A Playbook for Improving Unemployment Insurance Delivery

This playbook presents a bold yet achievable North Star vision for what unemployment benefits can be, and the concrete steps needed to get there — concrete steps that, in almost every case, individual states have already tested.

To create the vision and map the steps, we collaborated with more than 50 advocacy groups, partners, dozens of state leaders and many unemployed workers. We highlight lessons learned from past recessions, recent pandemic-inspired innovations, and complementary benefit spaces like SNAP and WIC.

This is a playbook of what’s working to deliver unemployment benefits to those who need them.

We hope that this living playbook will inspire and drive further improvements, building on the successes of individual states. New America, supported by the Families and Workers Fund, will continue to update these pages as states develop more solutions. We invite you to submit your suggestions and stories here.

Yes, a lot has gone wrong in benefits delivery.

And a lot has already been written about what’s gone wrong. Coming out of the last recession, the American Recovery and Reinvestment Act of 2009 (ARRA) provided up to a total of $7 billion as incentive payments for states to “modernize” state UC benefit provisions. Payments were available through September 2011*. As recently as 2016, the Government Accountability Office (GAO) asserted that 40% of states had completed successful “modernizations” of their unemployment systems. But looking at the experiences of states delivering unemployment this year, it’s hard to see $7 billion in improvements or modern benefits delivery. The Department of Labor Inspector General’s 2021 report highlighted that it took far too long for states to disburse payments (an average of 50 days for the PEUC program) and did not follow DOL guidance, leading to unnecessary hardship for workers while also increasing improper payments.

As of this writing, more than half of states have opted out of continuing Pandemic Unemployment Assistance (PUA) benefits, many citing the challenges of rampant fraud. If we don’t put effective solutions in place now, we are at real risk that essential emergent benefits will not be an option in the next recession.

Unemployment benefits helped millions of Americans through this crisis.

While there were delays and delivery challenges, the CARES Act kept 12 million families out of poverty and helped support the economy through the pandemic economic crisis. Every $1.00 spent on UI created $1.61 in local spending, which helped keep the economy afloat during this difficult time. Overall, nearly one in four workers relied on unemployment insurance to weather the pandemic, with insurance claims peaking around 30 million in late June of 2020. Less generous unemployment benefits would have made the recession even worse, demonstrating how necessary these benefits were to families.

So we learn from the past and build on what works.

We set a vision for what we want unemployment benefits to look like — a North Star that hasn’t been articulated before.

We want claimants to easily access and manage the benefits they’re entitled to, so we set claimant-centric metrics and launch demonstration projects to refine the best ways to achieve them. And we share what we learn, in ways that other states can readily replicate.

We want equitable outcomes for all populations, so we design processes for the hardest-to-reach 10%. We measure how those processes are working across demographics groups, and we hold systems accountable to equitable access.

Here is the future we can have.

“What is a state's primary mandate when it comes to UI: To administer the prompt delivery of unemployment benefits to eligible applicants or to focus on identifying potential fraud and minimizing payouts?” — New America

The report begins by describing a potential future for unemployment benefits, one where all claimants can navigate the benefits application process, where real-time data feeds make it possible for them to receive payments the very same day, and where anti-fraud measures are effective at stopping benefit and identity theft, without also stopping real claimants from gaining access. We will know we have reached these goals because we have consistent, reliable metrics that track claimant-centric success outcomes as well as equitable access metrics across all populations.

These are the issues we need to examine.

“Digitizing a broken process gets you a digitized, broken process.” — New America

We considered the unemployment process step by step, through the lens of claimants, partners, and the limitations of current systems. We uncovered success stories, case studies, and promising practices across the following areas:

• Fraud prevention and identity verification that don’t discriminate against populations that need benefits the most. (Surprise: the federal government already has an effective solution, but state unemployment systems haven't been using it.)
• Wage verification enhancements, like automated wage verification for 1099 (freelance) workers and the creation of an online unemployment account that people could review to fix their employment and wage data as needed, before they ever need to apply for unemployment.
• Claimant experience improvements across multiple dimensions, including mobile access, claim status tracking, recertification, supporting multiple languages, accessibility, plain language content, cross-benefit promotion, and password resets. In particular, we emphasize making it clear when claimants need to take action, and making it simple and fast for them to take that action correctly.
• Payment-related improvements, including existing ways for states to leverage instant and/or digital payment methods, further increase today’s payment timeliness in service of an eventual same-day payment goal, and avoid overpayments.
• Claim processing changes that can help states better prepare for the next recession by using data and practice exercises to develop scalable resources, and use workload management tools to more effectively deploy experienced staff.
• Customer service practices like measuring first contact resolution, offsetting high contact volume with improved self-service offerings for those claimants who prefer self-service, and providing service across multiple channels in a way that can resolve all inquiries back to the original individual for improved service.

We also spend time on specific technology best practices, like website instrumentation and back-end system monitoring, that can help systems collect and act on data to drive improvements. But we emphasize that technology by itself is not a solution. COBOL systems can process claims just fine, just as a new single-page JavaScript application can make it impossible for users to apply. The technology itself is nowhere near as relevant as the surrounding goals, metrics, policies, and processes.

Here are the steps towards the future we want.

In summary, our recommended approach for improving unemployment is:

• Define success in terms of claimant-centric outcomes, like percentages of claimants receiving unemployment benefits the same day they apply–not “how” measures like number of hours spent or lines of code written.
• Consider policy, practice, and technology together** holistically in developing hypotheses for solutions. Good implementation cannot fix poorly constructed policy.
• Launch demonstration projects with interested states designed to iteratively build, test, and further improve upon ways to achieve these goals.
• Promote successful demonstration projects to develop shared services when they make sense.
• Conduct a rapid ecosystem assessment to organize states into cohort models for building and deploying solutions towards these success criteria.
• Deploy strike teams to perform deeper state-specific work that needs to be consistent nationally, like defining and measuring backlogs and fraud rates.
• Build state integrated command centers to respond to incidents, working down fraud and backlogs.
• Improve the relationship between U.S. DOL and states through collaborative projects like demonstration pilots and shared services like central plain language and transadaptation teams from which states gain value.
• Deploy philanthropic support towards specific demonstration projects in service of specific success metrics, as well as a state working groups to advance best practices and share fixes.

This report was made possible through the support and partnership of the Families and Workers Fund.

* Source: https://www.naswa.org/system/files/2021-03/usdolreleasesnaswareport.pdf, pg. vi)

** “States often found that changes to one aspect of service delivery affected or influenced another. State leaders described the importance of strategically addressing the interplay of policy, business process, and technology barriers as they worked to implement their visions of service delivery. State policy staff usually participated in design sessions for new eligibility systems to ensure that new technologies were as compatible as possible with existing policies and planned policy changes. Policy staff often worked alongside operational staff to set policies for business improvement pilots or write new policies to fit new business procedures. Illinois changed its policy on verifying citizenship, allowing the state to eliminate a citizenship form that had slowed processing of benefit applications. Likewise, redesign of business processes sometimes called for changes in technology, and access to improved technology could spur changes in business processes. The development of a single computer system to handle all of Rhode Island’s social service programs facilitated the shift to an integrated business process using universal workers.” link