Thursday, October 30, 2008

20% high-yield defaults? Don't underestimate the power of the autos!

On Monday the yield on the Merrill Lynch High Yield Master index reached a shocking 19.6%. That yield is admittedly enticing, but the real question is, how many of those high-yield bonds might default? In short, if we're getting 20% yield, could we wind up suffering 20% losses in defaults?

According to Moody's, the largest default rate in history was 15% in 1933. In the post Depression era, there have been three years which produced double-digit default rates: 1990 (10.1%), 1991 (10.4%) and 2001 (10.6%). The average recovery rate (i.e., the amount the bond is worth immediately after default) is 32% for the three peak default years. So if a portfolio suffers 10% in defaults with 30% in recovery, it has actually suffered 7% in total credit losses.

That history would seem to favor high-yield, even admits an ugly economic forecast, given the initial yield of 20%. But risks remain. First, the proximate cause of most corporate bankruptcies is either an inability to roll over debts or a demand by creditors for collateral which the company cannot obtain.

Right now roll-over risk in high-yield is higher than any time since at least the early 90's. Junk-rated companies can obtain funding through one of two routes, either bank loans or the public bond market. But neither of those are open to lower-quality borrowers right now. There have not been any new high-yield issues for the entire month of October. And banks remain highly unwilling to lend to anyone, much less to high-credit risk firms.

Should the credit markets thaw somewhat in the near term, new deals may be possible. But even if that happens, how many companies will be able to operate where the cost of debt is 20%? Some companies look for ways to borrow in the secured market, where companies pledge specific assets to lenders, which in effect subordinates existing bond holders.

Then there is the 900-pound gorilla in the junk bond market: the autos. Ford and General Motors alone make up about 7% of the high-yield index. Recent stories in the media suggest that GM will need a loan from the government to complete a merger with Chrysler, and even then there are no assurances that the companies significant problems can be solved.

There are also stories that GM sought help from Toyota. Sounds awful desparate. In fact, I'd argue that a bankruptcy would be better for the American auto industry long-term, as it would allow firms to focus on production rather than dealing with an out-dated union structure. That's the path the airlines followed in 2001-2002.

If high-yield defaults follow the "normal" recessionary pattern of about 10% defaults, but GM and Ford default as well, that would bring total defaults to about 17%, disturbingly close to the 19.6% yield on the index currently.

Given the extremely high rate of interest on high-yield currently, the odds are good that high yield will produce positive returns over next 3-years. Even if defaults spike in the next 12-18 months, investors will likely be well-compensated for the credit losses over a longer period of time. But if one is to take that tact, bear in mind that the near-term could be very painful, and that market-quotes (or NAV values on mutual funds) could still fall from here.

Its probably best to think of high-yield as a low-beta equity investment rather than a bond investment. Go into it with the understanding that equity like gains and/or losses are possible, and size the trade accordingly.


PNL4LYFE said...

Are those peak default rates for single years? If so, a couple of those years in a row could easily wipe out the 20% yield.

Yesterday, MGM priced a senior secured deal. 750mm secured by the first lien on New York, New York casino. 13% coupon priced around 93 to yield 15%. First data point I've seen in awhile; obviously not that encouraging.

Regarding the autos, I agree with Phil LeBeau's argument that bankruptcy would have much worse consequences than it did for the airlines. A plane ticket is something that you only use once; price and flight times are the only things that people really look at. On the other hand, a car will be owned for years; you need to be confident that the company behind your warranty will be around for a few years. And whether there is any truth to it or not, bankruptcy will be viewed as a sign of low quality products. It's obvious that the government is going to step in to save the automakers at some point. I think it will be more productive and less costly if they do it in such a way that bankruptcy is not required.

Accrued Interest said...


Those are the single year peaks. Obviously 1990 and 1991 combined to be about 20%. But if you starting considering multiple years, you neeed to also consider multiple years of income generation.

As for the GM bankruptcy... its one thing for the government to support a company where the ongoing business model is sound, but there have been some past decisions that are causing problems. That's where we are with banks.

But GM's basic struction is broken. Its based on a 1970's union-based world, which is dead. Government proping that up is pointless.

Dividends Anonymous said...

Thanks for those insights AI,

I've never come across your blog before but certainly will return to read future content. I really appreciate the commentary on bonds.

Phani said...


