What can be confusing about the carnage in eurozone sovereign bond prices, is that there are so many factors at work all pushing the same way.
There are technical and fundamental factors on top of the blind panic and fear. And that’s not all. There are a number of regulatory factors too, and it’s these demons of our own design that FT Alphaville would like to discuss with you.
The regulatory push, occurring as it is in a vacuum of low liquidity, has the worst possible timing from the banks’ perspective. There has been a confluence of events: rising margins to trade debt, dysfunction in repo markets, even the fact that it’s the end of the year and few like to add on big positions at this time. Like we said — it’s a vacuum.
Furthermore, we’re in an environment where banks are dumping risky assets to deleverage anyway, in order to meet the capital ratio targets set by regulators and to make oneself more presentable to current and potential counterparties (or, dare we say, investors).
All of this with a rather depressing macro backdrop — all the real money investors who want German debt, ASAP; all the eurozone break-up trades; and all the shuffling of credit risk when sovereign restructuring is in the air.
These various factors are feeding on each other and since we are in a vacuum, there is a hell of a lot of feeding.
Enter the regulators, stage left
In a recent Rates Weekly note, Nomura analysts observed “breaking points” in the form of margin calls and ratings downgrades, which (in the case of margin calls) led to a perfect storm for Italian debt last week. But then they added a third, capital requirements. But not the requirement you might be thinking of:
The third aggravating factor is Basel 2.5 implementation which is due to come in for European Union countries on 31 December of this year. This may have crept under the radar screen a little due to the push back of the introduction of Basel III. The main sticking points here are to do with more stringent criteria for capital requirements. And this is probably compounding the pressure on funding markets that is already being felt.
More directly relevant to sovereign markets though is the enhancement of the so-called Incremental Risk Charge based on underlying default and migration risks. These will be tabulated from internal models subject to Committee approval. Presumably they will be premised upon drawing inference from implied probabilities as priced into the respective CDS curves. The latter have risen significantly for many of these markets since the provisions were first articulated. The increased pace of selling being observed at the moment is, in our view at least, partly a result.
It’s worth getting to know the Incremental Risk Charge (IRC), including where it comes from. Basel 2.5 overall is not that well known.
_________________
A Basel 2.5 fairy tale
Once upon a time (er, in 2005)…
Basel regulators wanted to make banks apply an Incremental Default Risk Charge to credit exposures in their trading books.
This was supposed to make the banks compute the capital they needed to hold with respective to specific assets, based on actual default risk.
Everyone took a long time to make their minds up about it, as is the wont with Basel regulations. But! Then the great 2007-2008 credit market implosion (CDOs, CDOs of CDOs, all that jazz) happened. Ye olde version of the IRC was promptly ditched in mid-2008, when it turned out that banks could be blown up through mere movements in the spreads and ratings migrations (not just to “D” for Default) of structured credit. Markets going illiquid made everything far worse and the Basel gnomes got rather cross with the bank quants whose Value-at-Risk (VaR) models failed to predict the losses which ensued. It was time for a new, much shinier way of modelling the risks…
And so the gnomes said: You must do a stressed VaR that looks deeper at historic loss data! And slap a capital charge on any securitisation trading you do! And model assets that might become correlated later! And you’ll make an Incremental Risk Charge model of combined default, spread risk, and ratings migrations too!
So it was that Basel 2.5 came to be. And the little too big to fail all the banks would have to look at their internal models to get ready for it.
This was all about structured credit, so no one really thought about how the change might affect sovereign credit. Even though the AAA bubble seemed to be strong with this sector…
_________________
And then Swiss banks had a go at applying Basel 2.5, because their regulators made them do it by the start of 2011, which was before everyone else. Duncan Wood of Risk wrote a great recent article on what happened next:
When the Basel Committee on Banking Supervision published guidelines on the IRC in March 2009, it said all positions subject to a specific risk charge for interest rate risk would fall within its scope. That includes highly rated government bonds, but because Basel II’s standardised approach to specific risk gives these bonds a 0% charge when rated AAA to AA–, both Swiss banks [UBS and Credit Suisse] interpreted that to mean they would not have to apply the IRC to similarly rated sovereign debt. More than a year later, in one of a series of updates, the Basel Committee published a clarification – all sovereign bonds should be included…
Here’s the clarification in question, from July 2011. (There’s also a more recent one here, hot off the printing presses.)
Frankly, FT Alphaville is consigning “developed country sovereign bonds were considered risk-free” to the “When We Were Your Age” list, and we still haven’t fully gotten over it, dinosaurs though we may be. As Wood says, the Swiss banks pondered how they should model default risk even for top-drawer sovereign bonds, like Germany and the United States. As he writes, they are still considering how to do it, but one possible solution involves CDS pricing data being added to models.
Surely though, banks are going to have look a lot harder at risk even in AAA sovereign bonds held on their trading books. Internal modelling of risk will matter a great deal here. It’s really something that should happen in a world where no one can ignore sovereign risk — but making it happen is difficult. David at Deus Ex Macchiato predicted back in July that banks may simply move sovereign exposure to the banking book.
But there is a slight rub there, we think. All the other factors now driving the crisis speak against holding sovereign debt in banking books as well, increasingly. No?
In the eurozone, spreads have been “migrating” all over the shop, even in AAA sovereign debt like France (and current CDS curves haven’t been kind either). Surely it goes beyond Italian debt now.
This can’t be compatible in the long run with Basel 2.5. Or can it? Or should it? What will we tell our kids when we tell them there are no more risk-free assets ’cause we killed them all? For shame…
By Joseph Cotterill and Lisa Pollack
Related links:
The AAA bubble – FT Alphaville
Fears rise over banks’ capital tinkering – FT