The Extent of Overall Health Insurance Coverage

At the end of 2006, the resident population of the United States was approximately 300 million. Roughly 250 million people had some form of health insurance. The operative word is roughly. People obtain health insurance from a variety of sources. Many of them have access to and sometimes coverage from more than one source. Moreover, they may not have coverage for the entire year, and there is no single repository of data on who has what sort of coverage over what period of time. Thus, a person living in a two-earner household may have coverage from both workers, from only one, or from nether. A retiree may have Medicare coverage and a private supplemental policy. An early retiree may not yet be eligible for Medicare but may have coverage through a former employer or may have purchased individual coverage. A college student may have coverage through her parents as long as she is a full-time student, but if she drops a class tomorrow, the coverage may lapse.

The most commonly used data source on overall health insurance coverage is the March Supplement of the Current Population Survey (CPS). This large, nationally representative survey is largely conducted by telephone but uses household visits for those without phones. Members of approximately 50,000 households are interviewed each month, using a somewhat complex set of rule for when respondents answer particular questions. The insurance questions are about coverage in the preceding calendar year, and in principle, the responses relate to the entire year. The Employee Benefit Research Institute (EBRI) releases an easily accessible summary of these data each year and provides a summary of the survey methods.

Because virtually all persons age 65 and older have Medicare coverage, most discussion of the nature of coverage focuses on the non elderly—those under age 65. It shows the number and proportion of the non elderly population that had health insurance in 1994 and in 2005. Note that in each year more than 100 percent of persons surveyed either had coverage or were uninsured. This is because some people have multiple sources of insurance coverage. Note, too, that the sum of the details of the number of people covered by each insurance category is greater than the reported total. The total reflects the unduplicated count.

Approximately 62 percent of the non elderly had coverage through an employer in 2005. Just over half (52 percent) of these had coverage through their own employment; the rest had coverage as a dependent. While the total number of people with employer-sponsored coverage has increased by 11.4 million since 1994, the percentage of non elderly persons with employer-sponsored coverage has declined from 64.4 to 62.0 percent.

The age distribution of those with employer-sponsored health insurance is generally U-shaped. Nearly 60 percent of children under age 18 have such coverage. The percentage drops to a low of 43.7 percent from ages 21–24, largely because children end their schooling, are dropped from their parents’ coverage, and either do not have a job that offers coverage or decline coverage that is offered. The percentage then rises through ages 45–54, at which point 71.7 percent have coverage, but declines among older workers. Those in the 55–64 age groups are somewhat less likely to report having employer-sponsored coverage (67.3 percent), in part because of early retirement and in part because of health problems that have led them to stop working.

Employer-sponsored coverage differs widely by industry. Slightly more than one-half of workers in the agricultural, forestry, fishing, mining, and construction industries had coverage in 2005. In contrast, public-sector workers are most likely to be covered; with nearly 90 percent of them having employer-sponsored health insurance in 2005.

Individual coverage is private insurance that is not purchased through a group. Some 6.8 percent of the non elderly had such coverage in 2005. The individual market is of considerable interest to policy-makers. Some would use tax credits to encourage the uninsured to buy individual coverage. Others would change the current tax incentives that favor employer-sponsored coverage to ones that encourage individual coverage. Young adults ages 18–24 had the largest probability of buying such coverage in 2004. This is undoubtedly a reaction by many to the lack of employer-sponsored coverage. Approximately 7.8 percent of older workers ages 55–64 had individual coverage in 2004. Many of these individuals had retired from an employer, some starting another career, and were not yet eligible for Medicare.

Medicare covered some 2.5 percent of the non elderly in 2005. Many of these people are eligible for Medicare due to disability; others are covered dependents or spouses of Medicare beneficiaries. The number of non elderly residents with Medicare coverage increased 75 percent between 1994 and2005. Much of this increase stems from Medicare’s more-generous interpretation of disability.

Medicaid is a joint federal-state insurance program for low-income individuals. The number of Medicaid recipients increased by over 19 percent between 1994 and 2005. The CPS data indicate that, in 2005, Medicaid provided coverage to some 13.5 percent of non elderly U.S. residents. This percentage is an understatement of Medicaid coverage for two important reasons. First, as noted later, Medicaid is the primary source of nursing home coverage in the United States. The vast majority of persons in nursing homes are over age 65. Second, there is some suspicion that the CPS methodology inadvertently under counts Medicaid eligible. The Congressional Research Service (2005) compared the CPS estimates with two other surveys that are designed to focus on low-income people and concluded that, in 2004, the CPS understated the number of Medicaid adults by some 5.8 million and overstated the number of uninsured by 3.7 million.

Military coverage refers to coverage of U.S. military retirees and dependents of active-duty, retired and deceased service members through the Tricare Program. It also includes CHAMPVA, the Civilian Health and Medical Program of the Department of Veterans Affairs for dependents with disabilities and certain survivors of veterans. In 2005, these programs covered almost 3 percent of the non elderly U.S.