Myth that could undermine credit derivatives
Saturday, 14 July 2007
Tony Jackson
There is a rather dispiriting resemblance between the latest credit crunch and the bursting of the dotcom bubble.
Ahead of each, there were plain warnings not just of what would happen but how and why. All that was missing was when.
But there is one fresh angle this time. The market has been struck by the disturbing notion that credit derivatives might not be worth what they were supposed to be.
This should hardly be news to anyone who read Warren Buffett's broadside against derivatives four years ago. There is no market, he pointed out, for complex derivatives. So instead of being marked to market, they are marked to model - or in some cases, marked to myth.
Suppose, he said, you write a contract specifying the number of twins that will be born in Nebraska in 2020. That will be taken up by a counterparty, in the usual way. Both of you might then devise different models, which showed you both making a profit for years. And why might you do that? Because, as derivatives traders, you get a bonus based on that profit.
The scope for this varies by derivative type. Much depends on whether the article on which the derivative is based is exchange-traded - that is, whether there is a fundamental market price.
Thus, currency or interest-rate derivatives are at least priced off a market rate. The same is true in principle of equity derivatives. But the practice may be very different. I was told last week by a former head of equity derivatives at a bulge-bracket bank that, as a manager, he had worked mostly in the dark.
The products devised by his dealers involved such abstruse calculations that he had to take them on trust.
Credit derivatives, though, are something else again. Loans are not traded and no two are alike. Even corporate bonds - in Europe anyway - are too thinly traded for the exchange price to be reliable.
Instead, the investment banks use the price at which they sell them to each other - which is not quite the same thing.
So credit derivatives are marked to model - or myth, depending on your point of view. And when that goes wrong, hedge funds and others hit trouble. Some will refer complacently to last year's Amaranth collapse, where a $6bn (£3bn) loss was smoothly absorbed by the market. But Amaranth lost its shirt on natural gas futures - a highly visible and widely traded instrument. These positions could be bought out with confidence by its rivals.
Contrast the Bear Stearns case, which triggered the latest mini-crisis. The other banks that inherited the subprime derivatives in question have held off selling them, precisely because they risk crystallising a much lower market price - which would then apply across the board.
Recall, too, that Amaranth was able partly to offset its losses by selling a $1.3bn portfolio of leveraged loans. These were greatly in demand at the time - for packaging into credit derivatives. It might be different today.
At this point, it is worth recalling what credit derivatives were originally for.
In the early days, banks would issue a loan to a customer, then buy credit protection on that loan in the form of a credit default swap, or CDS.
That protection would often be supplied, appropriately enough, by an insurance company. Insurers make their money from assuming risks. And the more diverse types of risk they take on, the safer they will be themselves.
But all that changes when we come to collateralised debt obligations (CDOs). In this case, the CDS is hoovered up by an investment bank, which jumbles it together with loans and bonds, then spits out a CDO in sliced-up form to investors.
This raises a serious question of whether credit derivatives are a suitable asset class for portfolio investors at all. For a start, they give hedge funds scope to book mythical profit, of which they can then take their 20 per cent cut.
So if my pension fund is invested in the hedge fund, I am down on the deal.
Or suppose the pension fund buys credit derivatives directly, then tells me it is fully funded. What does that statement mean?
All this suggests that once the markets have been purged of their excessive appetite for risk, the explosive growth in credit derivatives may be a thing of the past.
They will still be in demand for their original function of hedging risk. They may be less so as a means of blindly assuming risk in the hunt for yield.
As to where we are in that process, I cannot say. When the first spasm of risk aversion hit four months ago, I wrote that the repricing of risk would probably proceed by steps, rather than all at once.
That still seems the most likely picture. Recall that just before this latest episode kicked off, the cost of insuring risk through credit derivatives had dropped back to an all-time low. That might happen again. But not for long, I fancy.
There is a rather dispiriting resemblance between the latest credit crunch and the bursting of the dotcom bubble.
Ahead of each, there were plain warnings not just of what would happen but how and why. All that was missing was when.
But there is one fresh angle this time. The market has been struck by the disturbing notion that credit derivatives might not be worth what they were supposed to be.
This should hardly be news to anyone who read Warren Buffett's broadside against derivatives four years ago. There is no market, he pointed out, for complex derivatives. So instead of being marked to market, they are marked to model - or in some cases, marked to myth.
Suppose, he said, you write a contract specifying the number of twins that will be born in Nebraska in 2020. That will be taken up by a counterparty, in the usual way. Both of you might then devise different models, which showed you both making a profit for years. And why might you do that? Because, as derivatives traders, you get a bonus based on that profit.
The scope for this varies by derivative type. Much depends on whether the article on which the derivative is based is exchange-traded - that is, whether there is a fundamental market price.
Thus, currency or interest-rate derivatives are at least priced off a market rate. The same is true in principle of equity derivatives. But the practice may be very different. I was told last week by a former head of equity derivatives at a bulge-bracket bank that, as a manager, he had worked mostly in the dark.
The products devised by his dealers involved such abstruse calculations that he had to take them on trust.
Credit derivatives, though, are something else again. Loans are not traded and no two are alike. Even corporate bonds - in Europe anyway - are too thinly traded for the exchange price to be reliable.
Instead, the investment banks use the price at which they sell them to each other - which is not quite the same thing.
So credit derivatives are marked to model - or myth, depending on your point of view. And when that goes wrong, hedge funds and others hit trouble. Some will refer complacently to last year's Amaranth collapse, where a $6bn (£3bn) loss was smoothly absorbed by the market. But Amaranth lost its shirt on natural gas futures - a highly visible and widely traded instrument. These positions could be bought out with confidence by its rivals.
Contrast the Bear Stearns case, which triggered the latest mini-crisis. The other banks that inherited the subprime derivatives in question have held off selling them, precisely because they risk crystallising a much lower market price - which would then apply across the board.
Recall, too, that Amaranth was able partly to offset its losses by selling a $1.3bn portfolio of leveraged loans. These were greatly in demand at the time - for packaging into credit derivatives. It might be different today.
At this point, it is worth recalling what credit derivatives were originally for.
In the early days, banks would issue a loan to a customer, then buy credit protection on that loan in the form of a credit default swap, or CDS.
That protection would often be supplied, appropriately enough, by an insurance company. Insurers make their money from assuming risks. And the more diverse types of risk they take on, the safer they will be themselves.
But all that changes when we come to collateralised debt obligations (CDOs). In this case, the CDS is hoovered up by an investment bank, which jumbles it together with loans and bonds, then spits out a CDO in sliced-up form to investors.
This raises a serious question of whether credit derivatives are a suitable asset class for portfolio investors at all. For a start, they give hedge funds scope to book mythical profit, of which they can then take their 20 per cent cut.
So if my pension fund is invested in the hedge fund, I am down on the deal.
Or suppose the pension fund buys credit derivatives directly, then tells me it is fully funded. What does that statement mean?
All this suggests that once the markets have been purged of their excessive appetite for risk, the explosive growth in credit derivatives may be a thing of the past.
They will still be in demand for their original function of hedging risk. They may be less so as a means of blindly assuming risk in the hunt for yield.
As to where we are in that process, I cannot say. When the first spasm of risk aversion hit four months ago, I wrote that the repricing of risk would probably proceed by steps, rather than all at once.
That still seems the most likely picture. Recall that just before this latest episode kicked off, the cost of insuring risk through credit derivatives had dropped back to an all-time low. That might happen again. But not for long, I fancy.