Goldman Sachs announced stellar earnings last night of $1.8 billion, or $3.39/share. The expected number was a mere $1.64/share, which GS more than doubled. The press went crazy, although the reported earnings were in line with the whisper. Goldman used the positive momentum to push through a $5B equity offering, the proceeds of which will be used to pay back the TARP funds to the government. Banking crisis? Hardly!
This morning, an interesting point came up in Goldman's earnings call: what happened to December? You see, a funny thing happened on the way to profitability. When Goldman became a bank holding company in an emergency move last September, their fiscal year changed from ending in November to ending in December. It was widely assumed that after reporting 4Q results for the three month period ending November 2008, Goldman would report 1Q results for a "four month quarter" spanning December 2008 - March 2009.
Turns out, they didn't.
Goldman's stellar 1Q results only reflect the three months from Jan 09 to March 09. Their 4Q results terminate in November 2008. Where is December 2008?
There it is! In a filing reported to the SEC this morning (and not blasted to the press, of course), Goldman revealed that they lost over $1B (pre-tax) in December on writedowns, particularly on commercial real estate. In EPS terms, that reflects a $2.15 loss. And that means that Goldman actually made only $1.24/share in their first true "quarter" as a bank holding company, well under the estimated 1Q profit.
The truly bizzare thing is that every graph, table, chart and diagram I've seen thus far wholly disregards the December loss. And that, of course, is what Goldman wants. The 4Q and 1Q numbers, as reported, effectively eliminate December - and it's massive loss - from any calculation. It's no accident that Goldman publicly announced the huge 1Q number 12 hours ahead of schedule, and then hid the December loss in an SEC filing early the next morning. It is all too convenient that Goldman had until March to decide just how to recognize that December loss -- perhaps some February writedowns got mixed in somehow? We won't know because, critically, December will not be discussed at length in any public disclosures. Why? Because it never happened.
I am sure that Goldman would make the argument that a 3-month quarter is appropriate because it allows an apples-to-apples comparison with last year's first quarter. Oh, and it conveniently erases a billion dollar loss.
Update: The NYTimes live-blogged the earnings call with a similar frame of mind.
Update: The WSJ is catching on.