Is past performance working better than we thought?

Does the government's past performance evaluation system help the government during post-award contract management in terms of performance and/or contract cost more than is often imagined?

I don't have nearly enough information to answer that question with any certainty, but over the past few months I have picked up a few pieces of information suggesting this may be true.

In January, while the Harvard Kennedy School had a group of GS-15's in town for a few weeks for an executive education program, I had a long lunch one day with four IT program managers, from both military and civilian agencies, to discuss post-award contract management. It definitely caught my attention when one of the four stated that mentioning to a contractor that some performance problem would threaten their past performance evaluation was generally enough to change the contractor's behavior. The three others around the table agreed.

I didn't think about this again until Dee Lee, in our conversation about moving from government to industry, discussed the same phenomenon, but in a negative way.

The threat of a bad past performance evaluation could be used by the government, she worried, to force industry to "bend" (her word choice) to the government's wishes even when what the government was asking was inequitable. Her specific example involved pressuring the contractor to accept no additional fee on a contract mod. When she raised this with me, I was suddenly reminded of something an industry person had told me last year to the same effect – past performance evaluations could illegitimately be used to deprive the contractor of a right or benefit to which the firm was entitled.

I am in no position to judge the justice of such complaints from industry, or to judge how often the government uses past performance in an illegitimate way. The interesting fact is the confluence of these stories -- keeping in mind that we have a very small sample size here, making it really hard to generalize -- around the idea that past performance is a meaningful tool for the government.

The paradox here is that many government folks involved in source selection state that past performance is seldom a differentiator. So how can it affect contractor behavior even without being a source selection differentiator? Before these recent conversations, I assumed (as I think most would) that past performance cannot be an effective tool unless it actually affects the outcome of future awards.

Having thought about it, I have now concluded this is not necessarily true. Let's say that a contractor changes its behavior on a current contract to avoid a bad past performance rating. When that changes, the firm then gets a good rating. If contractors in general act that way, the average past performance rating for contractors will go up. Some of the good ratings would have occurred without the government's threat, others only because of the government's threat.

Under this scenario, differentiation among past performance ratings goes down, and it thus becomes less likely that past performance will be a differentiator for future contract source selections. But even without any differentiation that can be applied for future source selection, contractors are doing a better job on current contracts to satisfy their government customer than if there were no past performance system. They have a clear incentive to improve their performance, even if it ends up not being a differentiator for future source selections. For if they had not changed their behavior, their rating would have been worse, and those poor marks might then have become a differentiator for future awards.