The world according to 803

For contracting cognoscenti, "803" refers to a provision in the fiscal 2001 Defense Department authorization bill regarding competition for orders under the General Services Administration's schedule contracts. Section 803 came about in reaction to reports that many awards were being made when only one bid was received. Congress set up procedures that, in effect, require DOD buyers to get three bids before placing orders for serv.ices under multiple-award contracts. DOD has just issued regulations implementing these requirements.

The changes are good for the government. GSA schedule and other governmentwide contracts provide a streamlined way for agencies to access quality vendors efficiently. The Education Department and the Defense Logistics Agency have conducted procurements that were quick, quality-oriented and brutally competitive using the schedules. It can be done.

Because competition became associated with endless procedures that often allowed a poor contractor to win by gaming the paper proposal process, many have failed to realize that the streamlined, quality-based, vigorous competition that multiple-award contracts allow is very much in the government's interest. Agencies should want to get three bids. Government should realize this keeps vendors on their toes.

It may be harder, however, if an order is a recompetition of work in which the incumbent vendor has performed well. Such vendors should have an advantage. For this reason, other vendors may be hesitant to bid. But an agency has an obligation to scour the marketplace to look for other bidders. Otherwise, the incumbent vendor may become lazy.

As important as the substance of those changes is how they are framed. Those who want to return to the Dark Ages of bureaucracy — the triumph of process over results — want to frame Section 803 as a swinging of the pendulum, in which a period of greater freedom is followed by a return to controls in a never-ending cycle in which no progress is made. If we accept this, it can become a self-fulfilling prophecy.

Any time an organization changes how it operates, we should expect problems as well as benefits from the first round of change. Indeed, students of change argue that it's important to get something out there, knowing it won't be perfect, so you can learn from successes and mistakes and make the new system better. In the world of organizations, this is known as "continuous improvement." In the world of technology, it's moving from Version 1.0 to 2.0. The alternative — a "grand design" approach to change that requires something to be studied to death to get the perfect system the first time — is a recipe for failure.

We should not frame Section 803 as a swing of the pendulum. Instead, we should reaffirm our belief that government is capable of improving over time.

Kelman is professor of public manage.ment at Harvard University's Kennedy School and former administrator of the Office of Federal Procurement Policy. He can be reached at steve_kelman@harvard.edu.