Monthly Investment Commentary | August 2008

Amidst a lot of uncertainty in the markets, returns in July were all over the map. For equities, in addition to being a volatile month, there were sharp disparities in returns, with larger- and mid-cap domestic stocks experiencing losses, while small-cap stocks saw decent gains. The Vanguard 500 Index Fund, which is based on the S&P 500, lost 0.8%, while the small-cap iShares Russell 2000 gained 3.7% (some other small-cap indexes were up less than this, however). Mid-caps continued to march to a different drummer, this time on the downside, with the iShares Russell Midcap losing almost 4% for the month. For the first seven months of the year, domestic equity losses were lined up by market cap, with large-caps losing more than 12%, mid-caps off just over 10%, and small-caps down about 6%. Foreign stocks had a tougher month than domestic (overall), with Vanguard Total International Stock Index Fund off 3.7% in July and over 14% this year. REITs were a bright spot in July, with gains of just over 3% and for the year to date are now down only slightly. Short-term, emerging-markets local-currency bonds (based on the JPMorgan ELMI+ Index) were up 2.2% in July and now have a double-digit year-to-date gain. Vanguard’s Total Bond Market Index Fund was flat on the month, and has managed a modest 1.1% gain for the year through July.

High-Yield Roundtable with Edward Altman and Martin Fridson (July 22, 2008)

Given the unprecedented dislocation in the credit markets, and our belief that it is likely we will see a tactical opportunity in high-yield bonds at some point in this cycle, we asked two leading high-yield and distressed-investing experts to share their views. Edward Altman is the Max L. Heine professor of finance at the Stern School of Business, New York University, and director of the Credit and Fixed Income Research Program at the NYU Salomon Center. He is internationally regarded as an expert in corporate bankruptcy, high-yield bonds, distressed debt, and credit risk analysis (he developed the Z-Score, a tool used to forecast bankruptcies). Martin Fridson is the CEO of Fridson Investment Advisors and author of Financial Statement Analysis. Fridson has analyzed high-yield at Morgan Stanley, Salomon Brothers, and Merrill Lynch, and is similarly regarded as one the foremost high-yield experts.

How material are the compositional differences in the high-yield index today versus the past?

Fridson: The real change in the universe occurred between 1990 and 2000. There hasn’t really been a lot of change since then. Going into the [1990-1991] recession, we had a universe that was about 55% in the top [highest-rated] category and only about 2% in the bottom category. Ten years later, it was more like 35% in the top category and 15% in the bottom category. That’s about roughly where it’s stayed since then. So there is considerably more default risk, given the same economic conditions, than there was in 1990. It’s important to keep in mind that this is not a linear relationship. In other words, if a bond migrates from Ba to Caa, it more than doubles the risk of migrating from Ba to B. 

How material is the change?  How comparable is the index right now from a risk standpoint to what it was just prior to the recession in 1990-1991 and the 2000-2001 period?

Fridson:  If the default rates within the rating categories turned out to be comparable to what they were in the ’90-’91 recession ... then with the present ratings mix, instead of seeing a peak default rate in the 11% to 13% range, you’d see an average default rate in the mid teens range over a period of the couple of years surrounding the recession. That’s an indication of how much the ratings mix has deteriorated. So just as a matter of arithmetic, the impact of the deterioration in the ratings mix is very, very severe. The reality is that we had a very mild recession in 2001 after this deterioration occurred. We haven’t really tested the present ratings mix against a recession as severe as ’90-’91—which wasn’t all that severe by post-war standards. But it was more severe than the 2001 recession.

Altman:  I’ve checked out the numbers by the subsequent default rates by rating category, to the proportions that existed at the end of ’07.  You’d get as much as a 16% default rate, assuming that we had a severe recession like ’91 ... around 6% to 8% if we had a mild recession.

Do you have any thoughts about the likelihood of an extended deleveraging cycle—with a weak consumer and a sub-par corporate earnings environment—and how that might play out for the high-yield market?

