wisconsin_cur wrote:Swaps turn negativeSwap spreads turn negative
By Michael Mackenzie in New York
Published: October 23 2008 23:52 | Last updated: October 23 2008 23:52
The turmoil in the financial markets has taken hold of the strategically important trade in long-term interest rate derivatives, pushing rates to levels once thought to be a “mathematical impossibility”.
Such interest rate swaps are the most widely traded over-the-counter derivative and are crucially important for insurers, pension funds and other companies that need to fund liabilities decades into the future.
Investors use swaps to lock in interest rates for 30 years or more, trading a floating rate, based on the London interbank offered rate, for a fixed rate, typically based on US Treasury yields, plus a premium, called the swap spread, which reflects the risk of trading with a private counterparty as opposed to the government.
On Thursday, the 30-year swap spread turned negative after briefly flirting with such levels earlier this month. This implies investors are somehow reckoning that they are more likely to be paid back by a private counterparty than by the government.
I won't even pretend to understand exactly what this might mean or portend... but it sounds bad.
It is pretty simple really. Some swaps like options are done under ISDA standard documentation. Money market swaps including repos are done under ISMA standard documentation. Standard documentation means that you have two documents. One is your ISMA or ISDA that covers 99% of whatever will happen. It is based on past experience including previous crises. The confirmation on the other hand refers to the ISMA/ISDA, but lays out deal-specific details. It covers what I would call commercial risk versus legal risk 9 times out of ten.
As a default provision - the lawyers from dozens of investment banks sit around the table and hammer these argreements out - is LIBOR. The London Interbank Offered Rate that is published by the British Bankers Association (BBA). In normal times (at least up to now) this was a reliable indicator of where banks were lending to one another. They discarded the high and low offers, and then calculated the mathmetical mean of the rest of the submitted offers. This was fixed and published daily. As it was so reliable it found its way into standard agreements like ISMA/ISDA as this was something that lawyers could agree on.
So fast forward to this crisis. It breaks down because banks are reluctant to publish what appears to be a negative appraisal of where they are really getting interbank funding. Libor plus a spread that was getting wider. Worse, after Lehman, the lending disappeared, so in essence you had no reference rate. But thousands and thousands of contracts relied on some reference rate being published. Obviously if you are long or short it matters a great deal whether the rate accurately reflects actual deals being done or just a number out of thin air.
The fact is that banks - now - are quoting spreads based on where they can realistically get funding, but all the legacy contracts still refer to a rate that for all intents is meaningless. This is throwing all the mathmetical models off that calculate spreads and chances of defaults. We are talking about fat tails and events that statistically should not happen. There was no malintent. These standard agreements were negotiated in good faith with much, much open dialogue.
The reality is that the future does not always look like we expect it to. Stuff happens. No legal contract is ever 100% water tight. Certainly not a standard agreement that covers thousands of deals and is supposed to cover all eventualities. No agreement ever will. But one has to weigh the thousands of deals that were successfully transacted without any problem against the problems now when those terms may not reflect actual market conditions. It is like not giving out any home loans or mortgages - ever - because now we find out some were poorly written.
The organized state is a wonderful invention whereby everyone can live at someone else's expense.