• Covering Uninsured Children in the State Children’s Health Insurance Program

Congressional Budget Office
Testimony
Statement of
Peter R. Orszag
Director
Covering Uninsured Children in the
State Children’s Health
Insurance Program
before the
Subcommittee on Health
Committee on Energy and Commerce
U.S. House of Representatives
May 15, 2008
This document is embargoed until it is delivered at
10:00 a.m. (EDT) on Thursday, May 15, 2008. The
contents may not be published, transmitted, or otherwise
communicated by any print, broadcast, or electronic
media before that time.
CONGRESSIONAL BUDGET OFFICE
SECOND AND D STREETS, S.W.
WASHINGTON, D.C. 20515
Chairman Pallone, Congressman Deal, and Members of the Subcommittee, it is my
pleasure to appear before you today to discuss the State Children’s Health Insurance
Program (SCHIP). My testimony makes the following main points:
B SCHIP has significantly reduced the number of low-income children who lack
health insurance. According to the Congressional Budget Office’s (CBO’s) esti-
mates, the portion of children in families with income between 100 percent and
200 percent of the poverty level who were uninsured fell by about 25 percent
between 1996 (the year before SCHIP was enacted) and 2006. In contrast, the rate
of uninsurance among higher-income children remained relatively stable during
that period. The difference probably reflects the impact of the SCHIP program.
B The states’ outreach efforts and simplified enrollment processes for SCHIP appear
to have also increased the share of eligible children who participate in Medicaid—
and contributed to a decline in the percentage of children living below the poverty
level who are uninsured.
B The enrollment of children in public coverage as a result of SCHIP has not led to a
one-for-one reduction in the number of low-income children who are uninsured,
however. Almost any increase in government spending or tax expenditures
intended to expand health insurance coverage will displace private coverage to
some degree. In the specific case of SCHIP, the program provides a source of cover-
age that is less expensive to enrollees and often provides a broader range of benefits
than alternative coverage. As a result, the program displaces—or “crowds out”—
private coverage to some extent. On the basis of a review of available research,
CBO has concluded that for every 100 children who gain public coverage as a
result of SCHIP, there is a corresponding reduction in private coverage of between
25 and 50 children.
B CBO’s analysis of the Children’s Health Insurance Program Reauthorization Act of
2007, as passed by the House of Representatives, suggested that the legislation
would result in 5.8 million children gaining coverage under Medicaid or SCHIP in
2012. Of that increase, CBO estimated, 3.8 million children would otherwise have
been uninsured, and 2.0 million children would otherwise have had private cover-
age. In other words, about one-third of the children who would be newly covered
under SCHIP and Medicaid would otherwise have had private coverage. That
crowd-out rate is probably about as low as feasible for a voluntary program to
increase coverage among children, given the size of the proposed expansion. (Poli-
cies to reduce the rate below that level would most likely also reduce the number of
children enrolled in the program who would otherwise be uninsured.)
B On August 17, 2007, the Centers for Medicare and Medicaid Services (CMS)
issued a directive to state health officials that imposes certain minimum require-
ments on states seeking to enroll children in SCHIP whose families have income
above 250 percent of the poverty level. CBO’s analysis suggests that the directive’s
impact on enrollment is likely to be modest under current law, given the way CMS
appears to be implementing it and, more important, given the funding levels
assumed in CBO’s baseline. The directive could have a substantially larger impact
on enrollment in SCHIP if the Congress expanded the program significantly.
B On May 7, 2008, CMS released a follow-up letter clarifying certain aspects of the
August 17 directive. The clarifications issued by CMS are generally consistent with
how CBO originally interpreted the August 17 letter; therefore, CBO has not
altered its estimates of the policy’s impact on the cost and coverage of SCHIP.
Overview of the State Children’s Health Insurance Program
The State Children’s Health Insurance Program was established by the Balanced Bud-
get Act of 1997 to expand health insurance coverage to uninsured children in families
with income that is modest but too high to qualify for Medicaid. SCHIP is financed
jointly by the federal government and the states, and it is administered by the states
within broad federal guidelines. The Congress provided approximately $40 billion in
funding for SCHIP for fiscal years 1998 through 2007. The Medicare, Medicaid, and
SCHIP Extension Act of 2007 (MMSEA) extended funding for the program through
March 2009.
