Skip to main content
<Case Study>

Our Risk Model for PBGC Identified Potential Misrepresentations in Pension Plan Assistance Applications.

The American Rescue Plan (ARP) Act of 2021 appropriated up to $94 billion to the Pension Benefit Guaranty Corporation’s (PBGC) new Special Financial Assistance (SFA) program.

Multiemployer pension plans in financial trouble could apply to the SFA program for assistance to help pay benefits. Under the program’s rules, PBGC had 120 days to review applications and determine if the information applicants provided was reasonable.

What We Did

PBGC’s Office of Inspector General (OIG) had an intensive manual process to review the applications and identify and act on potential risks. The OIG asked us for help in identifying potential misrepresentations in the applications. Our approach was threefold:

  • Provided historical and current data for PBGC analysis.

  • Identified abnormal behaviors in SFA applications.

  • Created an interactive dashboard that the OIG could use to conduct analysis quickly and efficiently.

We used advanced data analytics to detect anomalies, and we developed a predictive risk model to highlight the anomalies in SFA applications. We then created data tools and resources to enhance the PBGC auditors’ ability to identify risks before paying out funds.

Why it Matters

As of November 2025, PBGC and civil settlements with the Department of Justice have recovered $262 million related to improper payments.

PBGC OIG continues to use the model and dashboard to identify potential high risk or fraudulent applications.

PRAC Advanced Analytics

  • Program: Special Financial Assistance Program = $94B

  • Data Analytics: We transformed information from large volumes of PDFs into data we could analyze.

Image
PRAC Risk Model
Image
Arrow
  • Rapid detection of risk factors.

  • Targeted insights into SFA applications.

  • $262M recovered through administrative and civil actions.

Page last modified: 04/02/2026
Thank you for your feedback!
Would you tell us more? Feedback
Was this page helpful?