Fridson:  For the moment, we seem to be limping along.  But I’d point out that in the latest quarter, we had the largest quarter-over-quarter contraction in outstanding bank credit since 1948. We’re really just seeing the beginning, evidently, of that effect of the contraction in credit. I think that’s been somewhat delayed. That’s going to affect not only the issuers of speculative-grade debt, but also both the business customers and the consumers of those companies.

I think it’s going to have a really compounding effect. We could be in for a very rough ride on the economy. … The impact of the consumer cutbacks is already being felt. We see very clearly a high level of distress in retailing, restaurants, and casinos.

Right now, it would not appear that we’re likely to be at a peak [in default rates] until, I think, as early as late 2009 … possibly even into 2010. 

Altman: Interestingly enough, the last two recessions actually were preceded by increasing default rates for two years or so … then continuing to rise for eight or 10 or 12 months after the recession started.

If you figure that we’re not in a recession yet—at least not statistically—then we’re talking about a shorter preceding period, perhaps less than a year. We can discuss why there are lags in the corporate default rates and why it’s not increased even more than it has. One reason is that a number of the companies that are in the low-[quality] spectrum were able to refinance at very attractive rates.

Fridson: The market seems to be anticipating a higher level of defaults than we’re seeing in spreads.  Risk premiums have increased faster than defaults have increased, if you compare this with past cycles.

There has been a pattern of how defaults have progressed during the last two big cycles. They would go from very low, spend a couple of years ramping up, followed by one or two years with big defaults, and then a very quick return to below-average defaults. Is there a structural reason for that pattern of defaults?

Fridson: After a large wave of issuance, inevitably there’s an echo effect of defaults picking up, just as a natural matter of course from the volumes ... then the gradual ... tightening up of monetary policy [that precedes a recession] … aggravates the problems of companies that were financially vulnerable to begin with [but that] were able to get past that the first year or two … because they had a lot of liquidity.

They’re now made all that more vulnerable by the weakness of the economy and the impact on their earnings and cash [flows].

What is your outlook for spreads—both shorter term and three to five years from now? 

Altman: The price of credit was the lowest—at least in the terms of high-yield bonds—in the history of the market on June 12 [2007]. I don’t think we’ll go back to that [and] the pendulum has swung quite a bit in the direction of much higher spreads.  Having said that, if the default rates escalate above 10%, which is entirely feasible over the next 18 months, then I think spreads will likewise rise rather than stay the same or fall. Historically, [spreads have] been as high as 1100 [basis points].

We’ve got a ways to go … to get to historic highs. That’s not out of the question in the next two years. If we say three to five [years], then I’d guess that we’d be out of the recession and out of the peak default rate period at some time, and they’ll fall back to more normal levels—which are around 500 or so.

If [the next] three to five years is a more normal period, then spreads will fall [approximately] 250 [basis points] from what they are now. However, in the interim shorter period, they could go 300 or more above what they are today.

Fridson: I absolutely agree with that. The default rates are likely to get somewhere in the vicinity of past cyclical peaks. In the peaks associated with recessions, it’s hard to imagine a milder recession than the last one, and the ratings mix hasn’t improved since then. We ought to see similar peaks in defaults and spreads.

When you’re looking at spreads widening in the bearish scenario, how do you see that materializing between high-yield and Treasuries?

Altman: I never forecast interest rates because I know I’m always wrong. Most other people are, too. I’m not saying anything about government yields.  You could play both ends of the assumptions with respect to the Fed actions. On the one hand, they want to keep us out of a recession … on the other hand, the weak dollar and the high price of oil and the inflation risk [suggest that] to lower interest rates would be a huge, huge mistake.

Fridson: Yes. It’s also important to keep in mind that the Fed only controls the short-term rates anyway. …There’s a tendency for Wall Street to explain everything away and say, “Treasury rates went down.  That’s why spreads are wide.”  You can probably very selectively choose a point in time to make that appear to be the case. But that’s not … analysis. That’s just cherry picking.

