As Zubin noted, the Journal has a piece today cataloguing the quick rebound in compensation at big Wall Street firms like Goldman and Morgan Stanley:

Based on analysts' earnings forecasts for 2009, Goldman Sachs Group Inc. is on track to pay out as much as $20 billion this year, or about $700,000 per employee. That would be nearly double the firm's $363,000 average last year, and slightly higher than the $661,000 for the average Goldman employee in fiscal 2007, according to analyst estimates reviewed by The Wall Street Journal.

Goldman's response, according to the paper:

"We've only accrued one quarter for compensation and benefits" for 2009, a Goldman spokesman said, noting that the 18% increase from a year earlier was "primarily due to higher revenues." He added that "compensation practices at Goldman Sachs remain fundamentally tied to the firm's performance."

I guess I don't see how that's a compelling defense. I'm not someone who begrudges Wall Street firms their ability to compensate employees well. But I do get exorcised when the structure of that compensation creates incentives to place reckless bets that taxpayers have to backstop. Which is to say, the problem with Goldman's likely payouts isn't that they're eye-poppingly high; it's that they appear to be based directly on short-term revenue, which is the opposite of what they should be based on if we want to eliminate those perverse incentives. And yet somehow the Goldman spokesman thought it was exculpatory to point out that the compensation is based on short-term revenue. Mystifying.

--Noam Scheiber