Uninsured Population Overstated

Uninsured Population Overstated
October 1, 2001

Since the early 1980s, the U.S. Census Bureau has used questions in its Current Population Survey (CPS) to arrive at estimates of the uninsured population. The Census Bureau estimates have consistently been higher than other estimates, such as those produced by the Urban Institute.

Now, the Census Bureau has recalculated its estimate, and it reports the uninsured problem may be less severe than previously thought.

  • The new count of the number of uninsured in 1999 reduces the number from 42.6 million to 39.3 million--a difference of 3.3 million people.
  • According to researchers, "private [health insurance] coverage appears to be the type of coverage most affected by reporting differences between the CPS and other surveys."

What’s a “Household”?

Contradictory answers given by CPS survey respondents help explain why the Bureau had to revise its estimate downward. Some respondents, who initially said they or someone in their household were uninsured, reported on a verification question that they were in fact insured. The contradiction is especially significant at higher income levels.

According to the National Center for Policy Analysis (NCPA), the verification question lowered the estimate of the number of uninsured living in households with annual incomes of $75,000 or more by 16 percent. The verification question lowered by 4 percent the number of uninsured living in households with incomes under $25,000.

The Census Bureau data may overestimate the rate of uninsurance among persons living in higher-income households because of a flaw in survey terms. The Census Bureau uses the term "household" to refer to all individuals living in the same residence, regardless of whether they are related. The Bureau considers all income of those in the household as "household income."

But “households” so defined can include a wide range of living arrangements: college roommates, multiple families, and adult relatives. Rarely do all members of such households have access to the total income of the household. This affects the ability to purchase health insurance. For example,

  • A household with an income of $80,000 could consist of a two-income family with two children where each adult makes $40,000; or it could be a family of two wage-earners making $30,000 each and a related adult family member (perhaps a niece or nephew) making $20,000 who cannot be covered by the same health insurance because he or she is outside the immediate family.
  • Three unrelated young adult roommates making $18,000 each would be listed as a “household” with income over $50,000, despite the fact that none can be a source of insurance for the others.
  • Two families, each with an income of $40,000, sharing the same house, would be counted as a household with $80,000 in household income.

Overestimating the Uninsured

Notes NCPA, “The failure to account for these differences overestimates the rate of uninsured in the higher income levels and paints an inaccurate picture of the uninsured problem.”

  • The Census Bureau reported 8.3 percent of individuals in households with incomes over $75,000 were uninsured. However, when families at that income level, rather than households, are considered, the uninsured percentage falls to 6.8 percent.
  • A family with two workers earning a total of $75,000 is less likely to be insured than a single individual making $75,000 because each only has access to $37,500 of that income. Thus, while 5.2 percent of workers earning $75,000 or more are uninsured, that number jumps to 9.7 percent for those earning between $25,000 and $49,999.

The NCPA analysis concludes, “When income is recalculated to exclude income of unrelated persons and related adult family members, the number of higher-income uninsured drops.”

The results of a new CPS survey will be released this month.

For further information . . .

see the report by Charles T. Nelson and Robert J. Mills, "The March CPS Health Insurance Verification Question And Its Effect On Estimates Of The Uninsured," August 2001, Housing and Household Economic Statistics Division, U.S. Bureau of the Census, available online at http://www.census.gov/hhes/hlthins/verif.html.

Additional information about the uninsured rate is available on the Web site of the National Center for Policy Analysis at http://www.ncpa.org/pi/health/hedex6a.html.

see “Census Bureau Reports May Overestimate Uninsured Rate in Higher Incomes Households,” BNA's Health Care Policy Report, March 20, 2000; Paul Fronstin and Rachel Christensen, "The Relationship Between Income and the Uninsured," EBRI Notes, March 2000, Employee Benefit Research Institute, 2121 K Street NW #600, Washington, DC 20037, 202/659-0670.