Is there any reason to think that defaults could be disproportionately higher than they have been in the past for BBs or Bs versus where they have been in the past? Is there any sort of structural difference or underwriting difference that went on when these bonds were issued that could make those bonds riskier this time around?

Fridson: One important factor to keep in mind is that if the rating agencies were to deliberately change their rating standards … it would be one of the stupidest business decisions I’ve seen in my career—given that their structured finance business is all based on the continuity of the rating agency. They might do it inadvertently. But I would just keep that in mind. I’d say that the continuity of the rating standards over long periods of time is fairly impressive. But you really have to test it after the fact. Anything we say now about how it will look when we look back on the present statistics five or 10 years from now is very speculative.

The other argument that gets advanced in this vein is that all of the defaults in that last cycle are all because of telecoms. Now telecoms are a much-reduced portion of the universe. We looked at Merrill’s index and found that in 2001—the peak year [for defaults]—taking out all the telecoms brought [the default rate] down to 9.9 [from 11.75%]. You’ll hear some people talk and say, “Oh, well—we never would’ve had more than a 4% or 5% default rate if it hadn’t been for those awful telecoms.” That’s simply not true.

You said you believe there’s going to be a longer, drawn-out default cycle. If we also have an extended period of deleveraging, what might the impact be on recovery rates?

Altman: A lot of the recovery rate has to do with the economic period that we’re in when the firm either defaults or emerges. For example, in today’s period, with a lack of liquidity and the very pessimistic outlook for the economy, you’ll find firms that are already in bankruptcy having trouble getting new financing. [Others] simply have lower expectations of their earnings and cash flows. Therefore, their prices on the securities plummet. I think everything is going to lower recoveries going forward. That’s a long-winded answer to say the LGD—loss-given default—is going to be exacerbated by higher defaults and lower recoveries over the next 18 months or longer.

Fridson: A couple other things I’d mention:  One is that there is generally more secured debt in capital structures than was true in past cycles. [This is] a function of a tremendous boom in the demand for loans … [and] the rise of collateralized loan obligations in this cycle. That would importantly lower recoveries, certainly, for the unsecured holders beneath the secured loans, and possibly even for the secured loans themselves.

The other point I’d make is that the market is clearly signaling that recoveries will be lower on this basis. Over the last four years, there has been a relentless widening of the spread between senior and subordinated issues of the same companies … and that would be an indication that the market is expecting lower overall recoveries.

By the way … it is possible … for the first time in this cycle that hedge funds may [be a material factor]. Hedge funds really were not significant players in this market up until the present cycle. One potential strategy is to go long the senior secured debt and short the unsecured debt—or even the equity of the company. To use the voting rights in that secured position to push for a liquidation, in which case the secured holders will get paid not only more quickly, but there will be no material recoveries for the unsecured holders—particularly the subordinated holders. So, here you wind up making money on both sides. It’s not illegal to do that. It will cause a lot of consternation if and when it happens. There’ve been some indications of trades like that having already been put into place. Similar kinds of strategies can be pursued using the credit default swaps, where you can take a much larger position on the short side, and use the voting rights to even lose money on that side to block a settlement and cause deterioration in the value of the unsecured debt, and increase the value of selling insurance on the company.

I absolutely agree that the average recovery over the full cycle has been pretty steady for several decades [but] we could see for the first time some material divergence from that stability.

What’s more important, absolute levels of debt, and with that, refinancing risk, or interest coverage and the ability to service existing debt? Do you feel strongly about one of those metrics versus the other being the more important variable to keep an eye on in the current environment? 

Altman: I’d say the servicing is much more [important]. The maturity schedule is definitely important [too]. It all depends on the ability of the firm to refinance, rather than the amount.  And the ability to refinance has to do with the ability to service the either continuing high level or even increasing level [of debt], if you’re going to refinance and try to grow, as well. Therefore you look at something like debt-to-EBITDA ratios or other means of cash-flow coverage as a better metric than simply [the absolute level].