Eligibility and Enrollment
States have considerable flexibility in designing their eligibility requirements for
SCHIP. According to the SCHIP statute, states may cover children living in families
with income up to 200 percent of the federal poverty level or 50 percentage points
above their Medicaid threshold.1 States are allowed to disregard certain types of
income and expenses in determining eligibility for the program. In 2008, 23 states
allow a maximum income equal to 200 percent of the poverty level, 20 states set the
limit above 200 percent of the poverty level, and 7 states set it below 200 percent of
the poverty level.2 North Dakota has the lowest threshold, at 140 percent of the pov-
erty level, while New Jersey has the highest, at 350 percent of the poverty level.3
1. States are required to maintain the Medicaid threshold (or level of income determining eligibility)
that was in place just before SCHIP was enacted. That requirement, for what is termed “mainte-
nance of effort,” prevents states from lowering their Medicaid threshold in order to receive a higher
matching rate under SCHIP for children who otherwise would have been covered by Medicaid.
2. See Elicia J. Herz, Chris L. Peterson, and Evelyne P. Baumrucker, State Children’s Health Insurance
Program (SCHIP): A Brief Overview, CRS Report for Congress RL30473 (Congressional Research
Service, March 12, 2008).
3. New Jersey has effectively expanded its threshold to 350 percent of the poverty level by disregarding
all income between 200 percent and 350 percent of the poverty level.
2
CBO
Table 1.
Enrollment in the State Children’s Health Insurance
Program, 1998 to 2006
Percentage Percentage
Number of Change from Number of Change from
Fiscal Children Previous Adults Previous
Year (Thousands) Year (Thousands) Year
1998 660 n.a. 0 n.a.
1999 2,014 205 0 n.a.
2000 3,358 67 0 n.a.
2001 4,603 37 234 n.a.
2002 5,354 16 374 60
2003 5,985 12 484 29
2004 6,103 2 646 33
2005 6,114 0 639 -1
2006 6,745 9 671 5
2007a 7,145 6 587 -13
Source: Department of Health and Human Services, Centers for Medicare and Medicaid Services.
Notes: n.a. = not applicable.
The figures for the number of people enrolled reflect enrollment at any time during the year.
The number of people enrolled in an average month would be about 60 percent of the above
totals. There was a change in reporting between 2004 and 2005. Prior to 2005, in states with
a combination program, children enrolled in both the Medicaid expansion and the separate
program during a given year were counted twice. Starting in 2005, however, those children
were counted only in the program where they were last enrolled.
a. Preliminary.
A number of states have used waivers of statutory provisions to expand coverage under
SCHIP to adults. About 80 percent of the adults who were enrolled in SCHIP in
2007 were parents, 19 percent were childless adults, and 1 percent were pregnant
women. Covering parents may help increase participation among children because
parents who are eligible may be more likely to enroll their children also.
The number of children enrolled in SCHIP at any time during the year increased
from 660,000 in 1998 to 7.1 million in 2007 (see Table 1). As states first imple-
mented their programs, enrollment grew very rapidly, reaching almost 6 million chil-
dren by 2003. Since then, enrollment has grown more slowly as states’ programs have
matured and some states have enacted policies to restrict enrollment in response to
budgetary pressures. About 587,000 adults were enrolled at some point during 2007.
Benefits
States can provide SCHIP coverage by expanding Medicaid to include children not
eligible for that program, creating a separate program under SCHIP, or using a combi-
nation of the two approaches. In 2008, 8 states are using an expansion under Medic-
aid, 18 states operate a separate program, and 24 states are using a combination
3
CBO
approach.4 States that provide SCHIP coverage by expanding Medicaid must provide
the same benefits that are available under their Medicaid program and follow all other
requirements of that program. States that create a separate program under SCHIP are
subject to certain minimum standards, including providing a benefit package that is
based on one of several specified “benchmark” insurance plans or an alternative that is
actuarially equivalent or otherwise approved by the federal government.
The Financing of SCHIP
The statute that established SCHIP set national funding levels for each year from
1998 to 2007. In addition, it specified a formula for determining each state’s share of
the federal funding, a matching rate for federal reimbursement of SCHIP spending,
and a mechanism for redistributing states’ unused SCHIP funds.
The annual funding levels specified in the original SCHIP legislation were as follows:
for 1998 through 2001, roughly $4.2 billion annually; for 2002 through 2004, about
$3.2 billion per year; for 2005 and 2006, $4 billion per year; and for 2007, $5 billion.
