New Census Estimates Show 3 Million More Americans Had Health Insurance Coverage in 2012

By: Kenneth Finegold

Abstract

Estimates from Current Population Survey Annual Social and Economic Supplement (CPS-ASEC) data for calendar year 2012, released on September 17, show that 3 million more Americans had health insurance coverage than in 2011. This ASPE Issue Brief includes estimates of the uninsured by age, race and ethnicity, employment status, and household income. It also traces trends in sources of health insurance coverage from 1999 to 2012 and discusses the implications of the findings in the context of the expansion of affordable coverage options under the Affordable Care Act.

The Census Bureau released new data on health insurance coverage and the uninsured for 2012 on September 17, 2013.¹  The new data come from the Current Population Survey Annual Social and Economic Supplement (CPS-ASEC), the most widely cited of the multiple national surveys of health insurance coverage. 

The key finding is that 3 million more Americans had health insurance coverage in 2012:  the number of Americans with insurance coverage increased from 260.2 million in 2011 to 263.2 million in 2012.  The corresponding percentage of people in all age groups without health insurance decreased from 15.7% in 2011 to 15.4% in 2012, a statistically significant change. 

The Census survey, fielded from February to April 2013, asked respondents about coverage during the previous calendar year, 2012, and classifies individuals as uninsured if they did not report any source of coverage for any time during the year.  Individuals are considered uninsured, therefore, if they reported being uninsured for the entire year.  Others who may have been uninsured for only part of the year are not counted as uninsured in these data.

Estimates from the CPS-ASEC give us a broad picture of patterns of uninsurance and sources of coverage prior to the establishment of the Health Insurance Marketplace and the expansion of Medicaid eligibility under the Affordable Care Act.  Open enrollment for the Marketplace began October 1, 2013 for coverage starting on January 1, 2014 (http://www.healthcare.gov).

Estimates of the Uninsured

The increase in health insurance coverage in 2012 represents a continuing and encouraging reversal in trends: from 2000-2010, the number of uninsured Americans went up each year.  Since 2010, as the economy improved and the early features of the Affordable Care Act took effect, the number of uninsured has gone down.  The decline in rates of employer-sponsored coverage, which dropped sharply in the previous decade, have leveled off, and public coverage under Medicare, Medicaid, and CHIP has increased.  The net result of these two trends is an increase in health insurance coverage. 

An estimated 48.0 million people were uninsured in 2012, 2 million fewer than the 50.0 million who were uninsured in 2010, when President Obama signed the Affordable Care Act.  The most recent estimates by the Congressional Budget Office (CBO) suggest that by 2016, the expansion of Medicaid and the establishment of the Marketplace under the Affordable Care Act will reduce the number of uninsured Americans by another 25 million.²

Age and Uninsurance

• The estimated percentage of children (under 19) who were uninsured fell from 9.7% in 2011 to 9.2% in 2012 (Table 1), a statistically significant change.  The Children's Health Insurance Program Reauthorization Act (CHIPRA), signed by President Obama in 2009, has improved public coverage under Medicaid and CHIP with provisions such as Express Lane Eligibility, which allows states to use information collected for other programs; bonus payments to states that do well in enrolling and retaining eligible children; and automatic eligibility for newborns whose mothers are enrolled.³  Over 4 million more children had public coverage in 2012 than in 2008, while the number with private coverage has remained about the same.  However, an estimated 7.2 million children were uninsured in 2012.
• Through 2009, uninsurance rates for 19-25-year-olds were about 5 percentage points higher than those for 26-34-year-olds.  The Affordable Care Act’s expansion of dependent coverage, allowing 19-25-year-olds to be covered under a parent’s employer-sponsored or individually purchased coverage, began to take effect in September 2010 and has completely eliminated this coverage gap, although an estimated 27.2% of each age group remain uninsured in 2012 (Table 1).  Uninsurance rates for these groups were statistically unchanged from 2011
• Estimates of uninsurance rates for ages 35-44 (21.1%), 45-64 (16.2%), and 65 and older (1.5%), also shown in Table 1, were statistically unchanged from 2011.
• Due to Medicare, the elderly make up a very small proportion of the uninsured (Figure 1).  Children, who are eligible for Medicaid or CHIP at higher income levels than parents, also make up a smaller share of the uninsured than of the general population.   Young adults (19-34) make up a larger share of the uninsured than of the general population and are a special focus for Marketplace outreach.  Adults between 35 and 44 are also somewhat more likely to be uninsured than the general population, while adults between 45 and 64 are about the same proportion of the uninsured and of the general population.

