Databases are the infrastructure of the modern administrative state and data is its lifeblood. When the data is contaminated with errors, federal agencies have difficulty performing even the most basic administrative functions such as managing its inventory of office space and protecting the personally identifiable information (PII) of social security number holders. The federal dissemination of unreliable data doesn’t just waste money; it undermines public trust in government and leaves it unmanageable.

There are federal laws requiring that agencies collect and disseminate only data that’s reliable. One of these laws, the Data Quality Act, is of particular importance to the federal IT community since as the Department of Justice explains, “The Federal Data Quality Act supports the Federal Government’s Fair Information Practice Principles (FIPPs), including the principle of Data Quality and Integrity, which are the foundation for the government’s use of personally identifiable information (PII).” Unfortunately, time and again, these laws are more honor’d in the breach than the observance. Public-facing federal databases are often unreliable, misinforming the public about everything from small businesses contracting opportunities to the incorrect use of radioactive medical materials.

Federal Offices of Inspectors General and the US General Accountability Office (GAO) have repeatedly documented problems in federal databases that are so severe that they render the databases unreliable. Five examples of studies discussing unreliable databases published by federal watchdogs in 2015 were included in public comments provided to the Administrative Conference of the United States, A Data Quality Act-Based Blueprint for Correcting the Crisis in Federal Databases. ACUS is “an independent federal agency dedicated to improving the administrative process through consensus-driven applied research, providing nonpartisan expert advice and recommendations for improvement of federal agency procedures.”

The public comments on the federal database crisis were provided in response to a project ACUS has underway on Agency Information Dissemination in the Internet Era. One of the examples of unreliable federal databases is the General Services Administration’s Federal Procurement Data System (FPDS). A report by the US Small Business Administration’s Office of the Inspector General states that congressional concerns about the accuracy of report based on the FPDS have “been substantiated by OIG audits and other Government studies, which have shown widespread misreporting by procuring agencies, since many contract awards that were reported as having gone to small firms have actually been substantially performed by larger companies.”

The SBA OIG report discusses the harm done to small businesses because of the inaccurate procurement data being disseminated via the database. “Awards made to ineligible firms impact procurement opportunities for small businesses and damage SBA’s credibility in reporting accurate small business contracting goals achievements… Without reliable data, SBA cannot accurately measure the Federal Government’s small business procurement goals achievements, which in turn weakens the ability of Congress and other Federal policy makers to determine whether the Government is maximizing contracting opportunities for small businesses.” In short, small businesses are being deprived of contracting opportunities because of inaccurate federal data. Thus, errors in federal databases are an issue that should be of front burner interest to small businesses and to the government contracting community.

The public comments also cited a Social Security Administration OIG report that should be of concern to banks, pension funds, insurance companies, and everyone from governments to private investigators, Social Security’s Death Master File. The public access version of SSDMF was established in response to the PATRIOT Act to help banks verify the identity of their customers. The OIG report states, however, that the database “includes approximately 6.5 million numberholders who, according to their dates of birth, were at least 112-years-old but did not have death information on their record.”

The problems stemming from this unreliable database are more than theoretical. The OIG report states that “[w]e matched the 6.5 million SSNs against SSA’s ESF and E-Verify systems and identified thousands of instances of potential identity theft or other fraud.”

The public comments also discussed a federal database problem that should be of tremendous interest to the commercial real estate industry, to public administration scholars and to taxpayer interest groups, errors in GSA’s Federal Real Property Profile (FRPP).

GAO testimony discussing the database explained that “as we reported in our 2015 High-Risk update, the data behind” the Freeze the Footprint “results were unreliable, resulting in a potential overstatement of the progress made to date in reducing the federal government’s real property footprint. Specifically, we examined data from four of the six agencies claiming the largest reductions in the first year of implementation of the Freeze the Footprint policy and found that the actual space reductions at all four were overstated…” GAO also noted that “[r]etaining unneeded real property results in significant costs to the federal government.”

Unreliable federal databases are a sign of problematic management of federal information resources. Fortunately, the public comments included a set of three practical recommendations for remedying the federal data crisis,

1. The Data (Information) Quality Act and its implementing guidance should form the basis of an updated ACUS Recommendation on Adverse Agency Publicity.

2. The updated Recommendation should advise agencies to subject each database they disseminate to a DQA pre-dissemination review process to assess its reliability; the review process should include the opportunity for public comment.

3. The updated Recommendation should advise agencies to inform the public that the DQA is the appropriate mechanism for anyone seeking corrections to an agency-disseminated database.