This paper reports the general results of research undertaken by Census Bureau staff. The views expressed are attributable to the author and do not necessarily reflect the views of the Census Bureau.

In 1980 the Census Bureau began asking questions about health insurance on the March Income Supplement to the Current Population Survey (CPS). Health insurance questions are asked each March, and refer to coverage at any time during the previous year. For sometime, the Census Bureau has published annual estimates of participants in specific health insurance programs such as Medicare, Medicaid, and employer-provided health insurance from the CPS. However, it wasn't until 1993 that the Census Bureau began publishing comprehensive estimates of health insurance coverage from the CPS. Even so, prior to their publication, the information has been widely available to private and government researchers through CPS microdata files and unpublished Census Bureau tabulations. The CPS has been a major source of information on the extent of health insurance coverage in the United States.

Underreporting of health insurance coverage in the CPS has been a persistent problem, as other surveys (such as the Survey of Income and Program Participation) have shown consistently higher annual coverage rates. Another issue involves the belief by some researchers and data users that CPS respondents tend to answer health insurance questions based on their current status rather than the previous year.

The main objectives of this paper are: 1) Analyze and compare recent trends in CPS and SIPP health insurance estimates. 2) Address the hypothesis that CPS respondents tend to answer health insurance questions based on their current status rather than the previous year. 3) Evaluate the findings. 5) Present recent developments in the CPS and SIPP.

Although the statistical testing for this paper shows some significant differences in estimates, analytically the differences are not considered significant. For the purpose of this paper only the analytical similarities or differences in estimates are discussed.