Race/Ethnicity and Uninsurance

• An estimated 29.1% of Latinos were uninsured in 2012 (Table 1).  This was a statistically significant 1 percentage point drop from 2011’s 30.1%, but Latinos continue to constitute a much larger share of the uninsured than of the total population.
• An estimated 19.0% of African Americans were uninsured in 2012 (Table 1), statistically unchanged from 2011.
• An estimated 15.1% of Asian Americans were uninsured in 2012 (Table 1).  This was a significantly lower rate than 2011’s 16.8%.  Due to small sample sizes, however, the Census Bureau recommends combining two or more years of data when analyzing changes in insurance rates for Asian Americans.
• An estimated 11.1% of non-Latino Whites were uninsured in 2012 (Table 1), statistically unchanged from 2011.

Figure 1. Profile of the Uninsured vs. Total Population by Age, 2012

Source: ASPE calculations from Carmen DeNavas-Walt, Bernadette D. Proctor, and Jessica C. Smith, U.S. Census Bureau, Current Population Reports, P60-245, Income, Poverty, and Health Insurance Coverage in the United States: 2012, U.S. Government Printing Office, Washington, DC, 2013 (http://www.census.gov/prod/2013pubs/p60-245.pdf), accessed September 17, 2013.

Employment Status and Uninsurance

• Among 18-64-year-olds, an estimated 15.5% of full-time workers and 27.7% of part-time workers were uninsured.  These rates were statistically unchanged from 2011.
• Also among 18-64-year-olds, the estimated share of nonworkers who were uninsured dropped from 26.7% in 2011 to 25.8% in 2012, a statistically significant change.

Household Income and Uninsurance

• In 2012, as in previous years, people living in lower-income households were more likely to be uninsured than people living in higher-income households.  Estimated uninsurance rates were 24.9% for people in households with less than $25,000 in income, compared with 21.4% for people in households with income of $25,000 to $49,999, 15.0% for people in households with income of $50,000 to $74,999, and 7.9% for people in households with income of $75,000 or higher. 
• These rates were statistically unchanged from 2011.
• Beginning in January 2014, many lower and middle income families will be eligible to receive affordable health insurance, either by enrolling in an expanded Medicaid program or by purchasing coverage in the Health Insurance Marketplace.  Families purchasing coverage in the Marketplace may be eligible for tax credits and cost-sharing subsidies.

Sources of Health Insurance Coverage

As discussed above, an estimated 15.4% of the population was uninsured, and this rate was significantly lower than the 2011 estimate of 15.7% uninsured.  The stabilization of employer-sponsored coverage and the growth in Medicare coverage both contributed to this result.

• An estimated 54.9% of the population was covered by employer-sponsored insurance in 2012.  This was not significantly different from the rate in 2011.  From 2000 to 2010, the percentage of Americans with employer-sponsored insurance dropped by about 10 percentage points (Figure 2).  The new CPS-ASEC data suggest that the erosion of employer-sponsored insurance leveled off in 2011 and 2012.
• An estimated 16.4% of the population was covered by Medicaid or CHIP in 2012.  As Figure 2 shows, Medicaid/CHIP coverage grew rapidly from 1999 to 2011, and the number of people covered by these public programs began to exceed both the number of Medicare enrollees and the number of uninsured.  The 2012 rate, however, was statistically unchanged from 2011.
• An estimated 15.7% of the population was covered by Medicare in 2012.  As the baby boom ages, more Americans turn 65 and become eligible for Medicare.  The share of the population covered by Medicare was higher in 2012 than in 2011 and the increase in the number of individuals covered by Medicare, from 46.9 million to 48.9 million, was also statistically significant.  This year-to-year increase in Medicare enrollment is likely to continue into the middle of the next decade.  
• An estimated 9.8% of the population was covered by directly purchased private insurance in 2012, statistically unchanged from 2011.  (Directly purchased private insurance is sometimes described as individual market or nongroup coverage).  The percentage of the population with this form of coverage has been very stable over the period shown in Figure 2.
• An estimated 4.4% of the population was covered by insurance related to current or previous military service in 2012, statistically unchanged from 2011.  The percentage of the population with this form of coverage has grown somewhat over the period shown in Figure 2 but remains small.

