Monday, November 24, 2008

High-yield bonds are a steal

I suggested a few weeks ago that the big decline in short-maturity swap spreads was a good leading indicator of a coming rally in high-yield (i.e., junk) bonds. So far that prediction hasn't born fruit, as this chart shows. But I also noted that sometimes the lag between swap spreads and other spread assets can be significant. It may well be the case that junk bonds are just going to be slow to react to what I view as a significant improvement in the underlying fundamentals.

Be that as it may, with junk bond yields now well over 20% (Lehman says the average junk bond yielded 22% last Friday, whereas Bloomberg says it was 25%), it's time to look at these bonds in isolation. Junk bond yields have never before been this high. Just what does that mean for an investor?

According to Moody's, the current default rate for junk bonds is about 4%. They expect it could rise to 10% by the end of next year, which is close to the all-time high cumulative default rate in the Depression of 14%. Let's say their decidedly pessimistic assumptions are right. Now, considering that defaulted junk bonds are likely to have a recovery value of at least 30% (which is very conservative), that means that an investment in junk bonds today is likely to incur losses of 10% * 70%, or 7% over the next year. In other words, you would lose 10% to defaults, but recover 3% of that as the defaulted companies liquidate their assets. So if you buy junk bonds today that are yielding, say, 22%, you can expect to receive, over the next 12 months, an income return of 22% less a loss of 7%, for a net return of 15%. And that's using some pretty pessimistic assumptions.

What would it take for an investment in junk bonds today to equal the essentially zero rate of return on cash today even if default rates are horrendous? The yields on junk bonds would have to rise from 22% today to about 26% for the one-year holding period return of junk to equal the current yield on cash. That's because an increase in the yield on junk bonds of 4 percentage points implies a price drop on those bonds of almost 15%. So things have to deteriorate meaningfully for an investment in junk to equal an investment in risk-free cash.

If junk yields don't increase at all, they will produce a return of 15% in the next year. If junk yields decline from 22% to 20% (a drop in the bucket if things in general start to improve), then junk returns will be almost 23% in the next year. If junk yields don't increase for a month, then an investment in junk bonds will produce a one-month return of at least 1.2%, which beats the annual yield on most money market funds! How much longer can investors ignore these numbers?

In short, it's not enough to be pessimistic these days in order to be bearish on the prospects of corporate bonds and equities. You have to be downright convinced that conditions in the next year are going to be far worse than anything the U.S. has ever seen.

As for investor psychology, all it would take for things to improve signficantly is for them not to deteriorate. Because if current yields hold steady, the returns to taking risk will rapidly undermine the confidence of those who are hiding out in cash.

10 comments:

prophets said...

I share your enthusiasm for HY debt, but is it fair to only look at a default rate of 10% in 2009?

For example, GS Credit Strategy also sees 10.9% default in HY in 2009, but a cumulative default of 30% through 2013 (next 5 years).

The Therapist Is In said...

Seems like investment grade would be a natural first step before jumping into high yield; won't investment grade bonds be the first to return to reality?

Scott Grannis said...

prophets: a 30% cumulative default rate in the next 5 years would be catastrophic--about twice the maximum experienced in the Depression when the economy shrank by 25%. Nevertheless, the HY market is already assuming that the cumulative default rate will be even higher than that.

To understand easily, suppose the whole 30% defaults at the end of the first year. If you invest $100 on day one, at the end of year one you will have $22 in income, less 30%*70% in net principal loss, leaving you with $101 to invest at 22% (assuming yields stay constant). That money will then earn 22% for the next four years. That's a pretty generous return on investment, wouldn't you say?

Scott Grannis said...

Therapist: HY bonds have a huge incremental premium built in, so I think they would be the biggest beneficiaries of any scenario in which the economy does better than the depression that is expected.

prophets said...

i see your point, though I think some of these HY companies have little recovery value. a lot of them are service businesses where human capital is the only value. company goes bankrupt, people walk out the door, and you have little more than customer lists or the auction value of an "Enron" sign on ebay.

i guess it also begs the question of what is the risk vs. the reward, given where equities are today. some equities have completely unlevered balance sheets (ex. intel).

how do you price the risk/reward of an equity vs. HY?

it also points to therapists question. sure HY has more yield (premium built in), but it also has more risk. how do you measure to the risk/reward in IG vs. HY?

thx for your thoughts, i'm glad i've found this blog!

prophets said...

just FYI I also calculate the following:

$100 invested:

30% defaults, with 30% recovery = $9 value

70% ($70 principal generates) 22% = $15.4 interest

+$70 remaining principal

nets out to $94.4 after first year, but then you do gain 22% for the remaining years thereafter. i realize they do pay interest payments throughout the year so maybe you do get that 22% on the defaulted loans but could be a little closer to ~90-95 perhaps. but very unlikely... thx for your help here.

prophets said...

IRR calculated here:

http://img529.imageshack.us/img529/2944/hyhypotheticalfx7.jpg

I think one mistake you are making is that you assume the dividends are automatically reinvested at the same rate and that is a separate investment decision.

It makes more sense to me to calculate the IRR on a single investment decision (ie. buying a portfolio of HY debt). At time=0, then the IRR is still low teens in a scenario of ~20-25% default rates after the first year.

So attractive, but not out of this world crazy attractive compared to equities, I think.

Clearly, the value of investing in HY or credit in general is simply finding things that will not default. (yes, I know that is quite obvious... ! )

prophets said...

url didn't take properly:

http://tinyurl.com/64g2m9

had to tinyurl it.

prophets said...

My last comment, as I am flooding your blog:

3 yrs of defaults, 5 yr holding period

http://tinyurl.com/625jsv

think you need a ~7% default rate on avg. for the next 3 years in order to justify HY investment. obviously defaults will be front loaded, but this scenario is probably closer to real world.

obviously credit selection is still paramount.

Scott Grannis said...

prophets: I haven't tried to use a scientific method of calculating returns given default rates and recovery values, and I don't have a good calculator available. But I do know that the default rate implied by current pricing is enormously high, and if we ever approached that in real life it would mean massive destruction in the economy. Since I don't think that is realistic then I have to believe that the yields available today on junk bonds are extraordinarily attractive. Another point that is important is that if yields don't rise further then junk becomes a compelling investment relative to other things. A company's bonds can pay off even in default and even if the equity goes to zero.

Anyway, thanks for your interest.