MMSEA provided $5 billion for 2008 and that same amount for 2009 (which is
available to the states through March 2009) and up to $1.6 billion in additional funds
for 2008 and $275 million in additional funds in 2009 to be used for states that
exhaust their federal funds.
Each year, federal funding for SCHIP is allocated among states on the basis of a for-
mula that takes into account the number of children in low-income families in each
state, the number of such children who are uninsured, and wages in the health services
sector in the state relative to the national average. States must provide matching funds
for expenditures from their federal allotments and have up to three years to spend
those allotments. Funds that are not spent within three years are redistributed to states
that have exhausted their allotments and are made available to those states for an addi-
tional year.
To encourage states to participate in SCHIP, the federal government pays a higher
share of their spending on SCHIP than it pays for Medicaid. The federal govern-
ment’s matching rate for SCHIP varies among states from 65 percent to 83 percent;
the federal matching rate for Medicaid varies from 50 percent to 76 percent. 5 The
national average matching rate for SCHIP is 70 percent and for Medicaid, 57 percent.
Although federal funding is made available on a matching basis for both programs,
the nature of the programs differs significantly because SCHIP is a grant program in
which federal spending is capped in advance whereas Medicaid is an entitlement pro-
gram with no predetermined limit on spending.
4. See Herz, Peterson, and Baumrucker, State Children’s Health Insurance Program: A Brief Overview.
5. SCHIP’s formula for determining the matching rate is based on the state’s federal medical assistance
percentage (FMAP), as used in the Medicaid program, and equals FMAP + 0.3 * (100 - FMAP),
with an upper limit of 85 percent.
4
CBO
Rules for the redistribution of unused funds have been amended a number of times,
both by extending and shortening the periods during which unspent funds are avail-
able. Because states were initially slow in spending their allotments, the Congress
allowed the states to retain some of their allotments longer than three years. In con-
trast, because recent spending has outpaced federal funding, the National Institutes of
Health Reform Act of 2006 (Public Law 109-482) required that a portion of unspent
2005 allotments be redistributed in 2007 instead of 2008.
The type of program that a state operates under SCHIP has distinct implications for
funding levels. States choosing to implement SCHIP by expanding Medicaid may
continue receiving federal matching funds at that program’s lower federal matching
rate once their SCHIP spending exceeds their available funds. In contrast, states oper-
ating a separate program receive federal matching funds (at the enhanced rate) only
up to the amount determined by the allocation formula (unless they convert their
program to a Medicaid expansion).
Expenditures for SCHIP
Initially, federal spending on SCHIP was well below the annual funding levels, as
states implemented their programs (see Table 2). However, since 2002, federal spend-
ing has exceeded the annual allotments every year. Because unspent funds from previ-
ous years and the redistribution of other states’ unspent funds provide additional
SCHIP financing for some states, those states have forestalled exhausting their federal
funds. Recently, however, some states have had insufficient federal funds available to
fully match their SCHIP spending. As a result, the Congress has acted several times to
provide additional funding. The Deficit Reduction Act of 2005 (P.L. 109-171) appro-
priated an extra $283 million in federal funding to support states’ SCHIP spending in
2006. The National Institutes of Health Reform Act of 2006 included provisions
modifying the redistribution of unspent funds from previous years to provide addi-
tional funds in 2007.6 The U.S. Troop Readiness, Veterans’ Care, Katrina Recovery,
and Iraq Accountability Appropriations Act, 2007 (P.L. 110-28) appropriated up to
$650 million in additional federal funding. Most recently, MMSEA provided up to
$1.6 billion in additional funds for 2008 and $275 million in additional funds for
2009 to cover states’ spending through March 2009.
6. The National Institutes of Health Reform Act of 2006 reduced the availability of 2005 allotments
in some states from three years to two and a half. Specifically, states forfeited half of their unspent
2005 funds (not exceeding $20 million) if their total available funds as of March 31, 2007, were at
least twice their projected spending in 2007. The law also specified that spending in 2007 from
redistributed funds on adults who were not pregnant would be reimbursed at Medicaid’s lower
matching rate.
5
CBO
Table 2.
