Opinion

Felix Salmon

Felix Salmon smackdown watch, cov-lite edition

By Felix Salmon
June 2, 2013

Stephen Foley was clearly thinking along the same lines that I was on Friday: shortly after my piece on cov-lite loans came out, he published his own, coming to much the same conclusion. (But expressing it better: I love “the bearable liteness of covenants”.)

Not everybody was convinced, however. David Merkel has the meatiest pushback, starting with this:

Ask a loan holder, “All other things equal, would you rather have a cov-lite loan or a normal one?” The answer will always be “Normal, of course. Why are you asking such a dumb question?”

Loan holders would prefer more defaults with lesser severity than fewer with higher severity. What is flexibility to the borrower is a higher degree of expected credit costs to the lenders.

I think both Foley and I would disagree here. All other things equal — including one’s position in the credit cycle — I for one would rather have a cov-lite loan than a normal one. For one thing, as Moody’s discovered, cov-lite loans do not default with higher severity than normal loans; in fact, their recovery rate, even for loans originated in the frothy years of 2005-7, was a whopping 90%. Intuitively, one might expect cov-lite loans to have “a higher degree of expected credit costs to the lenders”. But empirically, that just doesn’t seem to be the case.

More conceptually, if I’m lending to a company, I want it to concentrate on survival, when it gets into trouble, rather than being thrown to financial wolves whose interests are not even aligned with each other, let alone those of the company. Here’s Foley:

Potential covenant breaches excite lenders more for the opportunity they present to take some fees and reprice their loans at a higher interest rate.

And even this is not a straightforward matter. Since leveraged loans are parcelled out and traded on the open market, companies may have to satisfy many, often competing, interests – from fund managers concerned mainly with interest income, to vulture funds who may be playing in a company’s debt in the hopes of pressuring management into merger and acquisition activity. The process is about as far away from a friendly chat with your sympathetic bank manager as it is possible to get.

Such negotiations come at a time when a company’s energies might be better focused on steering the business through a tough patch.

In other words, I can see the attraction of covenants both to banks and to hedge funds who might be looking for the perfect leverage point from which to take over control of a company. But as a passive investor, I’m happier in cov-lite, especially if I have any exposure to the company’s equity.

Merkel then makes a second point, which is slightly stronger. Look at the credit cycle, he says: first debt is an attractive buy, and then yields fall, and then covenants weaken, and then defaults rise and the market crashes, at which point debt is an attractive buy again. If that’s the cycle, then a rise in cov-lite issuance is a clear sign that the debt markets are toppish.

I’m not sure I buy this either. For one thing, actual default rates turned out to be much lower, in the wake of the credit crunch, than the level of defaults that everybody was pricing in and expecting after Lehman Brothers collapsed. And the reason is simple: monetary policy, which loosened up considerably and unleashed a wave of liquidity into the debt markets, allowing troubled companies to refinance. Right now, monetary policy is still loose, and companies have never been more profitable. The combination makes it unlikely that we’ll see a spike in default rates any time soon.

And then there’s the case I was trying to make in my post — that investors have finally realized that cov-lite loans are just as good, if not better than, their covenanted siblings. As a result, the spike in cov-lite issuance isn’t a function of investors giving up protections they’d really rather retain; instead, it’s a function of investors happily giving up protections which in practice only serve to make the borrower’s life more difficult and expensive.

Merkel’s credit cycle makes sense, but it’s not inviolably true. And yes, we’re certainly in the bull-market phase of the credit cycle right now; whenever that happens, a bear-market phase is bound to arise sooner rather than later. Cov-lite loans will get hit in that bear market, along with all other credit instruments. But there’s no particular reason to believe they’ll be hit harder than loans with covenants. In fact, I suspect that, once again, they’ll end up outperforming.

Comments
8 comments so far | RSS Comments RSS

The key phrase here is “all other things being equal”. Right at the bottom of your first piece was the word “subordination”. As you explained, cov-lite loans tend not to cause as many problems as you might imagine, because they benefit from having junior layers of capital beneath them. Which is another way of saying that they have lower leverage (or senior leverage, if you will, or a lower loan-to-value ratio).

