Kelman: Are we better off now?

The 1994 Senate Governmental Affairs Committee's report "Computer Chaos: Billions Wasted Buying Federal Computer Systems" laid the foundation for the Clinger-Cohen Act, drafted 10 years ago. To paraphrase Ronald Reagan, are we better off now than we were 10 years ago?

In some areas, the answer is clearly yes. The most obvious is in purchasing off-the-shelf computers and software. "Computer Chaos" highlighted two problems in particular: clunky procurement rules and the scourge of constant bid protests. Those problems dramatically hurt the government's ability to get up-to-date technology and good prices for commercial items.

Today the world is different. Thanks to blanket purchase agreements negotiated under the General Services Administration's schedules contracts, governmentwide contracts with easy technology refreshment for new products and enterprise licensing agreements (morphing, one hopes, into a governmentwide SmartBuy program), agencies get great prices and quick delivery and service on the latest equipment.

The government is also in better shape now on a large number of smaller, relatively low-visibility bread-and-butter information technology services projects. I believe this is due to heeding three of the report's recommendations: to use past performance in making contract awards; to move to smaller, modular projects rather than big-bang grand designs; and to make greater use of commercial rather than custom-designed software.

Evidence for those improvements comes from a survey two students of mine conducted a year ago. They examined 100 IT services projects that made use of the GSA schedules and found that average customer satisfaction with the vendors' performance on those projects was 9.3 on a 1-10 scale — a dramatic increase from an average satisfaction of 6.9 for IT projects when I asked the same questions while doing research in the late 1980s.

But in an important area that "Computer Chaos" highlighted, progress has been only minimal. We still have major problems with many of our large, complex IT system transformations, epitomized by ongoing problems with the Internal Revenue Service's tax system modernization. I believe the government is doing a little better — now agencies tend to have projects coming in late and expensive rather than not working at all. And, to be fair, similar projects frequently run into problems in the private sector as well. But we need to do much better.

The strategy for greater success involves heavy doses of boring but important management blocking and tackling. The government needs to get better at defining what it wants and stabilizing that baseline, establishing and managing according to performance measures, communicating with vendors and government users, adhering to project management principles and offering incentives for vendors.

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