Allotments and Spending Under the State Children’s Health
Insurance Program, 1998 to 2007
(Millions of dollars)
Allotments
Fiscal SCHIP Unspent After Federal Funds
a b
Year Allotments 3 Years Spending Expiring
1998 4,235 n.a. 122 0
1999 4,247 n.a. 922 0
2000 4,249 n.a. 1,929 0
2001 4,249 2,034 2,672 0
2002 3,115 2,819 3,776 0
2003 3,175 2,206 4,276 0
2004 3,175 1,749 4,645 1,281
2005 4,082 643 5,089 128
c
2006 4,365 173 5,452 0
2007 5,040 62 6,000 0
2008d 6,000 58 7,094 0
Sources: Congressional Budget Office, Congressional Research Service, the Balanced Budget Act of
1997, the Deficit Reduction Act of 2005, and the Centers for Medicare and Medicaid
Services.
a. For both states and territories.
b. In general, states’ annual allotments are available for three fiscal years. Any funds unspent
after three years become available to other states with projected spending in excess of their
allocation plus any available funds from previous years.
c. Includes additional funding from the Deficit Reduction Act of 2005.
d. Projection by the Congressional Budget Office.
The Effect of SCHIP on Children’s Health Insurance
Coverage
SCHIP has significantly increased the number of children from low-income families
who have health insurance, but enrollment in the SCHIP program is greater than the
corresponding decrease in the number of uninsured low-income children. SCHIP
provides a source of coverage that is less expensive to enrollees and often provides a
broader range of benefits than private coverage; as a result, some people who other-
wise would have obtained private health insurance coverage have instead enrolled in
SCHIP. Estimates of the extent to which private coverage has declined in response to
the program vary, but the available evidence strongly suggests the net effect of the pro-
gram has been to reduce the number of uninsured children.
6
CBO
Changes in the Number of Uninsured Children
Information on changes in the number of children who are uninsured comes from
self-reported data collected in household surveys. The estimates presented here are
based on data from the Annual Social and Economic Supplements to the Current
Population Survey, conducted by the Census Bureau, which is the most widely cited
source of information on insurance coverage. Although the survey is intended to mea-
sure the number of people who were uninsured throughout the calendar year, it yields
estimates that are similar to other surveys’ estimates of the number of people who
were uninsured at a particular point in time.7
SCHIP should be expected to have had the greatest effect on the extent of insurance
coverage among children in families with income between 100 percent and 200 per-
cent of the poverty level because that was the group that had the greatest increase in
eligibility for public coverage.8 According to CBO’s analysis, the percentage of chil-
dren in that income range who were uninsured fell from 23 percent in 1996 (the year
before SCHIP was created) to 17 percent in 2006, a reduction of about 25 percent
(see Figure 1). The rate of uninsurance was relatively stable among children in families
with income over 200 percent of the poverty level. For example, among children
whose families had income between 200 percent and 300 percent of the poverty level,
the rate of uninsurance remained at about 10 percent from 1996 to 2006.9
Among children in families below the poverty level, the rate of uninsurance rose from
24 percent in 1996 to 27 percent in 1998 and then fell to 22 percent in 2006. The
increase from 1996 to 1998 in the percentage of such children who were uninsured
7. For a discussion of the strengths and limitations of the Current Population Survey and other house-
hold surveys for measuring insurance coverage, see Congressional Budget Office, How Many People
Lack Health Insurance and For How Long? (May 2003).
8. One recent study found that the rate of eligibility of children in families with income between 100
percent and 200 percent of the poverty level increased 70 percentage points from 1996 to 2002—
compared with an increase of about 30 percentage points among children in families with income
between 200 percent and 300 percent of the poverty level, an increase of 10 percentage points
among those below the poverty level, and an increase of 8 percentage points among those between
300 percent and 400 percent of the poverty level. See Jonathan Gruber and Kosali Simon, Crowd-
Out Ten Years Later: Have Recent Public Insurance Expansions Crowded Out Private Health Insurance?
Working Paper No. W12858 (Cambridge, Mass.: National Bureau of Economic Research, January
2007).
9. In its analysis, CBO accounted for the fact that a “confirmation” question was added to the Current
Population Survey beginning with the interviews that collected data for 1999. The new question
asked people who did not report having any of several types of insurance coverage whether, in fact,
they were uninsured. CBO compared estimates of uninsurance rates with and without the data
from the confirmation question and used those two sets of estimates to create an adjustment factor
(separately for each income group) that it applied to the estimates for years prior to 1999 to make
them comparable with estimates for later years.
7
CBO
Figure 1.
