What Price Performance?

If agencies can't come up with good performance measures, it may be because they haven't defined their desired results.

I

t's been more than a year since the Office of Management and Budget told agencies to boost their use of performance-based service contracting, an approach aimed at bringing private sector creativity to the federal marketplace. But many procurement professionals still are asking, "Can it really work in my area?"

The fundamentals aren't so difficult. You begin by defining the result you want, then identify the services required, the performance standards and a way to measure results. Then you devise incentives or disincentives to keep your contractor on track.

But a common complaint is that the right people aren't involved when the project gets started. Much of performance-based contracting is about defining expectations and ensuring everyone is on board with them. On the government side, you want subject matter experts, contracting personnel, program staff, the technical representatives, and, perhaps, legal and budget staff to be involved up front. Without them, it's impossible to come up with a basic, agreed-upon understanding of what the contractor is supposed to do. And if the government doesn't know what it wants, it's hard to see how the contractor will get it right.

International aid efforts, for example, require more than just a bilateral government-contractor relationship, according to William Reynolds of the Agency for International Development's office in Manila. Reynolds spoke about government contracting in the Philippines at an AID conference in March. The contracting and project officers for the host government as well as the U.S. government, he said, are in on the earliest discussions-when expectations are defined. Their willingness to participate and their understanding of performance-based service contracting techniques can make or break the effort.

Once you've determined the desired result for a project and expressed it in terms that everyone can understand, it's time to list all the services the contractor will perform. You must not leave anything out. If you do, and you're dealing with a fixed price contract, changes will cost the government a fortune. Nevertheless, agencies prefer fixed-price contracts, because they transfer the risk of cost overruns to the contractor. That only works, however, if the risk is manageable. Otherwise the government will likely not get the result it's looking for and the contractor will fail. If you are dealing with a lot of unknowns, as you might in the early stages of research and development on a new weapons system, then paying the contractor on a cost basis makes much more sense. In this case, you are paying the contractor for the effort it's expending, and not for a result that cannot be guaranteed.

A Health and Human Services agency seeking a contract for repairs to its elevators came up with a unique way to deal with excessive cost and technical risk. It needed a contractor for standard repairs and maintenance, and it wanted the work done on a fixed-price basis. In this scenario, if a major component in one of the older elevators failed, then the contractor would have to raise its price significantly to cover this contingency. So, the agency decided to take that risk, rather than transfer it to the contractor. The request for proposals stated that the cost of repairs for such a problem would be negotiated separately. That way, the contract could still be offered on a fixed-price basis, the competition could proceed and the agency didn't have to worry that an unseen repair would distort the price it would have to pay for the basic work.

Once an agency has identified the services it requires, the next step is setting performance standards against which the services will be measured. In some respects, this can be even more difficult than defining outcomes. A frequent problem is many agencies' inability to produce a solid baseline against which progress can be measured.

One Army facility wanted to shift a long-standing cost-type information technology support contract into a fixed-price performance-based agreement. Before Army officials could proceed, however, they needed to analyze their operations. It would have been hard to establish performance targets without defining what it was getting from its existing contract.

Carl DeMaio, director of the Performance Institute, a think tank based in Alexandria, Va., says the broader problem of agencies not being able to come up with good measures is that they haven't defined their desired results. "Agencies are finding that it is hard to draft and manage performance-based contracts with their vendors when their own programs don't have clear measures of results," he says. "Contracts serve programs, and if programs don't have clear performance goals, how can we possibly expect the contracts to have clear performance goals?"

Well-defined outcomes and solid performance standards won't mean much if agencies haven't developed sound quality assurance and surveillance plans to measure success. Even with a plan, agencies must have a knowledgeable and capable technical staff to monitor a contractor's progress. Oversight is a different role for employees who are accustomed to being doers, not managers. Changing that culture might be the toughest challenge in making performance-based contracting work.


Allan V. Burman, a former Office of Federal Procurement Policy administrator, is president of Jefferson Solutions in Washington. Contact him at aburman@govexec.com.

NEXT STORY: Big Plans for Small Business