The High Price of Low-Quality Data: Insights from a Congressional Hearing

Amidst recent federal budget discussions, Congress faces another significant question related to government spending: how can the country reduce improper payments? On September 10, 2024, the House Oversight and Accountability Subcommittee on Government Operations and the Federal Workforce held a hearing on the federal governments’ efforts to prevent and reduce fraud and improper payments – those that should not have been made, should have been made but were not, or were made in incorrect amounts. The hearing was intended to serve as a roll out of the scorecard on Program Integrity developed by the Committee. Tellingly, however, Chairman Pete Sessions explained the need to deviate from the original agenda, stating “it’s become apparent we simply lack the data to determine whether programs and agencies are getting better.” 

Data became the heart of this hearing. Whether to ensure federal funds are properly disbursed, determine fraud risk and employ preventative measures, assess agencies’ improvements in reducing improper payments, or provide transparency for the American public – the conversation came back to the data challenges for enhancing program integrity. With compelling witness testimonies and expert input from the Honorable Michael Horowitz (Pandemic Response Accountability Committee, or PRAC), Orice Williams Brown (Government Accountability Office), and Linda Miller (Program Integrity Alliance), lawmakers shifted their focus to addressing information needs and other issues that hamper the use and effectiveness of program integrity measures.

Data Challenges and Recommendations

The hearing’s discussions touched on some major data challenges, incIuding limited access to necessary data, data quality, and interoperability issues that prohibit agencies from adequately preventing and reducing improper payments – whether due to administrative errors or malintentions. More generally, the hearing underscored the need for robust data and IT systems to prevent fraud and inefficiencies, highlighting the broad impact of underinvestment in technology and data management across federal and state agencies. Antiquated and siloed data systems not only impede the use of emerging technologies to detect fraud patterns, but also hamper the effective execution of federal programs overall. As Brown aptly put it, “IT investments are … easy to ignore when things are going okay.” 

The witnesses outlined several actionable steps that Congress can take to address improper payments, including a call for investments in centralized solutions to provide access to the data necessary to manage federal programs. While there are ongoing efforts within the federal government to streamline data access and offer analytics capabilities – both for program management and oversight purposes – congressional action is required to strengthen or uphold some of these initiatives. For example:

• The Pandemic Analytics Center of Excellence (PACE), an analytics platform created by the PRAC to help the oversight community identify fraud and improper payments, has a statutory sunset in September 2025 and will be decommissioned unless Congress passes legislation to sustain the initiative beyond this date. 
• The Treasury Department’s Do Not Pay (DNP) portal was granted temporary access to a  key data source: the Social Security Administration’s Death Master File (DMF), a database of deceased individuals and their Social Security numbers. DNP’s access to the full DMF, which includes state death records, will expire in 2026, barring congressional approval for permanent authorization. 

Access to death data was brought up in relation to both initiatives. Cited as one of the root causes of improper payments, death data is critical to help prevent payments from being made to deceased individuals. As Horowitz pointed out, establishing access to the DMF can be a costly and time consuming process. To maximize the impact of payment integrity initiatives such as DNP and PACE, two factors are key: providing secure access and enhancing the quality and timeliness of death data – and vital records data more broadly. 

There are viable solutions to this dilemma. As the Data Foundation suggested in a 2024 report, creating a national, interoperable, and secure vital records service could streamline access to vital records and strengthen payment integrity.

Better addressing improper payments requires, as Miller put it, a whole-of-government approach. Efforts should be made to bridge silos and foster not only cross-agency, but also intergovernmental collaboration. States play an important role in administering major federal benefits programs, such as Medicaid and Unemployment Insurance, and as stewards of data for ensuring program integrity. Building a robust, secure data ecosystem together and investing in accurate, timely data for government agencies administering benefits is instrumental in preventing improper payments.

Where Can Congress Go From Here? 

While streamlined access to quality data may require an upfront investment, the opportunity it presents to significantly reduce improper payments has the potential to yield a tremendous return on investment. As Representative Jasmine Crockett observed, referring to GAO’s estimates on fraud, “I could guess that whatever the technology costs, it is less than $233 - 251 billion a year.” 

By investing in modern, interoperable data systems and timely data, and fostering stronger collaboration between federal and state agencies, there is a unique opportunity to enhance the integrity and efficiency of federal programs. Better data could not only help to keep ineligible recipients from receiving federal funds, but also reduce administrative burden and improve government service delivery for those who are entitled to them.  

The hearing provided lawmakers with valuable recommendations on where to go in their continued efforts to address improper payments and fraud. Now it is up to Congress to collaborate with the Executive Branch and chart a path to lead the way forward.