Can you please explain or share the information you have on CDS Pricing. For Example if you are given 6 months to 10 years spreads, recovery rates and yield curve, i s there is any algorithm to calculate CDS Price.

Can you please share the information you have on CDS Pricing.

Thanks & Regards,


Unknown said...

Accrued Interest,

Couldn't the government stipulate a final resolution (reduction) to the union cost issue as part of a bailout? Call it a public policy decision or some such...?

-Thomas Hudson

Accrued Interest said...

Well, the govt can do whatever the hell it wants. But let's be honest, in a Obama admin I'd expect the union to be protected.

Anonymous said...

«its one thing for the government to support a company where the ongoing business model is sound, but there have been some past decisions that are causing problems. That's where we are with banks.»

HAHAHAHAHA! That's so outrageously intellectually dishonest that it is even funny.

The business model of the financial sector has included:

* creating off balance sheet loss holding vehicles;
* selling massive amounts of debt against impossibly high real estate valuations;
* pretend to offload risk by selling each other CDSes not backed by capital;
* creating impenetrably complex derivatives and buying amazing AAA ratings from friendly raters;
* donating enormous amounts of campaign funds to politicians of both parties (mostly Republicans) to look the other way and restrain regulators;
* driving paper profits to a very large percentage of all USA profits and then taking 50% of paper profits as cash bonuses;

and that's a sound business model?

«But GM's basic struction is broken. Its based on a 1970's union-based world, which is dead. Government proping that up is pointless.»

GM's business model is broken only by hopeless mismanagement. There is nothing broken in manufacturing and selling cars, instead of scam-debt, and even the union aspect of it is not that big a cost, and fixing that could cost a lot less.

In a way the basic problems of the financial industry are that it overpaid executives with very bad incentives, and as a result overexpanded and resorted to massive shysterism knowing the government would rescue it; the basic problems of the automobile industry is that it has poor management and its workers compete with less paid foreign workers.

The problems of the financial industry are pretty fundamental to its business model, indeed have driven it, those of the automobile industry are fairly accidental to the business model, which is essentially sound.

So it is trillions for rich "sound" bankers but not a penny for "pointless" unions?

Accrued Interest said...

Have I mentioned how much I hate comments that start out by laughing at my argument? In what world is that anyway to convince anyone of anything?

Beyond that, are you seriously suggesting that GM has a better basic business model than Wells Fargo?

How can you dismiss GM's problems with "mismanagement?" How is banking any different? The fact is that we can wipe out all those problems you mention by simply injecting capital into banks. You can't wipe out GM's legacy liabilities in the same way.

Anonymous said...

«GM has a better basic business model than Wells Fargo?»

It surely has a much better business model than Countrywide, let's leave Wells Fargo alone, nobody is thinking yet to bail it out.

As to the Countrywide vs. GM business model, here is what an analyst thinks of the Countrywide one:
«I would argue that Countrywide is insolvent. Their only asset is their pricing platform, their business algorithm, and that's not working. The next biggest asset they have is the toner for their copiers.»

«How can you dismiss GM's problems with "mismanagement?"»

Well, easily: in good times GM is sort of viable. And in both and bad times Toyota is viable, and if GM was managed as well as Toyota it would be viable too, just less profitable. After all GM's labor cost problems exist in other countries (e.g. Germany) and they are not insurmountable.

«How is banking any different?»

But we are not comparing GM with the whole of banking (e.g. the few semi-sane bits of it like Wells Fargo), but a GM bailout with the financial industry bailout. That is the difference between say GM and AIG or GM vs. Lehman.

The GM business model is basically manufacturing and selling cars in the USA with USA workers, and that of AIG or Lehman is basically selling toxic paper based on fantasy ratings and the carry trade, and then paying executives cash from the paper profits resulting from insufficient provisions against risk.

The GM model may be a bit expensive but is sustainable, but the AIG or Lehman model is only sustainable if the fantasy ratings and the carry trade continue and the risk does not happen or is paid for by the taxpayers.

«The fact is that we can wipe out all those problems you mention by simply injecting capital into banks. You can't wipe out GM's legacy liabilities in the same way.»

This is so outrageous that is ridiculous, as it is precisely the opposite: the legacy costs of GM can be wiped out by nationalizing them, and it is not even that expensive, the problems with AIG and Lehman's business model is that it requires ever larger leverage and an ever growing bailout, which is now in the hundreds of billions.