Percentage of Children Who Were Uninsured, by
Family Income as a Percentage of the Federal Poverty Level,
1996 to 2006
(Percent)
30
25
Less than FPL
20 100% to 199% of FPL
15
200% to 299% of FPL
10
300% of FPL or More
5
0
1996 1998 2000 2002 2004 2006
Source: Congressional Budget Office based on data from the Current Population Survey for 1996 to
2006.
Note: FPL = federal poverty level.
was accompanied by a drop in Medicaid coverage, which some analysts have cited as
an unintended consequence of the welfare reform law that was enacted in 1996.10
The decline in the percentage of such children who were uninsured after 1998 was
accompanied by an increase in Medicaid coverage. In general, SCHIP did not make
more children in families below the poverty level eligible for public coverage because
most were already eligible for Medicaid. However, the percentage of children eligible
for Medicaid who participated in that program increased, which some analysts have
attributed partly to states’ outreach efforts for SCHIP (because applicants for SCHIP
were enrolled in Medicaid if they were found to be eligible for that program) and
the simplified application procedures that states adopted for both SCHIP and
Medicaid.11
10. See, for example, Karl Kronebusch, “Medicaid for Children: Federal Mandates, Welfare Reform,
and Policy Backsliding,” Health Affairs, vol. 20, no. 1 (January/February 2001), pp. 97–111.
11. See Thomas M. Selden, Julie L. Hudson, and Jessica S. Banthin, “Tracking Changes in Eligibility
and Coverage Among Children, 1996–2002,” Health Affairs, vol. 23, no. 5 (September/October
2004), pp. 39–50.
8
CBO
Those changes in the percentage of children who were uninsured do not yield an esti-
mate of the impact of SCHIP because there are many other factors—such as changes
in employment levels, family income, and health insurance premiums—that affect
children’s health insurance coverage. Nevertheless, the fact that the greatest reduction
in the percentage of children who were uninsured occurred among those who had the
greatest increase in eligibility for public coverage after SCHIP was established strongly
suggests that the program has reduced the number of children in low-income families
who are uninsured. As discussed below, however, estimating the effect of SCHIP on
children’s health insurance coverage requires a more sophisticated analysis that con-
trols for other factors that influence coverage and accounts for the program’s effects on
the number of people with private insurance.
Children’s Participation in SCHIP
The number of children who participate in SCHIP depends in part on low-income
parents’ awareness and understanding of the program, their attitudes toward public
insurance programs and health insurance in general, and the ease of the application
process. Nearly all states have promoted SCHIP through mass media campaigns, and
most have used community-based efforts such as educational sessions and home vis-
its.12 States have also implemented simpler enrollment procedures for SCHIP than
those used for Medicaid (although some have also adopted simpler enrollment proce-
dures for Medicaid). For example, most states do not require a face-to-face interview
for a parent to apply for SCHIP or to renew coverage but instead use simple mail-in
application forms, and most do not impose an asset test (that is, basing eligibility on
the amount of assets a family owns). Most states have a 12-month renewal period,
which enables children to remain enrolled in SCHIP for a year unless their family
reports a change in income or other circumstances.13 Since 2001, though, some states
have reduced their outreach efforts and retracted certain simplified enrollment proce-
dures in response to fiscal pressures.14
According to one study, 29 percent of the children who were eligible for SCHIP in
2005 on the basis of their family’s income participated in the program.15 Half of the
eligible children were covered by employment-based health insurance, 6 percent had
other coverage, and 15 percent were uninsured. According to that study’s estimates,
the uninsured children who were eligible for SCHIP accounted for over a fifth of all
12. Margo Rosenbach and others, Implementation of the State Children’s Health Insurance Program: Syn-
thesis of State Evaluations (report submitted by Mathematica Policy Research, Inc., to the Centers
for Medicare and Medicaid Services, March 2003).
13. Kaiser Commission on Medicaid and the Uninsured, Resuming the Path to Health Coverage for Chil-
dren and Parents: A 50-State Update on Eligibility Rules, Enrollment and Renewal Procedures, and
Cost-Sharing Practices in Medicaid and SCHIP in 2006 (January 9, 2007), available at www.kff.org/
medicaid/7608a.cfm.
14. Ibid.
15. Genevieve Kenney and Allison Cook, Coverage Patterns Among SCHIP-Eligible Children and Their
Parents, Health Policy Online, no. 15 (Washington, D.C.: Urban Institute, February 2007).
