Home / State Resource Center / Strategy Center / Strategies / Addressing Crowd Out / Data

Addressing Crowd Out

Data Highlights It is clear that SCHIP is a successful program, but less clear as to how much of an issue crowd out is. Most analyses seem to agree that some crowd out of private coverage does exist when states expand public programs, and that crowd out increases when they expand further up the income scale. But, there are widely varying estimates as to the magnitude of the crowd out effect, in part, because of varying definitions of crowd out and methodologies for estimating it. Despite these challenges, many researchers have investigated the magnitude of crowd out in public programs. The most widely used findings, which are discussed in more detail below, include: Data collected in ten states as part of the Congressionally mandated evaluation of SCHIP found that very few families—at most one in fifteen families— actually drop affordable private coverage that they already have in order to enroll in SCHIP. 1 The Congressional Budget Office, relying on a broader definition of crowd out that, for example, considers the response of employers to public expansions, has concluded that probably for every 100 children who gain coverage as a result of SCHIP, there is a corresponding reduction in private coverage of between 25 and 50 children. 2 It, however, assumes that initiatives to reach more uninsured children already eligible for Medicaid results in a reduction in private coverage of 20 children.

Table of Contents Summary Framing the Issue Legislative Authority Data Strategies Issues to Consider States Experiences Resources Download Brief (PDF)

Detailed Discussion of the Data
Studies on the crowd out effect of SCHIP rely on three kinds of data sources: population-based surveys, surveys of SCHIP enrollees, and state administrative data based on applications. As a result of different data limitations and methodologies, these studies have shown a wide range of crowd out estimates from zero to 60 percent. Given this wide range of findings, all crowd out estimates should be interpreted with caution.

• Population-based studies. One type of crowd out study uses population-based survey data and econometric techniques (which use assumptions to model changes in the behavior of families and employers) to determine the number of SCHIP and Medicaid-eligible children who would have had private coverage in the absence of public programs. Due to data limitations, estimates from population-based studies are imprecise and sensitive to assumptions; they also do not account for all the nuances of state-specific factors affecting eligibility determinations or economic conditions affecting insurance coverage.

• Population-based studies tend to find the greatest amount of crowd out, generally between 20 and 40 percent, 3 but as low as 10 percent 4 and as high as 60 percent. 5

• The one study to examine employer behavior in response to SCHIP found that SCHIP was not associated with an employer's decision to offer individual or family coverage, but was, however, associated with an increase in the employee contribution/cost. 6

• Applicant-based studies. Other crowd out studies use state administrative data based on applications to measure the percent of SCHIP-enrolled children who dropped private coverage or the percent of applications denied because of other coverage. Some states consider dropping of private coverage for certain reasons to be acceptable, and applicant-based studies generally reflect this state perspective, however these studies do not control for other factors, such as changes in employer behaviors.

• Using applicant-based studies in the first few years of SCHIP, Connecticut, New Hampshire, New Jersey, Oregon, and Pennsylvania reported virtually no crowd out; California estimated that four percent of enrollees dropped coverage; and the District of Columbia reported that 15 percent of applicants dropped health insurance within the three months before applying. 7

• In 2004, of 34 states reporting, almost all (33 states) found that 15 percent or less of applicants were denied eligibility because they had other coverage, and most (28 states) reported rates of less than 10 percent. 8
  

• Of the 23 states with data on applicants who dropped coverage before applying to SCHIP, about half (13 states) reported rates of less than 1 percent, and most (20 states) reported rates of less than 6 percent, but one state reported a rate as high as 16 percent. 9

• Enrollee-based studies. Another type of crowd out study surveys parents of children enrolled in SCHIP to measure how many SCHIP-enrolled children had private coverage before enrollment in SCHIP. Enrollee-based studies do not control for employer behavior or account for applicants who were denied SCHIP because of other coverage.

