Chmura Economics’ FedSpendTOP Database Provides a More Accurate Picture of Federal Contract Spending

Posted on March 31, 2017 by Chris Chmura

Many local economies are dependent on federal contracts, but the contract spending data available from government sources are often incomplete and inaccurate. Chmura Economics & Analytics created the Federal Spending by Time of Performance (FedSpendTOP) database to correct many of these issues.

Understanding trends in federal contract spending is important for many regions where federal contracts have significant impacts.

USASpending.gov is the publicly available official government source for data on grants, contracts, etc. but these data may be incomplete or inaccurate due to data quality problems.[1]

To resolve these problems, Chmura created the FedSpendTOP database. FedSpendTOP improves upon the USASpending.gov data by:

  • Adjusting for the length of the contract
  • Basing regional spending on the place of performance
  • Including purchases by non-Department of Defense (DoD) agencies which end up in DoD products as DoD spending
  • Identifying and correcting errors through quality control processes
  • Manually adjusting based on research

The following table shows the 10 metropolitan statistical areas (MSAs) with the highest DoD contract spending in fiscal year (FY) 2015 based on FedSpendTOP data and their corresponding rank based on unadjusted data from USASpending.gov.

Top 10 DoD Contract Spending by MSA, FY 2015
  Unadjusted FedSpendTOP
Washington-Arlington-Alexandria, DC-VA-MD-WV MSA 1 1
Dallas-Fort Worth-Arlington, TX MSA 2 2
Virginia Beach-Norfolk-Newport News, VA-NC MSA 7 3
Boston-Cambridge-Newton, MA-NH MSA 4 4
Los Angeles-Long Beach-Anaheim, CA MSA 6 5
San Diego-Carlsbad, CA MSA 3 6
St. Louis, MO-IL MSA 9 7
Baltimore-Columbia-Towson, MD MSA 5 8
Philadelphia-Camden-Wilmington, PA-NJ-DE-MD MSA 8 9
New York-Newark-Jersey City, NY-NJ-PA MSA 10 10
Source: Chmura FedSpendTOP & USASpending.gov

The unadjusted data put the Virginia Beach MSA 7th in the nation when it should actually be 3rd according to FedSpendTOP data. Adjustments based on company research and adjustments based on the length of contracts (shipbuilding contracts are typically multi-year) account for much of this difference. This is one of many examples of how FedSpendTOP data more accurately reflect where and when work supported by federal contracts is being performed.

[1] In 2014, the GAO estimated “with 95% confidence that between 2% and 7% of the awards contained information that was fully consistent with agencies' records for all 21 data elements examined.” http://www.gao.gov/products/D07755

This blog reflects Chmura staff assessments and opinions with the information available at the time the blog was written.