The Lectern: Sign of a government source-selection failure?

There was widespread dismay in the contracting community when the Government Accountability Office sustained Boeing's protest of the Air Force's award of the tanker contract to EADS/Northrup Grumman. Given the high visibility of this procurement, the general assumption was that the Air Force would clearly get the source selection right.

The sustainment of the protest produced a worry of the form, "If the government can't do a source selection that meets GAO muster on a procurement of this degree of visibility, how can we expect it to do source selection right on procurements that are of lower visibility?" The sustainment of the protest was presented in much of the media, such as The Washington Post, as a sign of government incompetence or of crisis in the procurement workforce.

Vern Edwards has written a very interesting and persuasive commentary in the new issue of the trade publication The Government Contractor, questioning this reasoning. Edwards is a prolific and intelligent commentator on contracting issues; I don't always agree with him, but he is always worth paying attention to.

Basically, Edwards' argument is that the source selection criteria for major weapons systems contracts are so voluminous and complex that it is virtually inevitable some mistakes will be made. The GAO found seven. Edwards' point -- although he doesn't put it in quite these words -- is that there were probably hundreds and hundreds of factors and subfactors that had to be evaluated in detailed
ways. Amidst the blizzard of paper the source selection process produced, a few problems didn't get caught.

In the past, source selection decisions on major weapons systems have seldom been protested. Protests on earlier source selection decisions, Edwards surmises, would have produced a similar pattern of needle in a haystack errors that probably also would have led to those decisions being overturned. It's not necessarily that government can't do source selection right -- it may be that nobody can do the kind of source selection used on major defense systems right. (It may be noted that one impact of 1990's procurement reforms was to move away from these massive, overly complex source selections in areas outside weapons systems -- so that source selection for IT has become simpler. Before the 1990's reforms, all government was subject to the problems Edwards describes for weapons systems.)

Edwards' lesson from all this makes sense. The source selection process itself is too complex. It is an accident waiting to happen.

Edwards thinks that, perhaps, given how close these tankers are to commercial items, the Air Force should have subsidized some development work by the two teams and just done a fly-off -- an approach, in my view, the government should consider more in an IT context as well.

One might also ask -- and I wonder how GAO would react to this suggestion -- whether GAO is asking for impossible standards of perfection from these complex source selections, the way the General Services Board of Contract Appeals used to do in the bad old days for IT procurements.