The International Swaps and Derivatives Association has been rather busy lately. In case you haven’t noticed, they’ve been on the public relations offensive, taking notes and calling out what they see as bad reporting, or even bad semantics.
It’s this report, which covers the sorry tale of $54bn of losses, that FT Alphaville would like to discuss with you. We think you’ll find it says a lot about credit derivatives history… and about Isda.
But first, to get a better sense of just how vocal Isda has gotten, here’s their advice on what to do if you don’t want to exchange your Greek bonds under the “voluntary” swap — the voluntary nature of which means that CDS contracts that reference the sovereign are unlikely to have to pay out (if the swap stays in its current form). An excerpt from their post entitled “The First Rule About CDS: Don’t Talk About CDS (Unless You’ve Read the Contract)”:
The fact remains, though, that the exchange is not binding on all debt holders. If you don’t like the deal, don’t exchange your bonds. Hold onto them. Whether you take the deal or not, you will keep your CDS. Collect the payments on the bond as long as they are being made. If a payment is missed, trigger the CDS and be made whole. Users of these products know the drill.
It’s funny, but we’re not sure that so many “users of these products” thought that “the drill” would involve so much political risk. Not to mention the fundamental reshaping of the eurozone as we know it. But sure, maybe the holders of Greece CDS weren’t purely seeking capital relief and an accounting hedge; maybe they know all about this “drill”.
Right, exposures to monolines! So the quick background is that the Office of the Comptroller of the Currency (OCC) in the US publishes regular quarterly reports on the trading and derivatives activities of banks. Isda did a study, released in August, using those reports and SEC filings to try to ascertain the amount of losses banks took on their exposures to monolines from 2008 to 2010. The focus was on credit mitigation techniques.
However, the study defined “banks” rather narrowly, the result being that when Isda did a second study (released Tuesday) with a broader set of entities, an extra $51.3bn of losses showed up. The original study had just $2.7bn in losses.
Study #1 had conclusions like this:
The banks themselves were not excessively involved with toxic mortgage products in derivative form.
But here “bank” meant: a bank under the OCC’s remit excluding non-bank affiliates of US banks, international banks and entities that were not banks when the losses were taken. Our guess is that Isda will argue that it was obvious to all, and clearly stated, that their first study was only ever intended to cover a narrow set of US banking entities. But one could argue against that by saying that those same banks have something of a crack habit when it comes to supporting their subsidiaries (and even their supposedly bankruptcy-remote ones). And let’s not start on how Isda defines “excessive”…
The monoline exposures arose in the first place because of the banks’ activities whereby they packaged subprime mortgages to sell to clients. Sometimes these securities were constructed by means of actual mortgages (“cash”) and sometimes they were just derivatives of those cash products (“synthetic”).
In many instances, as a result of their structuring of various securities, banks ended up holding large amount of exposure to subprime securities on their balance sheet — the afterbirth of the products they had created. No one wanted these so-called “super seniors”. Insuring these exposures, which the banks would otherwise be long, with monolines meant that exposures were effectively off balance sheet.
Everyone knows the punchline: when push came to shove and subprime started tanking, the monolines had too little capital to make payouts on all the insurance they’d sold.
Isda explains all of this very readable detail, with gems such as this (when explaining the further repackaging of subprime exposures):
(this was the financial world’s version of alchemy!)
FT Alphaville wishes to genuinely (don’t laugh, we have our game face on) applaud the organisation for Study #2. It’s a great read and has some very interesting tables:
We’ve wanted a table like this ever since we can remember and they must have had to go through a lot of SEC filings to compile it. They even included foreign banks in order to give a more complete picture. Thank you, Isda!!
The deal with the tables is that banks had to write off a lot of the value (“CVA charges” in the above) of the insurance contracts with the monolines once it became clear that they would be unable to pay out in full. The banks did get some payouts, though, not least because of the US government’s bailout of AIG.
That $54bn is a large amount of losses and Isda goes to lengths to explain that what we’re seeing is in part the consequence of what happens when you don’t collateralise your exposure. The monolines typically only started posting margin after severe downgrades and it was often too late by then anyway. And, as Isda points out, they were doomed by the sheer quantity of the notional on which they had written contracts. And they ask why wasn’t anyone at the banks modelling the loss given default or what would happen to the monolines if they had a wave of collateral calls?
But Isda also wants you to know how much bigger the mortgage losses that banks experienced were (hence the last column in the above table). In other words: well, look, derivatives on the crap caused some additional losses, but why don’t you go blame the crap rather than the derivatives? Stop being so mean about the derivatives, alright?!
Answer: cause it’s hard to justify the crap, and even harder to figure out why there should have been derivatives on it. That’s mostly why…
CDS demonization watch, ISDA vs Morgenson edition – Felix Salmon, Reuters
Things That May Not Blow Up The World: Derivatives? – Dealbreaker
Monoline Exposures Resulted in $54 Billion in Charges for Banks, According to New ISDA Study – MarketWatch