Nothing lies like statistics

The percentage of government contract dollars going to small businesses has been remarkably stable during the past decade despite reports to the contrary. A series of tortured attempts have been made to extract from the data a story of small-business woes that the numbers simply don't support.

The latest such "finding" is the seemingly dramatic revelation by Angela Styles, administrator of the Office of Federal Procurement Policy, that the number of new contract actions totaling more than $25,000 going to small business declined from 51,059 in fiscal 1992 to 20,736 in fiscal 2001 — and dollars also fell precipitously from $12.08 billion to $8.7 billion during that same period.

These numbers have been used to justify anti-taxpayer proposals on contract bundling, which combines smaller contracts into a single, larger one. The numbers are particularly unfortunate in the context of the wise efforts by the Office of Management and Budget's associate director for information technology and e-government, Mark Forman, to consolidate redundant buying. Furthermore, they are misguidedly used as a basis for the claim that the federal marketplace is growing less competitive, something that comes as news to any federal IT vendor.

The issue with these misleading numbers lies hidden in two little words: "new" and "contract." As contracting connoisseurs are aware, purchases from the General Services Administration's schedule contracts and work under task-order contracts are not considered contracts but instead are classified as "orders." Therefore, they are not included when the computer is asked to generate statistics on "contracts" (although they are included in procurement spending totals). And "new" excludes the exercise of options under existing contracts.

During the past 10 years, schedule- and task-order contracting have grown substantially, and contracts are often spread over many more years, meaning more options are exercised. (Ever hear a genuinely small business complain about the steadier cash flow coming from longer contracts?) As a result, the total number of "new contract actions" to large businesses, as well as small, has shrunk dramatically from 84,572 in 1992 to 34,621 in 2001.

However, when one looks at the percentage of new contract actions going to small businesses, the error of Styles' claim becomes apparent. Except for an aberrant decline in fiscal 2000, percentages have been stable since 1992, just like the overall small-business numbers. In 1992, small businesses received 60.4 percent of new contract actions and 29.7 percent of dollars — in 2001 the percentages were 60.5 and 33.1.

There may be some genuine problems, especially in IT, for small businesses. We as a community should deal with them constructively. Scare-mongering doesn't help that effort.

Kelman is 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.