Census Bureau Uninsured Numbers Raise Questions

Published October 13, 2009

The U.S. Census Bureau released new statistics on the number of uninsured Americans in its annual report on September 10, sparking discussion of what the statistics mean and some surprising information about the incomes of the uninsured.

The Census Bureau found an estimated 46.3 million Americans lacked health insurance in 2008, according to findings from the Current Population Survey (CPS). The proportion of Americans who lack health insurance grew by one-tenth of a percentage point, to 15.4 percent in 2008, up from 15.3 percent in 2007. According to the report, the uninsured rate has remained virtually unchanged for the past 15 years, hovering around 15 percent.

There is considerable disagreement about how the uninsured figures from the Census Bureau should be interpreted. According to Thomas Miller, a fellow at the American Enterprise Institute in Washington, DC, although the CPS is intended to track only individuals who are uninsured for an entire year, people tend to report being uninsured for shorter durations.

“The widely cited CPS statistic is considered closer to an estimate of those who were uninsured at the point in time surveyed, rather than the number of people uninsured for the entire year,” Miller said.

‘Massive Fraud’

Michael Cannon, director of health policy studies at the Cato Institute, believes supporters of government-run health care misuse CPS figures for political purposes.

According to Cannon, the CPS measures only the number of uninsured on any given day, not over a long period, and many of the uninsured quickly regain coverage.

“The Current Population Survey, and its estimate of 46 million or so uninsured, has been the occasion of a massive fraud on the public,” said Cannon. He says the number of Americans who are uninsured for the entire year “is more like 20 to 30 million—and many of those are already eligible for government programs or can afford to purchase health insurance.”

In 2003, the CBO compared data from the Survey of Income and Program Participation and data from the Medical Expenditure Panel Survey and concluded the CPS figures overstated the long-term uninsured by counting people as uninsured for an entire year who were actually uninsured for a shorter time. When those lacking coverage for only a short time were excluded, the true number of uninsured fell to between 21 and 31 million.

Overcounting the Uninsured

The September Census Bureau report also reported the large number of uninsured households that have what might be considered sufficient income to afford health coverage. Nearly 10 million uninsured individuals live in households earning more than $75,000 annually.

Approximately 13.7 million uninsured adults and children live in households earning less than $25,000 annually, a group who largely qualify for Medicaid or the State Children’s Health Insurance Program (SCHIP). Research by the BlueCross BlueShield Association has estimated up to 14 million uninsured individuals qualify for public coverage but have not enrolled.

Many of the low-income people counted as uninsured may actually have Medicaid coverage, according to Jacob Klerman, author of a new paper on the uninsured, published in the journal Health Affairs. “The undercount is three to six million,” Klerman said. “The Medicaid undercount refers to people who erroneously reported on the Census Bureau survey they were uninsured, despite being enrolled in Medicaid or SCHIP.”

Inflating the uninsured number still further, slightly more than 9.5 million of the uninsured are foreign-born residents who are noncitizens and may not be here legally, a question the Census Bureau does not ask in compiling its report.

Finally, the CPS figures note the uninsured tend to be young and relatively healthy—nine million are between the ages of 18 and 34—suggesting they may simply not find health insurance to be a good spending choice.

Devon M. Herrick, Ph.D. ([email protected]) is a health economist and senior fellow at the National Center for Policy Analysis in Dallas.

For more information …

“Understanding the Current Population Survey’s Insurance Estimates and the Medicaid ‘Undercount’,” by Jacob A. Klerman et al., Health Affairs, September 10, 2009: http://content.healthaffairs.org/cgi/content/short/hlthaff.28.6.w991

“Statement: Greenstein on Census’ 2008 Health Insurance and Poverty Data,” Center on Budget and Policy Priorities, September 10, 2009: http://www.cbpp.org/cms/?fa=view&id=628