SBA study shows perils of analyzing small business contracts

Federal small business contracting is never as simple as it seems. That was the lesson the Small Business Administration's Office of Advocacy learned with the publication of a new report on how small firms fare in outsourcing contests.

On Friday, SBA released a new analysis of small business winnings in competitive sourcing, the administration's initiative to boost the rate at which companies compete to take on government work.

Working with two different data sources -- civilian agency information in the Federal Procurement Data System-Next Generation, or FPDS-NG, and Defense records in the Commercial Activities Management Information System, or CAMIS -- the report assessed the rate at which small businesses prevailed in winning contracts stemming from outsourcing under the Office of Management and Budget's Circular A-76 rules.

First reviewers had to insert the caveat that FPDS-NG data from before fiscal 2004 could not be relied upon. "The new, comprehensive FPDS-NG reporting system renders data collected under the old system incomplete and unreliable," the authors warned.

"It is meaningless to compare absolute levels of contract spending in FY 2003 and earlier years to spending in FY 2004 and FY 2005," the authors said, citing a fiscal 2004 change to the dollar value threshold for reporting A-76 contracts. "The analysis of trends in various market shares over time remains relevant, though, because selected and overall data have changed by comparable percentages," they said.

Experts inside and outside government say that even the newer FPDS-NG system is flawed, as contracting officers have little incentive to record information correctly and, in some cases, data entry fields make it difficult or impossible to do so. In September, OMB Deputy Director for Management Clay Johnson promised that a new computer system for viewing federal contracts going live in 2008 would reflect accurate data.

For the SBA Office of Advocacy study, number-crunchers looked at what data there was for civilian agencies in FPDS-NG for the period from fiscal 2001 through part of fiscal 2006, and concluded that about two-thirds of A-76 contracts won by the private sector had been awarded to small firms, while those awards represented about one in five contract dollars.

For Defense, the CAMIS system reflected a slightly different study period that started in fiscal 2002, but showed a similar ratio, with small firms winning just less than two-thirds of A-76 contracts to take home nearly one third of their total dollar value.

The authors drew few conclusions from these statistics, but in a limited analysis of work converted directly to private performance without a lengthy competition process, they said similar figures suggest "that DoD relies on small firms for small outsourcing jobs and large firms for large jobs."

But the results get murky when you look at the total number of A-76 contracts. SBA's report indicated there were 3,735 civilian A-76 contracts awarded since 2001. But an early May OMB report on the government's A-76 activities said agencies have completed only 1,243 competitions between fiscal 2003 and fiscal 2006. Moreover, when measured by number of jobs (not contracts), OMB said agencies' in-house teams swept up 83 percent of that work, leaving just 17 percent to companies of all sizes.

Those figures compare apples to grapes and tomatoes, mixing contracts and competitions, jobs and contract awards, but the numbers do not seem to line up.

Eagle Eye Publishing, which wrote the report for SBA along with Bethesda, Md.-based Jack Faucett Associates Inc., clarified that those 3,735 contracts actually represent just 941 unique contract agreements. Contracts that extend over multiple years were counted multiple times in the study, said Eagle Eye president Paul Murphy (Eagle Eye also processes the data for Government Executive's annual Top 200 Contractors special issue).

On the Defense side, the contract figures represent 129 "initiatives," based on the best unique identifier that reviewers could pick out. But the initiative data included no deeper view of how long contracts lasted, among other missing fields.

Radwan Saade, an SBA contracting official who oversaw production of the report, said it would be impossible to compare the OMB and SBA figures without knowing more about how OMB's data was collected.

An OMB spokeswoman said factors contributing to different results could include how the studies account for private firms teaming with a government service provider, and the large number of contracts signed by NASA under a special waiver that resulted in multiple awards per competition.

What is clear is the low probability that any of the figures show the full picture.

By the time the report was published, FPDS-NG reflected almost $1.3 billion in A-76 contracts in fiscal 2006, according to Eagle Eye. A meager $8.98 million of that represented the first trickle into the system of Defense A-76 contracts. And in the most recent CAMIS report, the Defense A-76 spending for that year was listed as just $1.4 million.

It all depends on where you look.