How Agencies Can Leverage Advanced Analytics to Stop Fraud, Waste and Abuse in Unemployment Benefit Programs

How Agencies Can Leverage Advanced Analytics to Stop Fraud, Waste and Abuse in Unemployment Benefit Programs 

State unemployment insurance programs exist to mitigate the financial challenges many individuals and families face after a job loss. But state employment departments themselves face their share of difficulties in managing these programs, not least because they can be targets of fraud, waste and abuse.

That fact was driven home by the COVID-19 pandemic, which stretched agencies to the breaking point as unprecedented numbers of residents filed for claims. As if to add insult to injury, those claims involved no small amount of fraud. In September 2021, the U.S. Department of Labor (DOL) reported that an estimated 10% of $872 billion in federal pandemic unemployment benefits had been paid improperly, largely due to fraud.

Yet unemployment fraud, waste and abuse has been a perennial problem. The DOL estimates that inaccurate payments range, depending on the state, from as little as 5% to as much as 38%. That waste has a negative impact on state budgets and programs – and on the residents those programs serve.

But there is a solution. With the right technologies and approaches, states can apply advanced analytics to effectively identify and root out fraud, waste and abuse in unemployment insurance.

Advanced Techniques for Deterring Benefits Fraud

Unemployment fraud can take multiple forms. Employers might create false accounts to avoid tax liability. Claimants might submit false information or continue collecting benefits when they’re no longer eligible. Criminals might file false claims or defraud claimants through websites that mimic a state unemployment insurance portal.

State employment agencies typically work with federal, state and local law enforcement to identify and prosecute such fraud. Detection and prevention methods include cross-checking claims against employer data, verifying applicant identities, and investigating tips from members of the public.

But fraud schemes can be sophisticated. Fraudsters can access systems, acquire data and commit crimes across multiple channels – while remaining undetected by traditional business rules and monitoring tools. States need more effective approaches to identifying, detecting and stopping root causes of fraud, waste and abuse – before money gets paid out.

The solution is a purposeful curation of data assets, combined with rigorous application of advanced analytic methods and technologies, including predictive modeling, social-network analysis, risk scoring, machine learning (ML) algorithms and artificial intelligence (AI). ML classification algorithms alone have already stopped billions of dollars in tax-refund fraud at the state and federal levels.

Building a robust defense against unemployment benefit fraud starts with effective data management, which can identify new sources of data, enable efficient extract-transform-load (ETL) processes, and ensure ongoing effective data governance. For example, the DOL has encouraged all states to enter into participation agreements with the UI Integrity Center—a data sharing consortium managed by the National Association of State Workforce Agencies (NASWA)—to support the creation and use of its Integrity Data Hub (IDH), which enables cross-matching functionality to combat the challenges and urgencies of UI fraud.

Organizing this data in new and innovative ways—such as representing traditional data files as networks or “graphs”— can enable case analysts to identify suspicious patterns, search graphs for past instances of the pattern, and rank resulting cases. Team members can quickly assess the scope of noncompliance patterns, such as inconsistencies in reporting of employment by employer and employee.

AI techniques such as Beta-skeleton graphs can identify networks of fraud or noncompliance that aren’t revealed by human analysis. These fraud-detection models reduce the number of false positives and increase the efficiency of case analysts. Customized visualizations and dashboards then enable efficient reporting by team members and optimal actions by agency decision-makers. These techniques remain transparent to residents and don’t place unnecessary burdens on legitimate program beneficiaries.

Real-world Results of Fraud Detection and Prevention

The State of Arizona applied these advanced capabilities to identify and root out unemployment fraud. By mid-2020, the state’s Department of Economic Security (DES) suspected that up to 3.5 million claims for pandemic unemployment assistance (PUA) were fraudulent. It knew it would need to improve the speed and accuracy with which it detected and shut down new and evolving fraud schemes.

The state had already achieved measurable results by implementing modern technology to reduce income-tax fraud – preventing more than $130 million in fraudulent payments to date. In August 2020, the state began implementing new strategies and processes for unemployment fraud analytics. Improved technology enabled DES to capture and analyze large datasets to better understand PUA claimant eligibility. It also allowed the department to quickly generate predictive models, identify issues, generate risk scores, achieve new insights and reduce fraudulent PUA payments.

Unemployment insurance fraud, waste and abuse will continue to be a challenge for state governments. But with the right technologies to enable advanced analytics and the right strategies for putting analytics to work, states can make measurable progress in shutting down fraud, saving taxpayers money and serving residents more effectively.