Many Americans incorrectly believe that none of the poor has private health insurance. However, at least some of the new Medicaid enrollees would be those who previously had private coverage.

“Crowd-out” (or substitution) occurs when people who are already covered by employer-paid or individual insurance drop that coverage to take advantage of a public insurance option. Estimates of crowd-out are controversial among analysts. Some researchers find a high rate of Medicaid substitution for private coverage, while others believe it is negligible. Estimates of crowd-out for diverse populations vary.

Nevertheless, crowd-out is likely to be a significant problem for states that expand Medicaid eligibility to adults who are not disabled. For instance, analysis of past Medicaid expansions to mothers and children in the early 1990s by David Cutler and Jonathan Gruber, economists who have since served as Obama administration advisers, found that when Medicaid eligibility was expanded to children and pregnant women, about 50 percent of the newly enrolled had dropped private coverage.[38] A recent analysis by Gruber and Kosali Simon estimated that crowd-out for the Children’s Health Insurance Program averaged about 60 percent.[*]

Working adults are the target of Medicaid expansion under the ACA. The likely effect of crowd-out on working adults was analyzed for the Veterans Health Administration by academic researchers Steven Pizer, Austin Frakt and Lisa Iezzoni. They found crowd-out could reach 82 percent.[39]

A conservative estimate is that Medicaid rolls might have to rise by 1.4 people in order to reduce the uninsured by 1 person — a crowd-out rate of about 29 percent.[†] Medicaid, after all, will cost less money for many low-income people who are privately insured and provide, on paper at least, much better coverage.

According to the Kaiser Family Foundation, more than one in five nonelderly people with incomes of 138 percent or less of the poverty level in Michigan in 2011 were covered by private insurance.[40] This is about 512,000 people[41] — nearly half the 1.1 million nonelderly people in Michigan who are already enrolled in Medicaid and earning 138 percent or less of the poverty level.[42]

Approximately 685,526 Michigan residents were enrolled in Michigan degree-granting colleges and universities in the fall of 2010, according to the National Center for Education Statistics.[43] Many college students would be eligible for Medicaid under expansion rules, though many of them are likely eligible for coverage under their parents’ private insurance plans.

A decade ago, an estimated 29 percent of adults ages 25 to 64 with incomes below the poverty line purchased private health insurance.[44] Among individuals with family incomes less than 1.5 times the federal poverty level, an estimated 36 percent of individuals — more than one-third — had private insurance.[45] These are probably conservative projections of the percentage of people who would qualify under an expanded Medicaid program and who currently have private insurance. The definition of poverty in the ACA differs substantially from the official one. As a result, a preliminary analysis by staff from the U.S. Treasury Department found the Current Population Survey data could significantly underestimate the number that will be newly eligible for Medicaid, potentially driving up the financial burden on the public and increasing the crowd-out of the private insurance market.[46]

Crowd-out is an important concern. Substituting public coverage for private insurance shifts costs being borne by the private sector to taxpayers without boosting health coverage.

[*] The actual rate varied depending on the conditions governing expansion and the populations covered. The authors conclude, however, “Our results clearly show that crowd-out is significant; the central tendency in our results is a crowd-out rate of about 60%.” Jonathan Gruber and Kosali Simon, “Crowd-Out Ten Years Later: Have Recent Public Insurance Expansions Crowded Out Private Health Insurance?,” (National Bureau of Economic Research, 2007), ii, (accessed June 2, 2013).

[†] A ratio of 1.4 new Medicaid enrollees to reduce the uninsured by 1 assumes a crowd-out rate of 29 percent (1-(1/1.4)). One analysis found about one-quarter of the newly insured children in families earning less than 200 percent of poverty had substituted public coverage for private coverage. Peter J. Cunningham, James D. Reschovsky and Jack Hadley, “SCHIP, Medicaid Expansions Lead to Shifts in Children’s Coverage,” (Center for Studying Health System Change, 2002), 4, (accessed May 17, 2013).