9
CBO
uninsured children in 2005. Other studies have estimated that between 60 percent
and 75 percent of all uninsured children are eligible for either Medicaid or SCHIP.16
Although all of those studies were based on rigorous statistical methods, they have
important limitations because they relied on data collected in household surveys to
determine children’s health insurance coverage and to identify children who were eli-
gible for SCHIP or Medicaid. Coverage in public programs such as Medicaid is
underreported in such surveys, but the implications of that underreporting for the
estimated number of people who are uninsured is unclear. There is some evidence that
many people who are enrolled in Medicaid but who do not report having coverage
under the program may report having private coverage instead.17 There is also evi-
dence that some SCHIP enrollees report having private nongroup insurance, which is
not surprising in that many states design their programs to resemble private insur-
ance.18 Additional research is needed to fully understand the implications of the
underreporting.
Another potential problem is that survey data on such things as types of income and
expenses that may be disregarded for determining eligibility are also subject to misre-
porting. In addition, some major surveys (such as the Current Population Survey) col-
lect data on annual income but no information on fluctuations during the year, which
would be relevant for determining eligibility for SCHIP.
The Effect of SCHIP on Private Coverage
Determining the extent to which enrollment in SCHIP is offset by reductions in pri-
vate coverage is important for evaluating the overall effects of the program and for
assessing the extent to which government spending on the program has reduced the
number of children who are uninsured. The crowding out of private coverage can
occur through various mechanisms. For example, some parents who would have oth-
erwise had family coverage through their employer might decline it for their chil-
dren—or might decline coverage altogether—if their children are eligible for SCHIP.
In addition, previously unemployed parents might be more likely to decline coverage
at a new job if their children are enrolled in SCHIP. To the extent that SCHIP makes
private coverage less important for some families, the program might also increase the
likelihood that low-income parents take jobs that offer higher cash wages rather than
16. See Selden, Hudson, and Banthin, “Tracking Changes in Eligibility and Coverage Among Chil-
dren”; and Lisa Dubay, John Holahan, and Allison Cook, “The Uninsured and the Affordability of
Health Insurance Coverage,” Health Affairs, Web Exclusive, November 30, 2006.
17. See Kathleen Thiede Call and others, “Uncovering the Missing Medicaid Cases and Assessing Their
Bias for Estimates of the Uninsured,” Inquiry, vol. 38, no. 4 (Winter 2001/2002), pp. 396–408.
18. See Joel C. Cantor and others, “The Adequacy of Household Survey Data for Evaluating the Non-
group Health Insurance Market,” Health Services Research, vol. 42, no. 4 (August 2007), pp. 1739–
1757; and Anthony T. Lo Sasso and Thomas C. Buchmueller, “The Effect of the State Children’s
Health Insurance Program on Health Insurance Coverage,” Journal of Health Economics, vol. 23,
no. 5 (2004), pp. 1059–1082.
10
CBO
health insurance. Thus, even in the majority of states where SCHIP covers only chil-
dren, the program could reduce private coverage among adults as well as children.
SCHIP can also reduce private coverage by influencing the actions of employers. If
employers of low-wage workers believe that SCHIP makes health insurance less
important in attracting high-quality employees, some might reduce their contribution
to the premiums for family coverage, reduce the level of benefits offered, stop offering
family coverage, or stop offering insurance altogether. Such actions could lead to less
private coverage among families that are eligible for SCHIP as well as for those that
are not.
Families that substitute SCHIP for private coverage are generally better off because
the cost (to the enrollees) is lower and the package of benefits may be more extensive.
However, to the extent that employers respond to SCHIP by increasing premiums,
reducing benefits, or declining to offer coverage, other families could be worse off.
Little is known about how employers have responded to SCHIP. As discussed below,
the limited evidence that is available suggests that SCHIP has not affected employers’
decisions on whether to offer coverage but may have caused them to modestly raise
employees’ premiums for family coverage relative to the premiums for individual cov-
erage. The implication is that most of the reduction in private coverage associated
with the existence of SCHIP appears to result from parents choosing to forgo private
insurance for their children and instead enroll them in SCHIP, presumably because
the parents believe the program offers better benefits or lower costs than private insur-
ance.
