Census poverty figures may underestimate impact of pandemic unemployment insurance by half
The Census Bureau’s recent report on poverty for 2021 raises questions about the accuracy of the survey during COVID-19. Poverty levels in the United States have been found to have fallen significantly, largely due to stimulus payments and temporary extensions of cash transfer programs introduced under various relief programs. However, the assessed impact of pandemic unemployment benefits looks mysteriously disappointing.
While stimulus checks and the expanded child tax credit lifted 8.9 million and 5.3 million people respectively out of poverty, Unemployment Insurance (UI) lifted only 2 .3 million people above the federal poverty line (for each government benefit, the anti-poverty impact is assessed by counting the number of people in households whose total income is below the poverty line with and without the dollars provided by this specific program). Although all three policies have contributed significantly to poverty reduction, it is still surprising that unemployment benefits have not had an even greater effect given the scope and scale of the expansion. federal.
The anti-poverty effect of unemployment insurance is probably underestimated
For context, more than 10 million workers received unemployment insurance benefits every week in 2021 until the federal improvements expire permanently in September. People generally ineligible for unemployment insurance – gig workers, self-employed workers, etc. – were able to qualify, and the amount of regular weekly benefits was increased by $300 for all beneficiaries. Federal Unemployment Insurance expansions have been a clear pandemic success story, helping millions of families put food on the table during a time of massive employment disruption.
Still, evidence suggests that the Census Bureau numbers may be understating the success of unemployment insurance. In a new article by Jeff Larrimore of the Federal Reserve Board, data reporting issues associated with the Bureau’s Current Population Survey (CPS) are shown to be having a major impact on Census UI estimates for 2021. IRS tax data, derived from actual administrative records, indicates that more than $300 billion in unemployment benefits were distributed to 26 million people in 2021. But CPS data that the census used for its report Annual Report on Poverty appears to reduce the size of the unemployment insurance program by nearly $200 billion.
Data collection methods may be a potential source of this difference. The under-reporting of benefits from social insurance programs, such as unemployment insurance, has been a long-standing problem of household surveys like the CPS. When respondents are asked about their household income, they may forget to include unemployment insurance payments or small amounts received due to the stigma associated with benefits. The shortcomings of survey reporting could have been further exacerbated by the scale of COVID-19 unemployment insurance programs and the difficulty in reaching potential participants during the pandemic. While the IRS records aren’t perfect, the results are based on administrative tax data that aligns much more closely with numbers from other federal agencies.
The data discrepancy is not limited to last year
These reporting disparities also manifested themselves in the first year of the pandemic. According to IRS data analyzed by Larrimore and his colleagues, Jacob Mortenson and David Splinter, there were a total of 45.4 million unemployment insurance beneficiaries receiving a total of $565 billion in benefits in 2020. In contrast, CPS data indicates that 23.6 million recipients received $218 billion. The bottom two income deciles alone were recorded as having received $168 billion less in unemployment insurance benefits than the IRS figures for 2020.
In short, aggregate census data falls far short of IRS tax collection figures, especially for low-income workers. A similar discrepancy can be seen by looking at data from the Great Recession. During this period, unemployment insurance receipts and distributed benefits increased, but the CPS data did not correctly measure the increase among those with lower incomes. Much of the benefits distributed to people close to the poverty line are simply missed out.
IRS data can improve pandemic poverty estimates
The assessed anti-poverty effects of unemployment insurance are underestimated because they are based on data that underestimate the amount of cash assistance paid to recipients. When adjusting for benefit levels seen in IRS data, the Supplemental Poverty Measure (SPM), which accounts for tax credits not included in the official poverty rate, rises from 9.1 % to 7.3% in 2020 (a reduction of 20%) and from 7.8% to 6.6% in 2021 (a reduction of 15%), as shown in the table below.
These adjustments are presented with the population totals below. Officially, Unemployment Insurance is credited with lifting 5.5 million people above the MPS poverty line in 2020 and another 2.3 million people in 2021. When IRS data on beneficiaries unemployment insurance and benefit levels are applied using estimates from Larrimore, a Additional 5.9 million and 3.9 million people are kept out of poverty in 2020 and 2021, respectively. In other words, the impact of pandemic unemployment insurance on reducing poverty was more than twice as large as the census data suggest when supplemented with national tax data. .
An important caveat is necessary to contextualize these results. Just as the CPS data understates distributed benefits, the IRS data may overstate the numbers. The Unemployment Insurance program has battled identity theft and false payments during the COVID era. Notably, the employment and training administration of the Ministry of Labor assessed that there was an abusive payment rate of 18.7%.
Including this mispayment rate in the adjusted IRS calculations lowers the total poverty reduction to 9.3 million people in 2020 (originally 11.4) and five million in 2021 (originally 6.2). Yet under these conditions, levels of poverty reduction through unemployment benefits would still be 83% higher than census estimates. Even if you assume that false payments went to one of the three declared beneficiaries, the anti-poverty effects of unemployment insurance would still be 50% greater than the official figures indicate.
The take-out sale
Misleading data estimates can have significant effects on people’s perceptions and, eventually, political decisions. To date, political organizations and the media have taken the census figures at face value. These misconceptions risk entrenching an inaccurate portrait of the anti-poverty impact of unemployment insurance as a consensus. Reducing the real impact of unemployment programs by more than half may then lead policymakers to underestimate the importance of unemployment insurance reform on the list of legislative priorities. This does the workers a disservice. In the interest of defending future expansions, it is essential that we get these measurements correct in the future.
Photo credit: iStock