Performance-based contracting out of reach

Performance-based contracting for information technology in the federal government - contracting for results to which a vendor commits when it signs a contract - has a long way to go.

Although the government is making progress in asking for specific performance-oriented results, the typical request for proposals that I see for a contract or a task order still does not get very specific about the results the agency demands from vendors.

The good news is that the government has moved away from giving vendors design specifications or telling them exactly how it wants the work done. Unfortunately, those bad old days generally have not been replaced by results-oriented performance standards.

Instead, they have been replaced by descriptions of the problem the government faces and the bidding vendors' "approach" to solving it. An environment in which the government does not demand performance commitments is one where, by default, competitions gravitate toward cost shootouts or essay-writing contests. (The government's increasing use of past performance in source selection counteracts those tendencies somewhat.)

More modular-style task-order instruments and the spread of due diligence - giving bidding vendors a real opportunity to learn more about the agency's relevant business processes before submitting a bid - open the way for performance-based contracting. They do so by giving vendors more manageable chunks of work to bid on where it's more feasible to make performance commitments and by enabling them to learn more about the agency's environment, which reduces the risk of making commitments.

But the government is still not taking advantage of these possibilities. There is even some danger, as Chip Mather, senior vice president of Acquisition Solutions Inc., puts it, in borrowing the Defense Department's distinction that acquisition (which includes defining requirements) is broader than just procurement - that procurement reform can hurt acquisition reform. That is, the mantra of "speed above everything" provides an excuse for some in government not to fully consider what they're trying to buy.

That's in contrast to the desirable situation in which procurement reform serves acquisition reform by freeing program managers from needless bureaucracy,

giving more time to spend understanding the substance of what they're buying.

A historic watershed for the development of performance-based IT contracting occurred in the past few weeks. But it has received no attention up until now. In its winning bid for a contract to serve as "modernization partner" to the Education Department's Office of Student Financial Assistance, which runs the college student loan program, Andersen Consulting made three specific commitments to the department for results to which they would contractually commit if they won the job. In its bid, for which Andersen paid me as a consultant, Andersen committed to the following:

Annual savings of $100 million in the loan origination process (that is, the process of signing students up for their loans when they come into the system) through business process re-engineering and better use of IT.

One-time savings of $300 million to $700 million to be recovered through improved debt collection.

Eighty-five percent to 90 percent customer satisfaction ratings for Education customers, who, I believe, are students and school financial officers.

Andersen also offered to negotiate specifics with Education so that if the company does better than what it committed to, Andersen would be paid extra. If Andersen does not meet those commitments, they would be penalized. That is the essence of performance-based contracting.

Andersen also expressed an interest in using innovative incentive techniques - such as bonuses or penalties associated with specific completion dates for having systems up and running, share-in-savings contracting and per transaction contracting - to encourage meeting the results the parties agree to. My understanding is that all these commitments played an important role in Andersen's contract win.

To my knowledge (and I hope I'm wrong and others can point out when this has been done), this is the first time an IT vendor during a competition has gone beyond the requirements of the RFP and on its own initiative proposed specific results-oriented performance commitments. I believe these commitments made history. I will be right if others follow Andersen's lead.

Kelman was the administrator of the Office of Federal Procurement Policy from 1993 to 1997. He is now Weatherhead Professor of Public Management at Harvard's Kennedy School of Government.