Wolfgang Munchau provides nice succinct overview of some of the recent debate surrounding a source of financial system risk that has suddenly captured the popular imagination: credit default swaps. For those new to the concept, credit default swaps are effectively insurance. A protection seller (think insurer) takes the risk of default on a reference entity (a corporation, an index) in return for periodic payments (think insurance premiums).
But unlike insurance, this market is unregulated, and is about to be tested for the first time in a serious way (single names, like Adelphi, have gone bankrupt, but the market has yet to experience the consequences of a downturn-induced deterioration in credit quality).
Munchau’s piece does not break new ground, but it makes a tidy, high level summary of key issues. From the Financial Times:
If this had been a mere subprime crisis, it would now be over. But it is not, and nor will it be over soon. The reason is that several other pockets of the credit market are also vulnerable. Credit cards are one such segment, similar in size to the subprime market. Another is credit default swaps, relatively modern financial instruments that allow bondholders to insure against default. Those who such sell such protection receive a quarterly premium, based on a percentage of the amount insured.
The CDS market is worth about $45,000bn (€30,500bn, £23,000bn). This is not an easy figure to imagine. It is more than three times the annual gross domestic product of the US. Economically, credit default swaps are insurance. But legally, they are not, which is why this market is largely unregulated.
Technically, they are swaps: two parties swap payments streams – one pays a regular premium for protection, the other pays up in case of default. At a time of low insolvency rates, many investors used to consider the selling of protection as a fairly risk-free way of generating a steady stream of income. But as insolvency rates go up, so will be the payment obligations under the CDS contracts. If insolvencies reach a certain level, one would expect some protection sellers to default on their obligations.
So the general health of this market crucially depends on the rate of insolvencies. This in turn depends on the economy. The US and Europe are the two largest CDS markets in the world. It is now widely recognised, including by the Federal Reserve, that the US economy is heading for a sharp downturn, possibly a recession. The eurozone, too, is heading for a downturn, but possibly not quite as sharp.
According to the National Bureau of Economic Research, the average length of US recession, excluding the 2001 recession, was 11 months. The 2001 recession was shorter, which brings down the average to about 10 months. The US has been quite lucky. Germany, for example, suffered a downturn at the beginning of this decade. It lasted a near eternity – 15 quarters – and included two separate technical recessions. Interestingly, and perhaps most relevant to today’s debate, is the fact that this downturn was also aggravated by a national credit squeeze. German banks cleaned up their balance sheets after a decade of binge-lending.
The German experience has taught us that persistent problems in financial transmission channels cause long economic downturns. Today, the really important question is not whether the US can avoid a sharp downturn. It probably cannot. Far more important is the question of how long such a downturn or recession will last. An optimistic scenario would be a short and shallow downturn. A second-best scenario would be for a sharp, but still short, recession.
A truly awful scenario would be a long recession. The US did experience some longish recessions in the past, for example from November 1973 until March 1975, but there was no CDS market around at the time.
So what then would be the effects of these scenarios on the CDS market? Bill Gross of Pimco*, who runs the world’s largest bond fund, last week produced an interesting back-of-the-envelope calculation that received widespread publicity. He projected that the losses from credit default swaps caused by a rise in bankruptcies could be $250bn or more – which would be similar to the expected total loss as a result of subprime.
This is how he arrived at this estimate. His calculation assumes that the corporate insolvency rate would return to a normal level of 1.25 per cent (measured as the default rate of all investment grade and junk debt outstanding). As the entire CDS market is worth about $45,000bn, $500bn in CDS insurance would be triggered under this assumption. The protection sellers would probably be able to recover some of this, so the net loss would come to about half of that. This estimate is very rough, of course. Most important, it is based on the assumption that the hypothetical US recession would not turn into a prolonged slump. In that case, one would expect corporate default rates not merely to return to trend, but to overshoot in the other direction.
So one could take that calculation as a starting point. A downturn lasting two years could easily trigger payments streams of a multiple of $250bn.
At this point we might be tempted to conclude that this all is irrelevant, since this is only insurance, which is a zero-sum financial game. The money is still there, only somebody else has got it. But in the light of the current liquidity conditions in financial markets, that would be a complacent view to take.
If protection sellers were to default en masse, so too could some protection buyers who erroneously assume that they are protected. Given that the CDS market is largely unregulated there is no guarantee of sufficient liquidity behind each contract.
It is not difficult at all to see how the CDS market has the potential to cause serious financial contagion. The subprime crisis came fairly close to destabilising the global financial system. A CDS crisis, under a pessimistic scenario, could produce a global financial meltdown.
This is not a prediction of what will happen, merely a contingent scenario. But it is contingent on an event – a nasty and long recession – that is not entirely improbable.