With machine learning, risk scoring and automation, agencies can effectively identify fraud, waste and abuse in pandemic relief.
When the COVID-19 pandemic struck, stores and restaurants closed, factories were shuttered and millions of American residents suddenly found themselves out of work. The federal government and state agencies quickly offered pandemic-related benefits, including additional unemployment insurance, housing assistance, health care, food and nutrition and paycheck-protection loans.
But the surge in funding and benefits claims inevitably resulted in high levels of fraud, waste and abuse. For instance, one study found that as many as 15%, or $76 billion, of the federal government’s Paycheck Protection Program (PPP) loans were potentially fraudulent.
By September 2021, more than $87 billion in federal pandemic unemployment benefits had been paid improperly, largely due to fraud. Much of the abuse was linked to organized crime rings using stolen identity data.
State benefit programs also grappled with abuse. Washington State, for instance, called in the National Guard to help process 190,000 unemployment benefits claims that were flagged as potentially fraudulent. Meanwhile, residents across the country whose identities were stolen were hit with tax bills for benefits they never received.
Fortunately for beleaguered state agencies, there are solutions. With the right technologies and strategies, states can apply advanced analytics to effectively identify and root out fraud, waste and abuse in pandemic relief.
Tools for detecting and preventing benefits fraud
A common approach to identifying benefits fraud is to use details from known past cases to train a machine-learning (ML) algorithm. The algorithm quickly learns to determine the probability that a case is fraudulent. Of course, this technique requires knowing which cases are fraudulent, so it can be less effective against previously unseen fraud patterns.
Graph-based analytic approaches, such as social network analysis, can help. Rather than looking at a single claim as the unit of review, graph-based analysis looks for connections among multiple suspicious claims -- such as a large number of claims associated with a single username, email address, IP address or bank account.
Some of these patterns can be difficult for fraud analysts to recognize, particularly when fraudulent cases are hidden within a large population of legitimate beneficiaries. But ML-based pattern recognition can determine, for example, that an unusually high proportion of claimants are associated with a data breach at a single bank. That can be useful for uncovering claims from crime groups responsible for fraud clusters.
Combining such ML algorithms with the insights of fraud analysts can be highly effective. These “human in the loop” methods can also help agencies maintain public trust by adding transparency and explicability to the fraud detection process. This is especially helpful when using sophisticated artificial intelligence algorithms, which can deliver outputs in ways that are opaque to anyone but a data scientist. Adding the knowledge and experience of fraud analysts to the speed and number-crunching capabilities of AI can also help eliminate perceived bias in fraud detection.
Risk scoring to reflect agency priorities
Risk scoring is another useful tool against fraud. It can help prioritize suspected fraud cases for further investigation, focusing agency time and energy on identifying and stopping more cases of fraud more quickly. Risk scoring is especially helpful for crisis-related benefit programs, such as pandemic relief, that involve sudden spikes in the number of claims.
State agencies are often limited by the budget and staff they can dedicate to fraud investigations, so they need to focus resources on the most important cases. Algorithms can rank-order cases or assign probability scores to surface the most likely fraud scenarios.
These algorithms can be customized to reflect agency goals. For example, an agency might want to prioritize identifying and investigating the most probable cases of fraud rather than pursuing cases involving the largest dollar amounts. Agencies might also need to balance preventing fraudulent payouts against maintaining low false-positive rates so that payments to legitimate claimants aren’t delayed.
Automation and integration for speed and accuracy
Many agencies have invested in automation to process claims more quickly. But few have looked to automation to detect fraud at similar speeds. Paired with analytics, robotic process automation can accelerate fraud analysis at key decision points. And because RPA handles repeatable processes the same way every time, it helps avoid human error.
One challenge of emerging technologies like ML, RPA and advanced analytics is that they require substantial compute power to achieve the scale and speed needed in government programs. This barrier can be particularly acute for agencies hampered by legacy IT systems, and budget constraints that make it impractical to replace their server infrastructures just to automate fraud detection.
That’s where integration plays an important role. By integrating AI and RPA capabilities running on a separate, modern platform, while maintaining existing operations on legacy systems, agencies can reap the benefits of automation without having to make a large hardware investment. Even better, providers of AI-enabled analytics solutions are beginning to design algorithms that can be deployed effectively in existing production environments and quickly plugged into existing data sources.
The long-term impact of the COVID-19 crisis remains uncertain. For now, many pandemic-related benefits programs are beginning to wind down. But agencies can take away many lessons learned from their struggles against fraud during the height of the pandemic. And with the right analytics and emerging technologies and approaches, they can be more successful in their efforts to shut down fraud, save taxpayers money and serve residents effectively.