The existence of SCHIP may also affect private coverage by increasing enrollment in
the Medicaid program—a consequence of the outreach that states have conducted for
SCHIP and the simplified application procedures that many have adopted (in some
cases, for Medicaid as well as for SCHIP). That increased enrollment in Medicaid has
probably been offset to some extent by a reduction in private coverage, for the same
reasons that enrollment in SCHIP has probably been partly offset by a reduction in
private coverage. The reduction in private coverage associated with the increase in
Medicaid coverage is probably smaller than that associated with enrollment in SCHIP,
however, because people eligible for Medicaid have lower income and less access to
private insurance than people eligible for SCHIP do.
Efforts to Limit the Substitution of SCHIP for Employment-Based Health Insurance.
Federal law requires that the states have procedures in place to prevent people from
substituting SCHIP for employment-based insurance. The Congress included that
provision in the authorizing legislation because of concern about substitution, in part
resulting from a study that estimated that an expansion of Medicaid in the late 1980s
11
CBO
and early 1990s caused a decline in private coverage that was about half the size of the
increase in Medicaid coverage.19 Subsequent studies obtained much lower estimates
for the effects of Medicaid on private coverage.20
The potential for SCHIP to displace employment-based insurance is greater than it
was for the expansion of Medicaid because the children eligible for SCHIP are from
families with higher income and greater access to private coverage. According to one
study, 60 percent of the children who became eligible for SCHIP had private coverage
in the year before the program was established.21
States have included a variety of features in their programs to try to prevent SCHIP
from displacing employment-based insurance. A widely used approach is to impose a
waiting period—that is, a specified length of time that children must be uninsured
before becoming eligible for SCHIP. In 2006, 35 states had a waiting period, the two
most common being six months (imposed by 16 states) and three months (imposed
by 11).22 Only one state had a waiting period that was longer than six months. Many
states allow exceptions to the waiting period—when a parent loses private coverage for
reasons considered involuntary (by losing his or her job, switching to a job that does
not offer family coverage, or becoming disabled, for instance) or when the available
insurance is considered too expensive (if the employee’s premiums exceed a specified
percentage of income, for example, or if the employer contributes less than 50 percent
to the cost of coverage).23 Most states collect insurance information on the applica-
tion for SCHIP, and some verify that information with employers. Some states try to
limit the displacement of employment-based insurance by requiring premiums and
copayments within SCHIP.
Estimates of the Effects of SCHIP on Private Coverage. Estimates vary about the
extent to which SCHIP has resulted in less private coverage. The available studies,
which have focused on the effects of SCHIP on children, use various data sources and
methods. On the basis of a review of the available studies, CBO concludes that the
reduction in private coverage among children is most probably between a quarter and
19. That estimate included changes in coverage among children, women of childbearing age, and other
adults (who were not eligible for Medicaid). Among children, the study found, the reduction in pri-
vate coverage was equal to 40 percent of the increase in public coverage. See David M. Cutler and
Jonathan Gruber, “Does Public Insurance Crowd Out Private Insurance?” The Quarterly Journal of
Economics, vol. 111, no. 2 (May 1996), pp. 391–430.
20. See, for example, Linda J. Blumberg, Lisa Dubay, and Stephen A. Norton, “Did the Medicaid
Expansions for Children Displace Private Insurance? An Analysis Using the SIPP,” Journal of Health
Economics, vol. 19, no. 1 (2000), pp. 33–60.
21. See Julie L. Hudson, Thomas M. Selden, and Jessica S. Banthin, “The Impact of SCHIP on Insur-
ance Coverage of Children,” Inquiry, vol. 42, no. 3 (Fall 2005), pp. 232–254.
22. Kaiser Commission on Medicaid and the Uninsured, Resuming the Path to Health Coverage for
Children and Parents.
23. Rosenbach and others, Implementation of the State Children’s Health Insurance Program.
12
CBO
a half of the increase in public coverage resulting from SCHIP.24 That is, for every
100 children who gain coverage as a result of SCHIP, there is a corresponding reduc-
tion in private coverage of between 25 and 50 children.25
Measuring the extent to which SCHIP is associated with a decline in private coverage
is difficult because it requires comparing the insurance coverage of people under cur-
rent law with an estimate of the coverage they would have had if the program did not
exist. Analysts have estimated the reduction in private coverage attributable to SCHIP
by using various statistical models to try to remove the effects of other factors that
affect private coverage. All studies that have been conducted to date have estimated
the reduction in private coverage among children only; they do not capture any possi-
ble reduction in private coverage among parents or other adults. Consequently, the
available estimates probably understate the total extent to which SCHIP has reduced
private coverage.
Some studies have estimated crowd-out by examining the insurance coverage of p