A slew of competing research has attempted to estimate the number of Americans that lost job-based insurance due to the economic effects of the coronavirus, varying in projections from the millions to multiple tens of millions. But the losses may not be as drastic as initially thought, according to a new report from the Urban Institute.
Researchers performed a comparative analysis on four studies on COVID-19’s effects on job-based insurance coverage published between early May and mid-July, and found they differ wildly in methodology, assumptions and completeness. Smaller household surveys conducted during the pandemic have found little immediate change in insurance status, suggesting studies estimating more modest long-term changes in the number of uninsured people as a result of COVID-19 may be more accurate than others.
Despite the importance of waiting for definitive data next year, the models are still valuable for informing policymakers on the scope of the insurance crisis spurred by COVID-19. And researchers are continuing to pursue the question: New research released by the Economic Policy Institute on Wednesday found it’s likely 12 million people have lost employer-sponsored insurance since February.
Despite the Trump administration touting a rebounding economy, more than 1 million Americans applied for unemployment benefits last week, according to data released Thursday by the Department of Labor. The number of jobless claims has topped 1 million every week but one since late March.
Since the majority of Americans get coverage through their job, ongoing unemployment has given rise to concerns about an insurance crisis stemming from the pandemic and the threadbare U.S. safety net.
But recent household surveys suggest net changes in insurance coverage so far have been relatively minimal, Urban Institute pointed out in its brief. The surveys, conducted by groups like the Census Bureau and the Commonwealth Fund, are more timely but more limited than federal surveys with larger sample sizes. But none have found large increases in uninsurance at this point in the recession.
For example, the Coronavirus Tracking Survey conducted mid-May by Urban Institute and the Robert Wood Johnson Foundation found no change in employer-sponsored insurance, or in the number of uninsured people, for the overall sample. However, for those in families losing jobs, employer-based coverage dipped by 4.9 percentage points. That drop, though, was offset by a 3.5 percentage point increase in private non-group coverage.
More longitudinal estimates, like the four analyzed by the Urban Institute, are much more severe in their projections, but vary widely. They don’t all address the same exact questions, act on different methodology and assumptions and generally differ in completeness, Urban said.
Two earlier estimates from the Urban Institute pegged the number of people losing employer-based coverage over 2020 as 10.1 million workers and dependents, or between 17.7 million and 30 million workers and dependents over several months to a year. Both studies tried to incorporate projections of how many people would gain access to coverage through a family member, or through subsidized Medicaid or individual marketplace coverage.
One study from the Kaiser Family Foundation estimated 27 million workers and dependents were losing employer-based coverage over the remainder of 2020. However, the KFF study did not quantify the number of people who would become uninsured due to job loss.
Another from Families USA, looking at losses already occurred, estimated 5.4 million workers lost job-based insurance, but didn’t factor in family members and dependents who would also lose coverage as a result of the policyholder being let go.
The data is still shifting, and it’s important to wait for more definitive data points next year to get a fuller understanding of the pandemic’s effects on insurance coverage, Urban Institute said, though the projections are important for informing the nation’s public health response.