply don't have any alternative."
Rudolph-Shabinsky excludes CCCs altogether. Gaiser has a 10% bucket for CCCs but a 2% limit for individual CCC issuers. Muzinich & Co, which launched a short-duration high-yield strategy in 2010, is also focused squarely on B and BB-rated credits, according to portfolio manager Dave Bowen. By contrast, Huggins, who worries most about extension risk, is positively bullish about short-dated CCCs. "Agencies assign the same rating to a company's six-month bond and its 20-year bond - but there is clearly less risk in the shorter-dated bond," he says.
The basis for this argument is twofold. It is always easier to analyse the creditworthiness of an issuer over the next 12 months than over the next 10 years. But today, in particular, issuer balance sheets are in such excellent shape that the high-yield curve has flattened substantially. Huggins says he has found two-year and 10-year bonds from the same issuer trading at equal yields. If you expect spread tightening, it might make sense to maximise exposure by heading further out on that curve. Otherwise, why take more credit spread duration risk, and more volatility, for the same yield?
This is a very compelling argument in favour of short-dated high-yield at the moment. A flatter-than-normal high-yield curve and a steeper-than-normal government curve means a higher spread in shorter-dated than longer-dated bonds - at least relative to their respective price volatilities.
Against Huggins' example of two and 10 years with the same yield, AXA Investment Managers portfolio manager Ekaterina Findlow notes that some issuers' senior secured bonds will trade at 95 cents on the dollar with a 2013 maturity but at 80 cents at 2014. Matthew Eagan, a portfolio manager at Loomis Sayles, points to extreme examples like TXU, with 2014 bonds trading in the low-70s against its long-dated bonds in the low 30s. But even without the extreme examples, the high-yield curve is still upwardly-sloping. So holders of cheaper, longer-dated bonds enjoy a bigger cushion in the event of default or credit stress, when the issuer's entire curve will equalise towards recovery value.
"A 1-3 year maturity high-yield portfolio is like picking up nickels in front of a steamroller," says Eagan.
Huggins concedes that some portfolio managers "always want to be in the lowest-priced bond", but he finds it odd that an investor would assume the credits he picks are poor. "If we come to that conclusion, we should just get out of the name altogether," he says.
It seems a fair point. If your credit analysis is any good, unexpected defaults should only come from systemic risk events. While the curve for an individual credit will invert in the event of default, even the panic of 2008 led curves to invert only in the most pro-cyclical sectors. "[Curve inversion] doesn't ever seem to have happened at the market-wide level," says Rudolph-Shabinsky. "It seems to be idiosyncratic."
Ultimately, then, your perception of the credit risk in a short-dated high-yield portfolio will depend upon your sensitivity to losses from perhaps one or two defaults. "The chances are that, at the aggregate level, the short-to-maturity funds would still outperform the unconstrained ones because the defaults would be outweighed by the pull-to-par of the bonds that don't go wrong," reasons Milburn at Kames Capital.
So, in the context of a broad high-yield strategy, the trade-off with the tail risks of extension and default certainly looks to be a good one. But is that really the way investors think about the strategy?
Bowen at Muzinich says that the demand is coming from conservative wealth management clients. Others point to similarly income-driven insurance companies. "If [government bonds] were higher yielding, many investors in short-duration high-yield might never have taken the plunge into high-yield at all," says Bowen.
If full-market high-yield is compelling as an equity-replacement strategy, short-duration high-yield is apparently regarded much more as a cash-plus income strategy. "This kind of product is not about competing with the full-universe of high-yield," as Huggins puts it. "It's about a steady stream of decent income."
Similarly, Covode describes Neuberger Berman's short-duration high-yield strategy as "getting as close to ‘sleep-at-night' risk in high-yield as possible" - no wonder he insists on a "bullet-proof" portfolio. "If a bond drops 10-20 points there is nowhere else in the portfolio where you can make up that lost performance while bonds trade well above par and yields are in the 4-8% range." This is an absolute-return mindset, where a year in which one default wipes out 3-4% of return is a genuine worst-case scenario.
But there is no rule that says investors have to think of short-duration high-yield as cash-plus.
Why not look past the marketing spin of that ‘D' word and deploy the strategy as a straightforward high-yield exposure - working on loss and volatility assumptions derived from the full market?
As we have seen, the yield give-up against full-market high-yield is modest and, historically, the return has been just as good - if not better. Deciding how you feel about the risk you are taking with the strategy is important - and will probably be crucial to the trade-off that you will want to make between extension and credit risk.