The following op-ed by Harvard law School Professor Lucian Bebchuk LL.M. ’81 S.J.D. ’84, “Paying for performance at Goldman,” appeared in the July 24, 2009, edition of the Wall Street Journal.

Last week, after reporting stellar second-quarter profit of $3.4 billion, Goldman announced the setting aside of $11.4 billion for compensation – which, broken down per employee, is similar to what Goldman set aside in the first half of the boom year of 2007.

Goldman’s CFO argued that its pay decisions reflect the firm’s “pay for performance culture.” However, if Goldman proceeds to pay record cash bonuses this year, as many now expect, these payments would reflect a return to flawed pay structures, as well as a failure to implement effectively the compensation principles Goldman recently put forward.

The setting aside of $11.4 billion for compensation, it should be stressed, doesn’t yet commit Goldman to any amounts of cash bonuses. Goldman still has time to determine the magnitude and structure of its 2009 compensation. In doing so, it should give substantial weight to lessons drawn from the financial crisis.

The crisis has highlighted a substantial flaw in compensation structures that provide rewards for short-term performance – which is what Goldman’s paying super cash bonuses for 2009 would do. Such rewards can over-compensate executives as well as produce excessive incentives to take risks.

Rewards for short-term results can produce over-compensation by enabling executives to cash out large amounts of compensation on account of results that are subsequently reversed. In many financial firms whose aggregate earnings over the past several years are negative, executives have still been able to cash out large amounts of bonus compensation during the first part of this period – and they kept these amounts despite the large losses subsequently borne by the firms.

In addition, and perhaps most importantly, bonuses for short-term results provide incentives to seek improvements in short-term results even at the expense of excessive taking of risks of an implosion later on. The short-term distortion caused by standard compensation structures, which Jesse Fried and I first highlighted in our “Pay without Performance” book, has recently become widely accepted. Treasury Secretary Geithner stated last month that “[s]ome of the decisions that contributed to this crisis occurred when people were able to earn immediate gains without their compensation reflecting the long-term risks they were taking for their companies and their shareholders.”

Indeed, the flaws in the standard compensation structures of financial firms have been explicitly recognized by Goldman’s own leaders. Last April, in a widely praised speech before the Council of Institutional Investors, Goldman’s CEO Lloyd Blankfein called for compensation reform, stating that “[financial firms’] decisions on compensation … look self-serving and greedy in hindsight.” Evaluation of employees’ performance, Blankfein stressed, “must be made on a multi-year basis to get a fuller picture of the effect of an individual’s decisions.”

Goldman subsequently adopted compensation principles and announced them in its annual shareholder meeting last May. According to these principles, “cash compensation in a single year should not be so much as to overwhelm the value ascribed to longer term stock incentives that can only be realized through longer term responsible behavior.”

Paying record bonus compensation as a reward for 2009’s results would, at best, not sit well with Goldman’s recently adopted compensation principles and its leaders’ calls for compensation reform. To the extent that employees receiving cash bonuses can keep them regardless of subsequent results, recipients will end up generously paid even if the firm bears next year’s losses that fully wipe out 2009’s gains. And using a large fraction of the earmarked $11.4 billion for cash bonuses could “overwhelm” long-term considerations in the minds of many employees, providing them with incentives to focus on short-term profits.

To avoid the recognized flaws of past practices, Goldman should spend most of the compensation awards for senior executives in 2009 on equity incentives subject to long-term restrictions on executives’ ability to cash them out. And any amounts paid in bonus compensation should be based on multi-year results rather than performance in a single year. An examination of Goldman’s aggregate profits during the 2008-2009 period, say, provides a rather different picture of performance than an assessment based on 2009 alone.

At a minimum, even if some bonuses are based on 2009 results alone, Goldman would do well to prevent the immediate cashing out of these bonuses. Rather, following the examples of such financial firms as UBS and Credit Suisse, any 2009-based amounts should be parked in a company account for an extended period of time and adjusted downward if subsequent information indicates that the basis for the bonuses no longer holds up.

Notwithstanding its recent stellar short-term profits, Goldman should resist temptations to enable its employees to cash out large amounts of compensation on account of these profits. Doing so is necessary to avoid the risk that, down the road, Goldman’s 2009 compensation decisions would “look self-serving and greedy in hindsight.”