Kelman: History matters

In my most recent column, I discussed a fundamental challenge of managing the business relationship between the government and contractors. Mutual trust creates value in a relationship, so customer and supplier should act as partners. But that value also must be divided, which creates a conflict of interest between the two that can undermine the trust that helped create the value in the first place.

One basic strategy for dealing with that challenge is for customers to use past performance to influence decisions about their future dealings with the seller. If we decide on future dealings based on past conduct, we provide incentives for the supplier to create value through trust and not to exploit us as buyers in dividing up the value. If suppliers know that treating the customer well will create repeat business,  a supplier’s interests become aligned with those of the customer.

Before the procurement reforms of the 1990s, judgments about suppliers’ past performance played virtually no role in federal procurement because the judgments were considered too subjective and hence ripe for potential abuse. That was the central theme my 1990 book “Procurement and Public Management.” Past performance was a central part of procurement reform, highlighted both in a voluntary pledge by more than 20 agencies in 1994 to make past performance an important part of source selection in specifically enumerated contracts, and by provisions of the Federal Acquisition Streamlining Act the same year.

Since then, incorporation of past performance has made the system work better. This has particularly been the case where past performance is used informally in vehicles such as the General Services Administration schedules, where the ability to choose who can compete from a larger list of contract holders has meant that vendors with poor past performance don’t get asked to submit proposals.

Formal past-performance systems, with written report cards, have not worked as well as they should have, mainly because agencies have often proven unwilling to make distinctions between well- and poorly performing vendors. An important source of the problem is the regulatory provision allowing vendors unhappy with their grade to complain to a higher official.

That provision should be eliminated and replaced with one that allows a vendor to put its version of events into the contract file. I also believe that agencies, for major contracts, should have a third party — either an agency employee or a contracted interviewing firm — interview the program/contracting customer for past-performance report cards rather than having the customers fill them out themselves. Interviewers can be trained to elicit the necessary backup narrative and  encourage honest differentiation. Finally, for small purchases agencies should allow use of a simplified eBay/Amazon style rating system, specifically authorized by the Federal Acquisition Regulation, to allow quick judgments of vendor performance.

Revitalizing past performance should be a high priority for making the procurement system work better.

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