Cov-lite lenders are happy to let a borrower blow through whatever warning signs a covenant might provide because they’re comfortable that, at the bottom of whatever deep dark pit a borrower finds itself, there’s some value that they can recover.

The sections of the Moody’s study you quote don’t afford us a like-for-like comparison. Less highly leveraged loans will, naturally, achieve better recovery rates than highly leveraged loans (unless the recoveries cited include those junior layers of capital). So what happens when you control for leverage, expressed as debt to Ebitda, debt to EV, whatever metric you care mention?

Of course, there’s plenty of recent history in the ABS market of senior lenders losing their shirts despite having plenty of layers of junior capital below them…

Posted by gringcorp | Report as abusive
 

“For one thing, as Moody’s discovered, cov-lite loans do not default with higher severity than normal loans.”

Again, this is not evidence for or against pursuing one or the other. This is a fact about the *kind* of loans that end up in each bucket, and the standards it took to get them there. Cov-lite loans possibly *needed* higher underwriting standards because of the lack of covenant and consequently have a lower default rate.

This is not clean data coming from a careful experiment that you can use to argue your point. The empirics do not indicate what you think they indicate.

Posted by absinthe | Report as abusive
 

I was doing research on this in the 1990s and the subject was a well-worn path even then. Covenants don’t work. They don’t protect lenders – the only one that works even a little bit is the negative pledge, and there are lots of situations (mainly leveraged buyouts) where it’s obviously inappropriate to include a negative pledge. Felix is right here – what we’re seeing is not some kind of collapse in lending standards (if this was happening we would be seeing a lot of different pieces of evidence, not just this one point). We’re just seeing the gradual dawn of lenders actually thinking about what covenants a loan agreement really needs rather than using boilerplate.

Posted by dsquared | Report as abusive
 

@absinthe, I think you’re confusing default rate and recovery rate here. Everybody agrees that cov-lite loans have lower default rates, that may or may not be a function of them being more assiduously underwritten. But are you also saying that higher underwriting standards lead to a higher recovery rate? As high as 90%?

Posted by FelixSalmon | Report as abusive
 

I hate to show up in a second post with essentially the same point, but as someone who actually has traded leveraged loans (and make no mistake: the cov-lite loans being discussed in this context are to “high yield” borrowers)… the empirical evidence says lower defaults and lower recoveries. You can misread 90% to compare it against unsecured bond recoveries, but it doesn’t change the fact that an earlier default — all else equal — will mean a higher recovery for the levered lender. There is no magic outperformance to be had simply by dropping covenants from a credit agreement.

Also: let’s be clear about something when we’re talking about “lenders”… what we mean in this world is CLOs and levered loan funds — they do not care about the returns to the equity and yes, they are interested in collecting a fee to give a waiver for a busted covenant but — except in the most distressed situations, where you get real “distressed” players like Avenue, Oaktree, and Apollo involved — have no desire to use covenants to take control of a company.

As for calling the credit cycle, I definitely see Merkel’s point, though I would prefer to look to the issuance of dividend deals, PIK loans, and holdco loans as real signs that we’re getting toward a top.

Posted by CasualSophist | Report as abusive
 

Hello Felix… as a Brit you may be interested to know that I have published a book on Amazon called the “Patriarch” about the first Merchant Banking House in India, emerging from the aftermath following that one decisive battle on the afternoon of the 23rd of September, 1803 which changed the face of India forever… excerpts are posted on the website http://www.houseofshah.com and on twitter @pravbank

Posted by Bludde | Report as abusive
 

@FelixSalmon I have no idea. My point is just that the data’s not clean enough to say empirics are on your side. The most they’re good for is confirmation bias really. If you think there’s a clean analytical/causal explanation it should stand on its own.

Posted by absinthe | Report as abusive
 

I should also point out that underwriting standards are likely not the only confounding variable. Maybe if you could find them all there’s some statistical analysis that could prove your point; maybe there isn’t.

Posted by absinthe | Report as abusive
 

Post Your Comment

We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/
  •