Health Insurance Coverage in 2013: Gains in Public Coverage Continue to Offset Loss of Private Insurance
The economy has steadily improved since the Great Recession peaked in 2010, but recovery in employment and household income has lagged behind GDP growth. While real GDP recovered to its 2007 high by 2011, the unemployment rate declined but remained high through 2013 (7.4 percent), and median household income continued to decline through 2012. The recession accelerated the long-standing decline in employer-sponsored health insurance (ESI),1 and through 2013 most of the recovery in the uninsured rate was due to increased enrollment in public insurance, primarily Medicaid and the Children’s Health Insurance Program (CHIP). For adults, coverage through Medicare and military healthcare programs also increased slightly between 2010 and 2013, though not as substantially as Medicaid and CHIP coverage. With the exception of young adults ages 19 to 25, who are able to remain on their parents’ health plan until age 26 under the Affordable Care Act (ACA), ESI coverage rates for adults and children continued to decrease between 2010 and 2013. These declines in ESI coverage are partly attributable to changes in population characteristics among the nonelderly, including an increase in the number of low-income families, population growth in low-ESI regions, and workforce growth in low-ESI industries.
Data and Methods
This brief uses data from the American Community Survey (ACS), an annual survey conducted by the Census Bureau and designed to be representative at the national and state level. The public microdata sample for the ACS contains 1.9 million observations annually, making it by far the largest of the federal surveys. The ACS contains data on income, health insurance, demographics, work status, and industry sufficient to allow analyses of the differences in insurance coverage patterns across various populations. In addition, the very large sample size allows for state-level trend analyses (not included here).
Prior issue briefs in this series used the Current Population Survey Annual Social and Economic Supplement (CPS) to describe trends in health insurance coverage.2 However, long planned improvements to the insurance questions for that survey resulted in a break in trend between the 2013 CPS and the 2014 CPS, which collected data on coverage in 2012 and 2013, respectively.3 This brief therefore focuses on trends from 2008 to 2013 using the ACS. While the ACS has a significantly larger sample size than the CPS, it also has a few disadvantages. First, the income information in the CPS is much more detailed than that collected in the ACS, and income data from the CPS is therefore the source of official estimates of poverty in the United States.4 Second, the ACS does not collect data on employer size, so this brief does not present trends in ESI coverage by firm size.
This analysis uses the ACS iPUMS files created by the Minnesota Population Center, which have consistent variable definitions over time and include constructed variables on family relationships and income that are used to create Health Insurance Units (HIU)5 and calculate HIU income as a percent of the FPL.6 In addition, the Urban Institute has developed a series of logical coverage edits to the ACS designed to correct for known inaccuracies in survey-based estimates of health insurance coverage.7 In particular, the ACS over-represents private non-group coverage relative to other surveys and under-represents Medicaid and CHIP coverage among children relative to administrative data.8 These logical coverage edits reassign coverage types for respondents when other information collected in the ACS, such as receipt of Supplemental Nutrition Assistant Program (SNAP) or other public assistance, implies that a respondent’s coverage has likely been misclassified.9 Finally, all individuals reporting multiple health insurance types have been assigned to a single primary insurance type using a hierarchy, which further corrects for over-reporting of private non-group coverage. The hierarchy used for all analyses in this brief is as follows: employer-sponsored insurance, Medicaid or CHIP, military health care or Medicare, private non-group insurance, or uninsured.