Congress is engaged in an attempt to end the federal stalemate over reauthorization of the State Children's Health Insurance Program (SCHIP), which is set to expire at the end of the month. Rather than refocus SCHIP as a targeted safety net for low-income uninsured children, the two chambers passed bills to extend the safety net to children in families with significantly higher incomes.
Expanding SCHIP eligibility further up the income ladder is not a good way to help families that lack insurance yet have incomes above the current federal threshold. Enrolling children in families at these income levels is inefficient and will disrupt the private coverage many of them have today. This is because government programs and taxpayer dollars will increasingly become substitutes for private coverage and funding. This policy-induced phenomenon, known as "crowd out," substantially increases the cost of covering uninsured children.
The Heritage Foundation conducted an econometric analysis of the likely crowd out associated with the House and Senate bills. This analysis was based on a modified and extended version of the methodology developed by MIT professor Jonathan Gruber, a leading expert on the crowd-out effect. This analysis found that Congress's expansion proposals for SCHIP could cover as many as 2.4 million newly eligible children, but because of crowd out, the ranks of the uninsured would decrease by only 1 million. This is because, for every 100 newly eligible children in families with incomes between 200 and 400 percent of the federal poverty level (FPL), 54 to 60 children would lose the private coverage that they have today. 
To avoid undue costs and ensure that the program effectively and efficiently serves its intended purpose, Congress should change course and focus SCHIP on uninsured children in low-income families. To accomplish this, it should provide tax relief or direct assistance to needy families currently unable to afford to enroll their children in available private coverage.
When Congress reauthorizes SCHIP, it should keep in mind two important points. First, policies that expand eligibility thresholds cause children to lose private health insurance, which is often replaced by public programs. Second, estimates of this crowd out and its costs are significant and supported by most research in the field.
SCHIP and the Crowd-Out Effect
Most of the debate over SCHIP reauthorization hinges on expanding program eligibility to children in higher income families. While expanding SCHIP eligibility would, to some extent, reduce the ranks of uninsured children, these gains would be significantly offset-or even outpaced-by losses in private insurance. As the safety net is cast further up the income ladder, instead of complementing private coverage and reducing the ranks of the uninsured, SCHIP would increasingly become a substitute for it.
The Congressional Budget Office (CBO) recently conducted a literature review to estimate crowd out due to previous SCHIP expansions. CBO estimates crowd out for these expansions is between 25 and 50 percent. In other words, one quarter to one half of newly enrolled children would have otherwise had private coverage. Indeed, most leading studies of SCHIP expansions find crowd out of this magnitude. Moreover, it is generally agreed that the magnitude of the crowd-out effect will grow with further eligibility expansions because an overwhelming majority of newly eligible children already have private coverage to lose.  Yet recent studies have not estimated the crowd-out effects of SCHIP expansions relative to income eligibility thresholds. To fill this gap in the literature and to estimate the potential crowd-out effects if Congress were to expand SCHIP to children from families with higher incomes, The Heritage Foundation conducted its own econometric study.
Congress's SCHIP Eligibility Expansions and Crowd
The Heritage Foundation estimates show that SCHIP expansions have significantly substituted government programs for private coverage among newly eligible children. Moreover, this effect grows significantly in magnitude as children from higher income families become eligible. (See Table 1.)
On the aggregate, for every 100 children newly eligible for SCHIP, between 30 and 35 children lose private coverage. Disaggregating the analysis by income eligibility thresholds, however, indicates crowd out grows in magnitude when the program is extended beyond its intended focus of covering uninsured children in families below 200 percent of the FPL In summary, Heritage finds:
- For every 100 newly eligible children in families with incomes between 100 and 200 percent of the FPL, 34 to 42 children would lose private coverage;
- For every 100 newly eligible children in families with incomes between 200 and 300 percent of the FPL, 44 to 51 children would lose private coverage; and
- For every 100 newly eligible children in families with incomes between 200 and 400 percent of the FPL, 54 to 60 would lose private coverage.
The Cost of Senate and House Expansions
Estimating the magnitude of crowd out as a result of SCHIP expansions is important because, as the program becomes a substitute for private coverage, assistance flows to families whose children would have otherwise had insurance and away from children who currently go without. Because crowd out causes the ranks of the uninsured to decrease less than expected on a static basis, it increases the cost to the taxpayer of covering the uninsured. For this reason, despite what some in Congress might think, expanding SCHIP eligibility is a costly way to reduce the ranks of uninsured children.
Under the Senate's SCHIP expansion, an estimated 1 million to 1.2 million children would gain SCHIP coverage, but between 467,000 and 611,000 children would lose private coverage. Due to poor targeting and the relative cost of crowd out, the annual cost to taxpayers of covering an uninsured child under the Senate's expansion plan would increase from $1,418 to between $2,508 and $2,859. This is 1.8 to 2 times the cost of SCHIP coverage for a child in a family at this income level or almost 2.5 times the average cost of private insurance.
