It’s not all golden at Goldman

August 6, 2009

It’s hard to find much to quibble about with Goldman Sachs’ second-quarter performance–at least from a dollars perspective.

You just don’t expect to find many landmines in a 10Q, when an investment firm manages to take in more than $100 million in revenues from trading on 46 days. 

But there was one little hiccup for Goldman buried in its most recent quarterly report and that involves its Level 3 treatment of derivatives. Level 3, of course, is the category banks use for assets deemed too hard-to-value and all but untradeable. And at the end of June, Goldman reported having $15 billion in derivatives contracts that were classified as Level 3.

Overall, untradeable derivatives accounted for 27% of the $54 billion in Level 3 assets at Goldman in quarter two. And on fair value basis, Level 3 derivatives accounted for nearly 7% of the dollar value of all of the firm’s derivative contracts.

And there’s reason to worry about those untradeable Level 3 derivative contracts at Goldman because right now those transactions aren’t fairing well. In the second quarter, Goldman reported a net unrealized loss of $1.4 billion on its Level 3 derivatives–blaming the loss on “tighter credit spreads on the underlying instruments.”

The unrealized loss is significant because it represents a reversal of what had been a positive trend at Goldman. 

In the first quarter, Goldman reported a net unrealized gain of $975 million on its Level 3 derivative contracts, largely due to “increases in commodities prices … and changes in credit spreads corroborated by trading activity.”  And for all of 2008, Goldman reported a net unrealized gain of $5.58 billion. In 2007, Level 3 derivatives posted an equally impressive net unrealized gain of $4.5 billion.

Back in October 2007, Peter Eavis, part of The Wall Street Journal’s Heard on the Street team, did an excellent story for Fortune on how unrealized gains from Level 3 derivatives were juicing Goldman’s results while the first cracks were beginning to appear at other Wall Street firms. In that story, Eavis speculated some of that paper gain might be due to Goldman’s “stupendously prescient bet against mortgages.”

Maybe the second-quarter turn to an unrealized loss on untradeable derivatives was an aberration and things will revert to the norm in the current quarter. Or maybe this is an unwinding of some of those subprime hedges that worked so well for the firm during the height of the housing meltdown.

Either way, it bears watching and makes you wonder what are the underlying assets in those troublesome Level 3 derivatives on Goldman’s books.


Comments are closed.