In a Wall Street Journal piece yesterday, Phil Gramm and John Early expose a weakness in the way that some politicians and pundits interpret Census Bureau data to talk about the child poverty rate. The weakness is this: No matter how much money the government pours into refundable child tax credits, the consequences won’t show up in the Census Bureau’s numbers measuring child poverty. That’s because the agency, when assembling data on poverty, doesn’t include transfers in their data on household incomes.
Gramm and Early explain:
House Speaker Nancy Pelosi and Senate Majority Leader Chuck Schumer assured us in July 2021 that expanding the child tax credit would “cut the nation’s child poverty rate in half.” Shortly thereafter, President Biden proclaimed that the expanded credit would “cut child poverty in half this year.” …
As we pointed out on these pages, the Democrats’ rosy promise wouldn’t be recorded in the official Census Bureau poverty numbers, because the income numbers used to calculate the official poverty rates don’t count refundable tax credits as income to the recipients. No matter how much money the government pours into any of these tax credits, it will never raise the official income measure given the way the census defines income.
This is not unique to the measurement of child poverty:
The Census Bureau fails to count two-thirds of all government transfer payments to households in the income numbers it uses to calculate not only poverty levels but also income inequality and income growth. In addition to not counting refundable tax credits, which are paid by checks from the U.S. Treasury, the official Census Bureau measure doesn’t count food stamps, Medicaid, the Children’s Health Insurance Program, rent subsidies, energy subsidies and health-insurance subsidies under the Affordable Care Act. In total, benefits provided in more than 100 other federal, state and local transfer payments aren’t counted by the Census Bureau as income to the recipients.
This is why it matters:
If the Census Bureau had included the missing $1.9 trillion in transfer payments, child poverty would have been only 3.2% in 2017, compared with the official rate of 17.5%. Government transfer payments that were distributed in 2017 had already cut child poverty by 82%….
Last year, the official census numbers for 2020 failed the laugh test. They showed that household income was down by 2.9% and the poverty rate was up by 1 percentage point in a year when federal transfer payments expanded by 36%. For the first time ever, the Census Bureau included the supplemental estimate in the same release as the official number, showing that income had actually risen by 4% and the poverty rate had fallen from 11.8% to 9.1%. Had it counted all the transfer payments, the poverty rate would have been about 2%.
By the way, this measurement convention isn’t unique to the Census Bureau. I have pointed out many times that using the Bureau of Labor Statistics’ (BLS) measurement of the prevalence of paid leave in America paints a picture that’s utterly incomplete. I once wrote for the Acton Institute:
First, according to the Bureau of Labor Statistics, 17 percent of workers have access to a paid leave program (an increase from 13 percent in two years). However, this number is highly misleading since it severely underestimate the actual number of workers that benefit from leave due to BLS’s peculiar survey methods which require paid leave to exist separately from “sick leave, vacation, personal leave or short-term disability leave that is available to the employee.” Proper accounting using several government surveys about worker benefits reveals that a majority of workers have access to paid family leave benefits and three out of four who take leave in a given year get full or partial pay.
The way the Census and BLS measure these things is not convenient. For instance, I would prefer for the Census to report both strict income and tax and transfer data. However, the agencies themselves are transparent about the choices they make. The problem is with the people who use these raw numbers to advocate for more child-poverty subsidies or a federal paid-leave program without acknowledging the measurement caveat.
Gramm, Early, and Robert Ekelund have a new book out that looks at how measurement choices affect the debate over inequality.