Under the House's SCHIP expansion, an estimated 2.2 million to 2.4 million children would gain SCHIP coverage, but between 1.2 and 1.5 million children would lose private coverage. Due to poor targeting and the relative cost of crowd out, the annual cost to taxpayers of covering an uninsured child under the House's expansion plan would increase from $1,612 to between $3,485 and $4,008. This is 2.2 to 2.5 times the cost of SCHIP coverage for a child in a family at this income level or almost 3.5 times the average cost of private insurance.
Expanding SCHIP to cover children in higher income families is not an efficient or cost-effective way to reduce the ranks of uninsured children. As the safety net is cast further up the income ladder, it will increasingly substitute government programs and taxpayer dollars for private coverage and funding. In order to avoid significant and increasing crowd out, and to optimize the program's "bang for the buck," Congress should abandon its current course before SCHIP expires. Congress should, then, restore SCHIP's purpose as a targeted safety net for uninsured children in low-income families and work to more efficiently and effectively direct assistance to those most in need.
Paul Winfree is Policy Analyst in the Center for Data Analysis, and Greg D'Angelo is Research Assistant in the Center for Health Policy Studies, at The Heritage Foundation. The authors gratefully acknowledge the assistance of Andrew Nowobiliski, intern in the Center for Data Analysis at The Heritage Foundation.
 In 2004, the average annual premium for private health insurance for children under 18 was $1,183. See America's Health Insurance Plans, "Individual Health Insurance: A Comprehensive Survey of Affordability, Access, and Benefits," August 2005, p. 5, at www.ahipresearch.org/pdfs/Individual_Insurance_Survey_Report8-26-2005.pdf.
 "Section 2101(a) of the Social Security Act describes the purpose of the SCHIP statute 'to initiate and expand the provision of child health assistance to uninsured, low-income children in an effective and efficient manner that is coordinated with other sources of health benefits coverage.'" See Letter from Dennis Smith, Director of Center for Medicaid and State Operations, Center for Medicare & Medicaid Services, to State Health Officials, August 17, 2007, at www.cms.hhs.gov/smdl/downloads/SHO081707.pdf.
 See Congressional Budget Office, "The State Children's Health Insurance Program," May 2007, at www.cbo.gov/ftpdocs/80xx/doc8092/05-10-SCHIP.pdf, and Jonathan Gruber and Kosali Simon, "Crowd-Out Ten Years Later: Have Recent Public Program Expansions Crowded Out Private Health Insurance?," National Bureau of Economic Research, Working Paper No. 12858, at www.nber.org/papers/w12858.
 Congressional Budget Office, "The State Children's Health Insurance Program."
 "Fully 61 percent of children who became eligible for public insurance due to the creation of SCHIP already had private coverage. As SCHIP grows to allow children from wealthier families, this figure will rise. According to a CBO analysis of Census data, current proposals in Congress to expand SCHIP eligibility would reach children in income groups in which 89 percent or more of children currently have private coverage." See Andrew M. Grossman and Greg D'Angelo, "SCHIP and 'Crowd-Out': How Public Program Expansion Reduces Private Coverage," Heritage Foundation WebMemo No. 1518, June 21, 2007, at www.heritage.org/Research/HealthCare/wm1518.cfm.
 To measure the level of crowd out, we employ the 1996 and 2001 panels of the Survey of Income and Program Participation (SIPP), which capture the years between 1996 and 2003. In part, the SIPP was chosen because it is known for producing relatively small crowd-out effects. Also SIPP data best capture the overlap group reporting public and private coverage, which is critical to accurately measuring crowd out. Furthermore, in order to strengthen the validity of our crowd-out estimates, we employ an instrumental variables approach to estimate the impact of changes in SCHIP eligibility on insurance status. For further discussion of this approach, see Gruber and Simon, "Crowd-Out Ten Years Later."
 Under the Senate's expansion proposal, children in families with incomes up to 300 percent of the FPL could become eligible for the program, and the legislation would not affect states with program eligibility already beyond that threshold.
 See America's Health Insurance Plans, "Individual Health Insurance," p. 5.
 Under the House's expansion, there is virtually no upper bound on eligibility thresholds. Although the bill is widely considered more expansive than the Senate's, it is somewhat unclear exactly how far up the income ladder it would expand the program. Early versions of the bill would have expanded SCHIP to children in families with incomes up to 400 percent of the FPL-$82,600 dollars for a family of four. Subsequently, the explicit eligibility expansion threshold was removed from the legislation. Instead, the bill would allow and encourage state governments to significantly expand their programs. Because expanding SCHIP to 400 percent of the FPL was the House's stated policy goal, and because no state has gone beyond that level to date, our analysis assumes that states could expand their SCHIP programs to cover children up to that eligibility threshold.
 America's Health Insurance Plans, "Individual Health Insurance."