Governmentwide contracting: In praise of duplication and overlap

A long-awaited GAO report on governmentwide contracting vehicles is finally out, and the verdict is a mixture of good and bad news.

The good news is that the report, titled "Contracting Strategies: Data and Oversight Problems Hamper Opportunities to Leverage Value of Interagency and Enterprisewide Contracts," does not recommend eliminating governmentwide vehicles, such as those run by NASA and the National Institutes of Health, that compete with GSA’s governmentwide acquisition contracts. Given the hostility in parts of Congress toward competition among governmentwide contracts, that’s good news, especially given that the Office of Federal Procurement Policy needs to decide soon on renewing the authorities for some of these vehicles. It is also sensible to worry about whether the government is getting good enough prices on these contracts, about lack of easily available and comparable data on the contracting shops, and about potential risks from these vehicles that they might dilute the government's buying power.

The big flaw in the report, and in many of the commentaries on these vehicles, is conceptual. Both GAO and several members of Congress criticize competing governmentwide vehicles for being duplicative or overlapping -- having duplicative administrative costs, for example. This is a conceptually flawed argument. It is the argument that used to be made in the Soviet Union for why there should only be one brand of toothpaste made by a giant government monopoly -- no need for separate corporate headquarters, duplicative research efforts, wasteful marketing, etc. etc.

But that also means no forces of competition for customers, to drive down prices and encourage innovation.

GAO auditors are concerned that they have located different prices for the same item on different governmentwide contracts. That is far better than the likely alternative in a governmentwide monopoly -- the worst price being the only price.

Sure, there can indeed be a tradeoff between customers leveraging buying power and suppliers reducing administrative costs (which both argue for fewer contracts), and gaining the benefits of competition. Mergers among competing firms in the private sector sometimes make sense. But I would guess -- though this is an empirical question -- that the size of contract that is needed to leverage the government's buying power is less than one governmentwide contract for the whole government.

And, sure, like GAO I would love to see better data on what prices the government is actually getting, performance measures for these contracts associated with price and contractor performance, and transparency across vehicles so that customers could easily compare prices and terms/conditions. But, to repeat a tired but appropriate phrase, let's not throw out the baby with the bathwater. I am all for duplication and overlap -- a.k.a. vigorous competition across a limited but greater-than-one number of governmentwide contract vehicles among which agency customers can choose.