Fridson: I would say that the market tends to focus more on the coverage ratios. In our shop [Fridson Investment Advisors] we actually do put a greater emphasis than most do on the debt values, as relative to the enterprise value. We want to be very sure that we have a good cushion beneath us in the capital structure whenever we’re in an investment.

Spreads and yields look fairly attractive right now, but we haven’t really seen the defaults starting to flow through, and we know that that’s going to have an impact on performance on the downside. In the prior two recessions, it looked to us like the big returns for the asset class didn’t really materialize until defaults were near or past their peak levels. Does it make sense to overweight high-yield in front of the default wave since you have a bird in the hand with the attractive yields and spreads? Or does it make sense to wait until the defaults have really started to materialize and then get into the asset class? 

Altman: Well, I’d like to expand the discussion, if you don’t mind, to both high-yield and distressed debt, in this respect. Distressed debt is where I’ve actually done more work. My feeling is that it’s the absolute best time ... to pile into both asset classes near the peak of the default rate. The best returns ever were just after the peak in 1991. My guess is that it’s too early for both high-yield and distressed. This could be very early if the cycle is as long and drawn out as I think. 

It’s interesting that those entry points, it sounds like, were very close to the bottom for the stock market also.

Fridson: Yes. The returns have tended to be pretty mediocre during the rising phase of the default-rate cycle. I concur with what Ed says. In the last couple of cycles, the 800 basis points [spreads] have been a turning point. The caveat this time around would be that the spreads are already at a level that was consistent with considerably higher default rates in past cycles. So it looks like there is an additional liquidity premium in the picture, and that may indicate that really it will be somewhat wider before you get to the point where you can [invest] with high confidence that … it will [turn out to be] a good entry point. So that is a caveat you have to attach to it. To some extent, though, the mere fact that you’re collecting a higher yield improves the odds in your favor.

Is it reasonable to expect that the factors that would cause high-yield to weaken before it recovers would be factors that would also have a negative impact on the stock market?

Fridson: I think generally that’s correct. They’re probably going to have a reasonably close connection this time around.  It hasn’t been true in every period. There have been times at which factors very unique to the high-yield market have caused it to diverge from the overall trend.

But I think recently and in the near future, it should be the case that the general weakening in the economy and an extension of earnings weakness to some other sectors that have not felt it yet probably would have an adverse effect. The big difference is that the industry distribution is very different: high-yield doesn’t have anything like the representation of financials that have become so predominant in the S&P and Dow and those indexes.

Do either of you have a total return expectation for the broad high-yield asset class over the next five years? 

Fridson: I think it’s likely to be up in the range of historical averages [of] 9% a year, but with very, very wide variance around that. We’re likely to have another of those years like 28% in 2003 in there. 

Altman: I agree with the latter. Behind the whiplash [we could see] a huge … upswing [a few years out] that’s going to do wonders for the average. But it’s going to be painful getting there.

Are there any big risks that could cause that five-year number to be materially lower than that average?

Altman: One is the length of the recession. We haven’t had a long recession in a long time. If we do have a two-year recession, which is not out of the question, then assuming that the default rate is going to peak a little bit after that, that might be a little bit different than the past recessions and high-default periods. That’s one thing that could be different than the past. Having said that, you’re still going to have your positive period in there.

Fridson: I guess that if you’re thinking about really dire scenarios—if we have a really serious spike in inflation and therefore long-term interest rates—even a contraction of risk premiums that I think Ed and I both foresee could be overwhelmed [by these negatives]. If you’re talking about interest rates going back to the levels of the 1970s—yes, that would be pretty hard to overcome.  That could pull down the returns substantially.

Thank you both very much. We appreciate your being so generous with your time.

—Stapp Financial Planning, PLLC

This information is also available at www.stappfinancial.com, or you can download a PDF version. To contact us about the newsletter, send an e-mail to bstapp@stappfinancial.com. Before acting on any advice it is recommended to seek appropriate counsel applicable to your individual circumstances.