GAO: Weekly Jobless Claims Data ‘Flawed’

Inaccurate reporting, fraud skews unemployment reports

Roy Maurer By Roy Maurer December 9, 2020
DOL headquarters

​The U.S. Government Accountability Office (GAO) cited the federal government's closely watched weekly jobless claims data as being flawed, containing overestimates, underestimates, inconsistency and a high prevalence of fraud among claimants.

The GAO said much of the inaccurate data published each week by the U.S. Department of Labor stems from a faulty method of reporting from the states, many of which have been overwhelmed by processing backlogs.

States reported that 712,000 U.S. workers filed for new unemployment benefits during the week ending Nov. 28. The total number of workers continuing to claim unemployment benefits was 5.5 million. Another 4.5 million are enrolled in the Pandemic Emergency Unemployment Compensation program for people whose benefits have run their course, and 8.8 million claimed benefits as part of the Pandemic Unemployment Assistance (PUA) program for people not typically eligible for jobless benefits, such as self-employed and certain gig economy workers who are unable to work because of the COVID-19 pandemic. Approximately 20 million people overall are in the process of collecting some kind of state or federal unemployment aid, according to the DOL data.

But that number is likely inflated, the GAO concluded. The government watchdog agency discovered that the DOL has been reporting the number of claims it receives each week from state agencies as individuals, a method which provided a good approximation of the actual number of people claiming benefits before the pandemic. But that method has since been proven problematic, said Thomas Costa, acting director of the GAO's education, workforce and income-security team.

"Prior to the pandemic, the number of continued claims approximated the number of people claiming benefits because each week individuals typically filed a claim for just the previous one week of continued unemployment," Costa explained. "However, this has not been the case during the pandemic because of challenges implementing the newly created PUA program and backlogs in processing historic numbers of claims in many states. If an individual claims benefits for multiple weeks of unemployment during a single reporting period, each week is counted as a separate continued claim. This could happen if an individual was unemployed for multiple weeks before their application was processed—due to claims backlogs—or if the individual claimed benefits retroactively in the PUA program."

For example, a person filing for four weeks of benefits in a single week would be counted as four separate people by the DOL, instead of one person.

Multiweek claims are especially prevalent in the PUA program because individuals accumulated weeks of unemployment as states hurriedly tried to implement the new program, Costa said. When GAO analyzed data for 20 selected states, it found that the number of continued claims submitted in the PUA program in its first three months exceeded by almost 20 million the total number of people who had submitted an initial claim.

Meanwhile, sometimes claims are underreported, especially in the PUA program. That's because "states started reporting PUA claims data at different times, depending on how quickly they were able to implement the new program and establish reporting processes," Costa said. "In addition, even after implementing the program, in certain weeks some states did not report data to DOL to include in its weekly news releases."

As a result, the DOL reported inaccurate decreases in claims overall. Then there have been other problems with the weekly unemployment claims process, from millions of suspected fraudulent claims to numerous reported delays and denials for scores of legitimate claimants.

Unemployment fraud, which appears to be particularly prevalent in the PUA system, has skewed claims counts, the GAO said.

The DOL's weekly jobless claims reports have been one of the most eagerly followed economic data points tracked during the pandemic, providing a nearly real-time picture of changes in the labor market.

In a response to the GAO findings, the DOL said it planned to clarify its future unemployment releases to note that the number of continual claims does not accurately estimate the number of people claiming benefits. But it doesn't anticipate any changes to how the claims are counted, noting the imminent expiration of pandemic-related unemployment insurance benefits and stating that making any changes could be difficult for state unemployment offices.

Both the PUA and the Pandemic Emergency Unemployment Compensation program are set to expire at the end of the year unless Congress intervenes to extend them.



Hire the best HR talent or advance your own career.


Find your peers in SHRM's online community.

Find your peers in SHRM's online community.

Join SHRM Connect


HR Daily Newsletter

News, trends and analysis, as well as breaking news alerts, to help HR professionals do their jobs better each business day.