As spreads of all colours blow out due to the perpetually unresolved sovereign crisis in Europe, FT Alphaville has been wondering what non-fundamental factors are driving these moves. The bond market is in some places broken and in other places potentially being driven by regulation.
To the extent that the market for credit default swaps influences the bond market, we ponder the technicals of these derivatives that reference it. Here we look at the role played by trades directly between banks and sovereigns.
The sovereign CDS feedback loop
Previously we had concluded that the counterparty risk management (i.e. “CVA” desks) at banks serve to push CDS spreads ever wider as spread-widening is taken to mean increased riskiness of a sovereign (as a counterparty to a trade with a bank), leading to buying pressure on sovereign CDS as the bank hedges itself against the sovereign, leading to spreads widening out further and so on. Basel III, when it kicks in, is only going to bake this in further as volatility serves as an input to the CVA recipe too.
So are sovereigns big counterparties to banks? If so, are banks in-the-money or out-of-the-money to them?
Collateral? We never!
One of the reasons this matters so much is that sovereigns do not post collateral on their trades. If the sovereign owes the bank money, i.e. the mark-to-market of the bank is positive, then no collateral goes to the bank. If, on the other hand, the sovereign is owed money by the bank, then the bank does have to post collateral. This is the dreaded one-way credit support annex (CSA). Such agreements are usually two-way with both parties posting collateral in the name of reducing counterparty risk to each other.
There are a couple of exceptions. Portugal and Ireland do post collateral, but that’s a fairly recent innovation. Other sovereigns are resisting this while banks have argued that by not posting collateral, the sovereigns are only serving to push up the cost of the services the banks offer them, as well as their own CDS spreads, which again may influence bond prices.
But again, how big of an issue is this anyway? Are there many derivatives between sovereigns and banks?
Ok, how big is it?
FT Alphaville was busy being hung up on by banks who didn’t want to discuss their derivatives exposures to sovereigns with us when Christopher Whittall of IFR came along to remind us that such figures were available in the European bank stress tests. (We seriously owe Chris a beer.)
So among the various types of sovereign exposures that the results of the tests outlined, “direct sovereign exposures in derivatives” was listed, that being the “net position at fair values (Derivatives with positive fair value + Derivatives with negative fair value)”.
Risk did a story at the time about select exposures at the time and even outed BCPE for misreporting its exposure to France by a mere €3.4bn (they got notionals and mark-to-market confused).
It’s the positive values that we are most interested in, since those are the cases where the banks are “owed” money by the sovereign. This is in the mark-to-market sense. That is to say that this isn’t necessarily that the banks have lost money on the trades, it’s more about what they stood to lose in the extreme event that the sovereign couldn’t or wouldn’t pay. Pretty standard risk management in a distinctly less risk-free world.
Here’s our table of some of the biggest (positive and negative) exposures and the banks that hold them (click to expand):
A reminder that the data comes with an as-of date of December 31st, 2010.
The unlucky winners
At the bottom of the table we’ve sum-iffed in order to look at the quantity of positive exposures, i.e. the sum of how in-the-money the banks listed are. For your viewing pleasure, we’ve also put in the CDS spreads of the sovereigns as of the end of last year and yesterday. We also wanted to show off our conditional formatting skillz by highlighting the positive exposures over €500m in red and those over €250m in yellow.
Banks are indeed in-the-money to Italy — to the tune of €5,119m. Dexia, Deutsche Bank, and BNP Paribas have the largest exposures. To give a sense of scale compared to CDS outstanding (hence how much this amount can drive the market), the current net notional outstanding that references Italy is $20,530m, which is €15,201m at today’s rates.
Interestingly, there are also large positive marks to banks against Germany. Dear readers, any ideas what Germany is managing with this? With the Italian sovereign, as we’ve stated before, we think the two main types of derivative that Italy is entering into are swaps to get variable rate debt into fixed and foreign exchange swaps to turn their foreign currency bonds into euros.
The table clearly shows significant home bias in Germany, France and Greece.
CDS go boom
Now, we’d love to have another snapshot of the value of these derivatives to know whether they’ve increased or decreased in value. What we do know, however, is that CDS spreads have blown out. To the extent that banks are actively managing their direct sovereign exposure, they will be in the market as protection buyers, hence pushing spreads out further.
There are, of course, ways to hedge other than by using credit derivatives. Banks could look at the overall allocation of risk to a given sovereign and decide to reduce it by dumping the bonds of the sovereign. Particularly appealing if said bonds attract a larger haircut than they used to anyway. Or a bank could decrease their overall exposure to the country by reducing loans there… we think we hear a sucking sound where sovereign exposures used to be.
Related links:
Peripheral exposures: mind the cash – FT Alphaville
It’s a capital ratio of two halves - FT Alphaville
Banks to dump more Italian debt - IFR