That doesn’t mean we can’t have a crisis, but the threat is less than it might be.
There’s been quite a reaction to the posts from two weeks ago about the potential that our next financial meltdown might be quietly germinating in the dark recesses of the credit derivative market, and rightly so. There are still people linking to the second in the series today.
After the break, I’ll follow some clues we can look at to see whether another mega-mushroom of risk is growing in the dark basement of finance, the way it was in Finance Monsters….
Pain ahead, but not panic.
Over the past week, I’ve been looking at statistics on credit trading, both Credit Default Swaps (CDS) and CDOs (Collateralized Debt Obligations), paying attention to the growth of “synthetic” bonds and bonds created from corporate loans (CLOs, or Collateralized Loan Obligations).
My verdict: There’s not enough exposure floating around the financial system to assure we have another avalanche of defaults, even if the energy complex suffers substantial credit losses. That said, we could still have a Herstatt Event, where one participant in the system fails, triggering a cascade of failures by others. (The linked article marks the 37th anniversary of that event, kicked off by the failure of Germany’s 25th-largest bank.)
There are two basic reasons that I think a catastrophic systemic failure is unlikely right now. First is that the size of the exposure is not on the scale that we saw before the 2008 event. The second reason is the advent of exchange-traded “single name” CDS.
When Wall Street dealers and banks stood as counterparties to the bond-by-bond single-name CDS before the 2008 crisis, there were enormous demands for capital as the prices went down on those bonds. As a result, undisclosed exposures became the reason to fear having any bank as counterparty for anything, even cash deposits or Letters of Credit.
Today, about two-thirds of all CDS trades are booked with central clearing operations and subject to calls for margin capital. Failure by one participant results in that participant being wiped out, but not a daisy chain of other participant failures.
For the other reason, I checked records that FINRA, the securities dealers’ self-regulation organization, keeps of various deal types, broken out by year of issuance, and what I could glean from the MarkIt web site without being at an institutional client that pays for details.
When I look at the FINRA spreadsheet for CDO issuance from 2000 through today, I can see the historic rise (and fall) of structured finance (SF) CDOs, the kind that AIG insured. A typical CDO deal built from structured finance assets had 100 bonds, all rated BBB and BBB-, of which as many as 50 were subprime MBS.
Those mortgage bonds were attractive because they sold for higher yields than BBB bonds from other kinds of deals. But the supply was limited, since the BBB and BBB- tranches of a typical subprime MBS deal might be only $5 to $10 million in size. During the height of the madness from late 2005 through the first half of 2007, hundreds of billions of dollars’ worth of deals were sold based on those bonds. But the demand was so high that another kind of deal became popular – the “synthetic” CDO, which used CDS to generate the cash flows, as a replacement for real bonds.
In the year 2000, just over $1 billion of Structured Finance CDOs were issued, heavy with subprime MBS. That dropped below $800 million in 2001, rebounding, and then off to the races, with $17.5 billion in 2002, $35.1 billion in 2003, $83.3 in 2004, $157.6 in 2005, and $307.7 in 2006, peak year for the housing market. In 2007, there were still enough new bonds to create $259 billion of SF CDOs, but in 2008, SF CDO issuance had dropped to $18.4 billion and then to just $331 million in 2009.
In addition to all that SF CDO deal issuance during the mania, the “synthetic” CDOs were peaking around the same time, with nearly $160 billion issued between 2005 and 2008. All told, the SF CDO business and its synthetic CDO proxies accounted for close to $1 trillion in credit exposure, with half of it concentrated in the lower-rated pieces of the original Mortgage-Backed Securities deals. Add to that the $1 trillion or so in higher-rated mortgage bonds, and the bond market represented around $1 trillion in potential losses.
Adding in unknown trillions in CDS payoffs that were coming from bilateral contracts, and it was almost inevitable that one or more of those counterparties would fail. When AIG failed to meet margin calls against their CDS positions, a global crisis was off to the races. If taxpayers hadn’t paid those margin calls, it was almost certain that a dozen or more TBTF banks would follow them into bankruptcy, since without those AIG payments, they were mostly insolvent.
At present, we don’t see the same rise of “synthetic” CDO deals that extract the theoretical value of using junk-rated energy debt as collateral. We do see nearly 70% more issuance of junk corporate loan CLOs in 2014 than we saw in 2013, but the issuance is “only” $105 billion through the third quarter of 2014 so we might be on track for $130 billion for 2014, well under half the SF CDO peak year with $308 billion. And the junk corporate lending party hasn’t gone on nearly as long as the junk mortgage lending party lasted. The issuance was “only” $64 billion in 2012 and $86 billion in 2103.
The current distress will lower the chances that the following prediction from the securitization trade press last June comes true. Quoting a lawyer that gets his business from these arcane deals “Derivatives Week” and “Total Securitization” shared his opinion that “there is enough liquidity in the underlying CDS market to warrant the CDO market’s revival.”
But even newsletter cheerleaders for the industry admit the potential for danger. The first sentence of the article says “The synthetic collateralised debt obligation market, not seen since and often blamed for the onset of the global financial crisis, could see a revival in Europe and the US, following implemention of the International Swaps and Derivatives Association’s 2014 credit derivatives definitions in September.”
So it looks like there is about $200 billion of exposure to weak energy junk bonds in the US market, and potentially another $200 billion or so coming from CLOs and “synthetic” bonds. If those bonds and CDOs drop to half their value in this oil slump, that would translate into $200 billion of losses scattered around the system. That’s probably not enough to bring on a second global crisis.
I did say “probably” and I still can’t know which bank or hedge fund might get wrong-footed. But if no single outfit has more than a quarter of that risk, energy bonds alone shouldn’t trigger the next avalanche.
For avalanche fans, don’t worry. It’s still snowing. Russian may default.
When Russia defaulted in 1998, spreading shock waves took down several hedge funds, including Long Term Capital. Those shock waves even washed over the US commercial real estate market, leading the largest buyer of lower-rated Commercial Mortgage-Backed Securities (CMBS) to seek bankruptcy protection.
So there’s still a decent, though small, chance for people who want to profit off the end of the world as we know it over the next few months, and there are some taxpayer-guaranteed banks ready to take your bets.