Saturday, January 19, 2008
credit default swaps
i've talked about it a little -- corporate defaults are rising, and promiscuous financing has probably created a backlog of weak companies that survived for years on the easiest credit terms ever seen, with only shallow economic washouts since the 1990 recession. with the credit crunch finishing that, it would not be surprising to see corporate defaults "surprise" to the upside. as jubak notes, 12% or even more is not unrealistic; banks are currently forecasting just 5% by the end of 2009; bill gross has noted the scale of the issue.
Bill Gross, chief investment officer at Allianz SE's Pacific Investment Management Co, or Pimco, recently told investors that if defaults in investment-grade and junk corporate bonds this year approach historical norms of 1.25% (versus a mere 0.5% in 2007), sellers of default insurance on such bonds could face losses of $250 billion on the contracts. That, he said, would equal the losses some expect in the subprime-mortgage arena.
With no central trade processing of credit-default swaps, defining trading-partner risks can be a Herculean task. Mr. Buffett learned the difficulty of unraveling such complex instruments in 2002 when he directed General Re Corp., a reinsurer that had been acquired by his Berkshire Hathaway Inc., to pull back from the business of these swaps and other derivatives. It took General Re four years to whittle the business from 23,218 contracts to 197 by the end of 2006.
Doing so involved tracking down hundreds of counterparties to General Re's trades, many of which Mr. Buffett and his colleagues had never heard of, he says, including a bank in Finland and a small loan company in Japan, to name just two. One contract, Mr. Buffett says, was designed to run for 100 years. "We lost over $400 million on contracts that were supposedly" safe and properly priced, "and we did it in a leisurely way in a benign market," Mr. Buffett says. "If we had to unwind it in one month, who knows what would have happened?"
as we saw around the time ACA became incredible, parties are going around trying to find duplicate insurance. the cost, however, has exploded and in the end insurance simply isn't going to be available for most CDS as defaults stack up just as there isn't sufficient fire insurance if 12% of the country burns down -- the sums involved are much too large.
if any event has the potential of turning a recession into a depression, this is it. if several large counterparties start welching en masse, risk aversion will go through the roof.
it's often said -- and rightly -- that derivatives are a zero-sum game, and for every winner there is a loser. but that rather misses the point in my view.
for every loser there is a winner -- until some losers can't pay. most banks hold both winners and losers, as any bank balance sheet's derivative breakdown will show you. most of these massive portfolios are constructed to balance risks on the presumption that contracts will pay out.
but if a bank's winners turn up to be uncollectable and worthless, even a responsible portfolio becomes a net loser -- and the bank then has to move to protect itself in an environment, once sanguine, that will resemble a war zone.
this is not a remote possibility. among large CDS writers are the bond insurers and structured-finance hedge funds -- some of whom have surely made common practice of using leverage to write vast volumes of CDS to collect premium on the presumption that they would never have to pay. the use of leverage ensures that small premiums result in big returns on equity, and the likelihood of bankrupting the fund can be -- through the lens of efficient market theory -- be wrongly seen as a "six-sigma event". in any case, the operators behind the funds can count on the event to be at least a few years off, which is more than enough time to make a ridiculous fortune thanks to the heavily skewed reward system that lies at the heart of american-style irresponsible capitalism. it's a classic example of "picking up nickels in front of steamrollers".
i agree with jubak and gross -- the mortgage bust is a depressing prospect in and of itself, more than capable of leaving banks capital-impaired for many years as was seen in japan from 1992 forward.
but the true barely conceivable disaster in all this is that the housing bust would provoke a consumer-led recession that forces corporate defaults in sufficient number to push several large CDS counterparties -- like ACA, ambac, MBIA but also invisible hedge funds -- to insolvency. that is an process which could trigger the kind of cascading bank failures that haven't been seen in the united states since 1933. money center banks simply don't have anything like the capital to sustain the probable losses that would emerge from an unbalancing of their derivative portfolio.