Three ways to revitalize the use of past-performance data
I have been attending the annual acquisition research conference at the Naval Postgraduate School in Monterey, Calif., and spoke on the first-day morning plenary on ways to improve the acquisition system. Judging from comments I have received since during breaks and meals, the part of my remarks that attracted the most interest was my suggestions for making past performance a more meaningful part of contracting decisions.
The introduction during the 1990s of past performance as a criterion for making contract awards improved the way the procurement system works. But the system hasn’t lived up to its potential. A warning sign is that it works best when it is most informal and least-bureaucratic. A few years ago two of my students examined a random sample of 100 IT services task orders awarded under the General Services Administration schedule contract. They found an enormous improvement in customer satisfaction with contractor performance since the research I had done in the late 1980s – from 6.8 to 9.3 (on a scale of 1 to 10).
This was due, in my view, to the ability of the schedules to work as a great informal past-performance system. With agencies able to choose three schedule vendors among whom to organize a competition from among the many on the schedules, they didn’t choose anyone with poor past performance (unfortunately, there are people who are constantly trying to inhibit agency ability to use such an informal past performance system). By contrast, the more that past performance is based on the government’s formal past performance databases, the less of a differentiator it is – because there isn’t enough variance in the scores contractors receive on their report cards.
I have a number of reforms in mind.
One is to eliminate the strongest disincentive for government folks to provide honest report card ratings – the provision in the FAR that allows a contractor unsatisfied with their rating to appeal to a higher level in the organization. This means that a contracting official who gives a bad rating is asking to spend hundreds of hours of work to justify their decision. Replace this with a provision that allows contractors to place their version of events in the file, but eliminates the appeal.
Second, on important contracts have report cards compiled using oral interviews by interviewers trained in eliciting variation in answers rather than plain vanilla ratings.
Third, as part of a larger reform (which I will discuss in my next blog) in how vendor invoices are processed, add a simple 5-point scale (“excellent” to “poor”) that program officials can use to rate the work for which the invoice was submitted. This would provide the contractor with ongoing, real-time performance feedback.
Can the Office of Federal Procurement Policy take up the cudgels on these reforms to revitalize past performance? Judging from comments I have received in Monterey, lots of people would like to see this happen.
Posted on May 13, 2010 at 7:26 PM