Synopsis: A simple rule derived from the definition of distressed debt has consistently generated alpha for tactical asset allocators in past high-yield recoveries from market lows.
The challenge: When to pull the trigger?
The large variance in short-run returns among asset classes during major market turns produces huge payoffs for successful tactical asset allocation. Consider, for example, what a fixed-income manager could have achieved following the 2001 recession by astutely varying the relative weights of investment-grade and high-yield corporates. In the second quarter of 2002, the investment grade BofA Merrill Lynch US Corporate Index trounced the BofA Merrill Lynch US High Yield Index, 4.37% to -6.98%. One year later, the tables turned, with the HY index crushing the IG index, 10.02% to 2.47%, in the second quarter of 2003.
Given these rewards, why doesn’t every money manager grab oodles of alpha by reallocating during major market turns? The problem is that it requires calling the turn. Classically, the relatively undervalued asset may get more undervalued before the market comes to its senses. Jumping in too early can be disastrous for managers who do not have the luxury of being evaluated only on a multiyear basis. For those trying to put up good quarterly numbers, it is not good enough to say, “I am confident that in the fullness of time the market will recognize the wisdom of my positioning, even if I suffer a year or two of underperformance in the interim.”
Here is the good news: A startlingly simple timing signal has consistently paid off over a three-month horizon in recoveries from high-yield cyclical bottoms. Recognize upfront that the methodology does not guarantee capturing the cyclical rebound’s highest three-month return and the historical sample size is small. With those caveats, the trading strategy is a bona fide no-brainer and has reliably generated very substantial alpha.
Breaking through 1,000
The easy-to-remember rule for deciding when to step up high-yield exposure is to pull the trigger the day after the option-adjusted spread (OAS) on the BofAML High Yield Index falls below 1,000 bps.
High-yield cognoscenti will recognize the 1,000 bps level as the threshold for defining distressed bonds, a standard I introduced some 25 years ago. At an OAS of 1,000 bps, the market is essentially saying that the high-yield asset class as a whole is distressed. When the spread moves out of that territory, the market is signaling that financial distress is receding.
To test this trading rule, I measured three-month returns on the BofAML High Yield Index, the BofA Merrill Lynch US Non-Distressed High Yield Index, the BofA Merrill Lynch US Distressed High Yield Index, the BofA Merrill Lynch US Treasury & Agency Index, and the investment-grade BofAML Corporate Index. I started each trial on the first trading day after the spread first fell below 1,000 bps. Future users of this strategy do not have to predict anything or exercise any judgment, but can instead wait to act until the decisive piece of information has arrived.
For the period preceding OAS availability on the BofAML High Yield Index, that is, prior to Dec. 31, 1996, I used the yield-to-maturity (YTM) difference between the BofAML High Yield Index and the BofA Treasury & Agency Index. (Yield-to-worst figures are also unavailable prior to Dec. 31, 1996.) This substitution is justified by a comparison of the YTM spread and OAS during the period in which both are available. From Dec. 31, 1996 to Dec. 31, 2014, the median monthly difference between the two versions of the spread, when OAS was in a range of 900-1,000 bps, was – remarkably enough – one basis point. (The YTM version was higher by that amount.) As a point of interest, the gap increased as OAS declined.
Note, in addition, that the inception date of the BofAML Non-Distressed High Yield Index and the BofAML Distressed High Yield Index is, you guessed it, Dec. 31, 1996, so returns are not available on those indexes for the first two trials depicted in the table. Finally, the high-yield market had just one cyclical low in the early 1990s, but the high-yield spread exited the distressed zone twice. After falling below 1,000 bps on Dec. 24, 1990, the spread later rose above that threshold before definitively dropping below it on Feb. 12, 1991. (The first trial in the table commences after the market’s closure for Christmas on Dec. 25, 1990.)
In all five trials high-yield outperformed governments by at least 5.10 percentage points and in the most extreme case, by 13.28 percentage points. Note that these are non-annualized, three-month returns. On an annualized basis, the high-yield returns during the five episodes documented here ranged from 30.32-78.99%. High-yield similarly outperformed IG corporates every single time, by margins ranging from 2.75-12.13 percentage points.
In the three trials in which returns are broken out by distressed and non-distressed, dynamic asset allocators following the 1,000 bps rule and achieving average results would have beaten governments by 2.51 to 4.90 percentage points without needing to own a single distressed issue. The non-distressed high-yield performance edge over IG corporates was likewise substantial in both 2001 and 2009. Naturally, coming off the bottom, the distressed component of the high-yield market turned in exceptionally high returns, ranging from 63.29-174.40% in annualized terms.
Not engineered to catch the absolute bottom
Based on an admittedly limited historical record, the simple 1,000 bps rule generates a substantial three-month performance bonus for tactical asset allocators. Adopters of the trading rule should be forewarned, however, that I have not solved the problem of precisely calling the bottom. Indeed, the high-yield OAS may narrow by hundreds of basis points, producing sizzling high-yield returns, before it finally cracks the 1,000 bps barrier.
The table below compares the high-yield returns captured by the 1,000 bps rule during the high-yield market’s four cyclical recoveries with the peak quarterly returns observed in those cycles. In 2001 the 1,000 bps rule actually produced a return slightly higher than that of the best quarter of the cyclical rebound. In the others, an asset allocator astute enough to have entered at the start of the peak quarter did materially better than followers of the 1,000 bps rule. For mere mortals, however, adding hundreds of basis points over comparable-period returns on high-quality bonds is an outstanding achievement.
Marty Fridson, Chief Investment Officer of Lehmann Livian Fridson Advisors LLC, is a contributing analyst to S&P Capital IQ. His weekly leveraged finance commentary appears exclusively on S&P Capital IQ LCD. Marty can be reached at firstname.lastname@example.org.
Research assistance by Yueying Tang and Zizhen Wang.