Figure 2. Sources of Insurance Coverage, 1999-2012

Source: ASPE calculations from Carmen DeNavas-Walt, Bernadette D. Proctor, and Jessica C. Smith, U.S. Census Bureau, Current Population Reports, P60-245, Income, Poverty, and Health Insurance Coverage in the United States: 2012, U.S. Government Printing Office, Washington, DC, 2013 (http://www.census.gov/prod/2013pubs/p60-245.pdf), accessed September 17, 2013.

The estimates shown in Figure 2 are based on the percentage of people for whom each source of coverage was reported.  These percentages sum to more than 100 percent because some people report multiple forms of coverage.  Some of the elderly and disabled, for example, are “dual eligibles” enrolled in both Medicare and Medicaid.⁶   Because the CPS-ASEC asks about coverage in the prior calendar year, people who had different types of coverage in different parts of the year (for example, people who directly purchased individual market coverage and then got jobs providing employer-sponsored coverage) may also report multiple sources of coverage. 

Figure 3 shows the distribution of sources of coverage when individuals with multiple sources of coverage are assigned to one primary source.  The methods we used to assign primary source of coverage to individuals with multiple sources of coverage are explained in the Appendix.

Figure 3. Primary Source of Coverage, 2012

Source: ASPE computations from Current Population Survey Annual Social and Economic Survey (CPS-ASEC) microdata for Calendar Year 2012.

Table 2 compares the number of people reported as having a particular source of coverage with the number to whom that source was assigned as primary coverage.  The difference is most striking for directly purchased private insurance, which is reported for 30.6 million people, but assigned as primary only to the 11.1 million people who reported it as their sole source of coverage.  This estimate is very close to the 10.9 million covered lives in the individual market reported  in the 2012 administrative data submitted by insurers in connection with the Affordable Care Act’s Medical Loss Ratio (MLR) requirements.⁷

As Table 3 shows, most of the people with directly purchased coverage in the CPS-ASEC data also have something else.  The most common combination is direct purchase and Medicare.  Many Medicare enrollees purchase Medicare Supplemental Health Insurance policies, also known as “Medigap,” to cover costs they would otherwise bear out-of-pocket, ⁸ and some respondents may have  reported such policies in their CPS-ASEC interviews.  Similarly, persons who reported both direct purchase and employer-sponsored coverage may have been thinking of the wrap-around policies they purchased for services such as dental or vision care that were not covered by their employer plans.  And since the CPS-ASEC asks about coverage over the course of the full calendar year, some of the people reporting multiple forms of coverage in 2012 may have had directly purchased policies for some months and other coverage, such as employer-sponsored coverage from a new job, for other months.  People reporting directly purchased policies along with Medicaid  or CHIP may have bought coverage while waiting for public coverage to kick in.

Conclusion

CPS-ASEC health insurance coverage estimates for 2012 continue a two-year reversal of a decade-long increase in the number of uninsured Americans.  There were significant declines in uninsurance rates overall and among subgroups such as children, Latinos, and nonworkers.  Uninsurance rates did not get significantly worse for any of the subgroups discussed in this brief or in the longer Census Bureau report on which it is based. 

Yet in 2012, as in 2011—and in all previous years for which the CPS-ASEC has collected data on health insurance coverage—substantial proportions of Americans in all categories except the elderly were  uninsured.  The upcoming expansion of Medicaid (in participating states) and the establishment of the Health Insurance Marketplace (in all states) under the Affordable Care Act will give millions of Americans affordable options for coverage they did not have before. 

