Tennessee Reduces Improper Unemployment Payments through Data and Determination
Connecting state and local government leaders
COMMENTARY | The Tennessee Department of Labor and Workforce Development implemented technology solutions to help mitigate improper unemployment insurance payouts.
The multi-billion-dollar problem of improper unemployment insurance payouts isn’t a new issue for state governments. The same web portals that make it easy for workers to apply for benefits also create an opportunity for fraud. Without cutting-edge identity verification tools, online identities can provide a smokescreen for criminals, allowing them to appear eligible for benefits and putting the entire system at risk.
Mitigating the risk of fraudulent claims centers on ensuring that the person on the other end of the transaction is truly who they say they are. Since claimants are signing up online more regularly, rather than walking into a government building, it has become much harder to verify their identities.
How hard is it? According to the U.S. Department of Labor, over the last three years, half of all states disbursed improper payments ranging from 10 to 26% of total unemployment dollars. The causes vary from errors by claimants and employers to outright deception and identity fraud. For example, a “fictitious employer scheme” that ran from 2009 to 2011 filed more than 900 claims in three Midwest states, totaling $8.7 million of taxpayer funds stolen.
Tennessee was no exception to this. In 2013, Tennessee’s state comptroller found that over $73 million in unwarranted payments had been made over the prior six years.
Back then, the Tennessee Department of Labor and Workforce Development (TDLWD) followed the standard protocol for identity verification by entering personal identifiable information like name, address, Social Security Number, etc. into the TDLWD’s registration system. This information was only checked against the Social Security database and State Information Data Exchange System (SIDES) to match applicants against a limited set of existing datasets. Then, employers were contacted to confirm unemployment. If the agency learned the applicant was still working and didn’t apply for benefits, it may indicate the individual’s identity was stolen, triggering immediate action. This was the standard process that many states followed for confirming applicants.
But a criminal can create an online identity that looks legitimate and still be partly or completely fictitious. What was needed was a way to immediately flag suspicious entries by verifying each applicant’s identity at the point of entry into TDLWD’s registration system.
Comprehensive Data is the Key
Relying on the system to authenticate identities and work status was necessary due to resource limitations. But that also meant that there was no agency follow-up until an issue was flagged by the system, which had only the information entered by the applicant to draw on. In contrast, a digital identity network has the ability to detect and block complex fraud in near real time by looking at a wide range of additional factors, including geolocation, time of access and IP address.
TDLWD implemented an expanded identity verification solution that drew on the organization’s 10,000 external data sources to provide a more complete picture of individuals applying for unemployment benefits. This enabled the system to almost instantaneously determine if someone creating a profile is potentially a criminal, triggering additional verification steps and alerting the TDLWD team.
The system also automates much of the claim verification process, including electronic follow-up with applicants, freeing up employees to focus on other tasks and providing a frictionless customer experience for real clients.
Of course, the state wants to support legitimate claims for unemployment benefits. But even claimants that initially meet the standards may continue to receive benefits after they are no longer eligible. The state can also identify these individuals through a thorough assessment of the digital identities used to access the TDLWD system.
Verification Saves Millions
The results in the four years since the identity verification technology was implemented have been exceptional. In 2017 alone, TDLWD prevented nearly $30 million in improper or fraudulent payments. During the first six months of 2017, the system pinpointed 2,400 identities that were not eligible for payments. TDLWD also was recently recognized by the National Association of State Workforce Agencies for their ongoing efforts to combat waste, fraud and abuse.
Essential Steps for States
A comprehensive identity verification system will take a commitment to pre-planning, collaboration, and training. For state labor agencies looking to shrink losses due to improper payments, a full understanding of their internally captured data is essential, as is having access to a broad range of outside data sources. This may require a technology partner for successful implementation.
Preventing fraud also requires a collaboration with employers, who are naturally concerned with limiting illegitimate payouts. In fact, without notification from employers around the state, TDLWD wouldn’t have been aware of approximately one-third of the fraudulent claims prior to implementing the identity verification technology. Fortunately, the Tennessee business community has been vigilant in looking for anomalies that might indicate an issue.
Even though statewide unemployment is down, the risks continue, with millions of dollars at stake. While acknowledging that there is still more to be done to reduce the number of improper claims, TDLWD considers its revamped system a platform for success. The strategic use of data for identity verification has proven to be essential to its efforts.
William Ornelas is Director, State & Local Government, Health & Human Services for LexisNexis Risk Solutions, which provides government agencies with access to the most data, analytics and linking technology.
NEXT STORY: How Good Is Your Phishing Training? A New Tool Could Help Figure That Out.