• An analysis of data collected in ten states as part of the Congressionally mandated SCHIP evaluation found that among the 28 percent of SCHIP enrollees that had private coverage in the six months before enrolling in SCHIP, half (14 percent) lost private coverage due to a change in employment or family structure, one quarter (8 percent) cited affordability issues for ending private coverage, meaning that less than seven percent of recent SCHIP enrollees had dropped their private coverage for reasons unrelated to affordability or to loss of coverage due to changes in family structure or employment. 10 (See Graph below.)
  

• Other state-based surveys of SCHIP enrollees have found crowd out rates of less than 10 percent. 11
  
  

 

Go To Next Section:

Strategies

---

Footnotes

 1. A. Sommers, et al., “Substitution of SCHIP For Private Coverage: Results From A 2002 Evaluation in Ten States,” Health Affairs, 26: 529-537 (2007). Back

2. Congressional Budget Office, “The State Children’s Health Insurance Program,” (May 2007). Back

3. For example, see P. Cunningham, J. Reschovsky, & J. Hadley, “SCHIP, Medicaid Expansions Lead to Shifts in Children’s Coverage,” Center for Studying Health System Change (December 2002); P. Cunningham, J. Hadley, & J. Reschovsky, “The Effects of SCHIP on Children’s Health Insurance Coverage: Early Evidence from the Community Tracking Survey,” Medical Care Research and Review, 59: 359-383 (December 2002); H. J. Lee & W. Tian, “The State Children’s Health Insurance Program: Participation and Substitution,” Economic Research Initiative on the Uninsured, Working Paper #53 (October 2004); A. Davidoff, G. Kenney, & L. Dubay, “Effects of the State Children's Health Insurance Program Expansions on Children With Chronic Health Conditions,” Pediatrics, 116: e34-e42 (July 2005); J. Hudson, T. Selden, & J. Banthin, “The Impact of SCHIP on Insurance Coverage of Children,” Inquiry, 42: 232-254 (Fall 2005); and C. Banksak & S. Raphael, “The Effects of State Policy Design Features on Take-up and Crowd-Out Rates for the State Children’s Health Insurance Program”, Journal of Policy Analysis and Management, 26: 149-175 (2006). Back

4. A. Lo Sasso & T. Buchmueller, “The Effect of State Children’s Health Insurance Program on Health Insurance Coverage,” Journal of Health Economics, 23: 1059-1082 (September 2004). Back

5. J. Gruber & K. Simon, “Crowd-out 10 Years Later: Have Recent Public Expansions Crowded Out Private Health Insurance,” Journal of Health Economics, in press (2008); also available as NBER Working Paper #12858 (January 2007). Back

6. T. Buchmueller, et al., “The Effect of SCHIP Expansions on Health Insurance Decisions by Employers,” Inquiry 42: 218-231 (Fall 2005). Back

7. M. Rosenbach, et al., “Implementation of the State Children’s Health Insurance Program: Synthesis of State Evaluations,” Mathematica Policy Research, Inc., (March 2003). Back

8. S. Limpa-Amara, A. Merrill, & M. Rosenbach, “SCHIP at 10: A Synthesis of the Evidence on Substitution of SCHIP for Other Coverage,” Mathematica Policy Research, Inc., (September 2007). Back

9. ibid. Back

10. op. cit. (1). Back

11. For example, see E. Shenkman, et al., “Crowd-out: Evidence from the Florida Healthy Kids Program,” Pediatrics 104: 507-513 (1999); E. Feinberg, et al., “Family Income and Crowd Out Among Children Enrolled in Massachusetts Children’s Medical Security Plan,” Health Services Research, 36: 45-63 (December 2001); D. Hughes, J. Angeles, & E. Stilling, “Crowd-Out in the Healthy Families Program: Does It Exist?,” Institute for Health Policy Studies, University of California, San Francisco, (August 2002); and L. Shone, et al., “Crowd-Out in the State Children's Health Insurance Program (SCHIP): Incidence, Enrollee Characteristics and Experiences, and Potential Impact on New York's SCHIP,” Health Services Research, 43: 419-434 (February 2008). Back