One very important component of the Affordable Care Act’s coverage expansion, the extension of dependent coverage to 19-25-year-olds, has already been in effect for several years.  The new CPS-ASEC data for this age group, compared with data from earlier years, provide fresh evidence that this policy has had the desired impact of increasing health insurance coverage.

Sources of coverage among the insured were fairly stable in 2012.  Most Americans continue to have employer-sponsored insurance, and the long period of decline in this source of coverage appears to have leveled off.  The only statistically significant change in the sources of insurance coverage, the increase in the number of people with Medicare, reflects the demographic shift that we have long known to be coming rather than recent public policy.  As two million or so individuals move into Medicare each year, however, the number of people who benefit from the Affordable Care Act’s improvements in Medicare’s prescription drug coverage (closing the Part D “donut hole”) and preventive services (expanding the types of care available with no cost-sharing) will grow accordingly.

Appendix: Methods for Assigning Primary Source of Coverage

To assign persons for whom more than one source of coverage is reported in the CPS-ASEC to the single, “primary” source of coverage shown in Figure 3 and Table 2, we created a hierarchy.  Because Medicare is considered to provide primary coverage for Medicare-Medicaid dual enrollees, we placed Medicare at the top of the hierarchy.  Everyone who reported Medicare (alone or in combination with other sources of coverage), is thus assigned as Medicare, and the Medicare “Any” and “Primary/Only” estimates in Table 3 are the same.  Next, we assigned anyone not reporting Medicare, but reporting Military coverage, as Military.  We then assigned persons not reporting Medicare or Military, and reporting Medicaid or CHIP, as Medicaid/CHIP.  Persons not reporting Medicare, Military, or Medicaid/CHIP, but reporting either Employer-Sponsored coverage or coverage from outside the household (which we assumed was Employer-Sponsored coverage) were assigned as Employer-Sponsored.  Direct purchase was at the bottom of the hierarchy, so the only people for whom Direct Purchase was assigned as primary were those who did not report any of the other types of coverage.

The particular hierarchy we developed did not affect estimates for those reporting only one source of coverage, or for those reporting as uninsured.

Endnotes

¹ Carmen DeNavas-Walt, Bernadette D. Proctor, and Jessica C. Smith, U.S. Census Bureau, Current Population Reports, P60-245, Income, Poverty, and Health Insurance Coverage in the United States: 2012, U.S. Government Printing Office, Washington, DC, 2013 (http://www.census.gov/prod/2013pubs/p60-245.pdf, accessed September 17, 2013).

² Congressional Budget Office, Effects on Health Insurance and the Federal Budget for the Insurance Coverage Provisions in the Affordable Care Act—May 2013 Baseline(http://www.cbo.gov/sites/default/files/private/cbofiles/attachments/44190_EffectsAffordableCareActHealthInsuranceCoverage_2.pdf, accessed September 17, 2013).

³ See http://www.medicaid.gov/Medicaid-CHIP-Program-Information/By-Topics/Childrens-Health-Insurance-Program-CHIP/CHIPRA.html, accessed September 17, 2013.

⁴ Uninsured rate is the number of uninsured people in the specified category divided by the total number of people in that specific category.    For example, 9.2% of those age 0-18 are uninsured.

⁵ Percent of uninsured is the number of uninsured people in the specified category divided by the total uninsured population.  For example, 15.0% of the uninsured are age 0-18.  Race/Ethnicity data for American Indians and Alaska Natives, Native Hawaiians and other Pacific Islanders, and non-Latinos reporting more than one race are not shown, and for that reason the Race/Ethnicity percentages in Table 1 do not sum to 100%.

⁶ http://www.medicaid.gov/Medicaid-CHIP-Program-Information/By-Population/Medicare-Medicaid-Enrollees-Dual-Eligibles/Seniors-and-Medicare-and-Medicaid-Enrollees.html.

⁷ The MLR data are available at http://www.cms.gov/CCIIO/Resources/Data-Resources/mlr.html.

⁸ Steven Sheingold, Adele Shartzer, and Dan Ly, Variation and Trends in Medigap Premiums, ASPE Report, December 2011 (http://aspe.hhs.gov/health/reports/2011/MedigapPremiums/, accessed November 5, 2013).