The Heritage Foundation

Backgrounder #2180 on Health Care

September 9, 2008

September 9, 2008 | Backgrounder on Health Care

SCHIP: How Congress Can Avoid Repeating Last Year's Mistakes

Key Points

  1. Last year’s SCHIP debate was about politics, not poor children. $1 trillion is already committed to children on Medicaid and SCHIP over the next 10 years. According to the NPR/ Kaiser/Harvard survey, most Americans do not support expansion to children in families making $60,000.
  2. Eligibility for enrollment in SCHIP was not capped at 300 percent ($61,950 for a family of four) of the federal poverty level. Under the 2007 legislation, states could continue to expand eligibility for government coverage through “income disregards” or even Medicaid.
  3. Expanding government coverage to higher income levels leads to loss of private coverage, which also drives up the cost of private health insurance for all Americans.
  4. Performance Bonus and Express Lane eligibility provisions favored wealthier states and states with large non-citizen populations.
  5. SCHIP should include premium assistance to help families with children get or keep their private health coverage.

Congress may soon return to reauthorization of the State Children's Health Insurance Program (SCHIP). Extension of the popular program should have been a simple, quick, and bipartisan victory last year, but pol­icy took a back seat to politics and raw emotion, leav­ing confusion even among proponents on both sides of the aisle. There was little time for serious discus­sions about important policies.

As a result, congressional leaders were unable to resolve smoldering controversies about the inclusion of higher-income children and adults, the "crowd out" of private health insurance because of public program expansions, or the expansion of eligibility to non-citi­zens through mechanisms buried in the bill, as well as the imposition of tax increases that were as unneces­sary as they were unpopular. Moreover, congressional consideration of the SCHIP legislation was marred by the leadership's insistence on relying on accounting gimmicks that would enable the bills to meet pre­scribed budgetary rules.

In putting politics before policies, congressional leaders undercut any serious consideration of the underlying issues and obscured key policy questions for Members and taxpayers alike.

Last year's debate was, of course, never about cov­ering poor children. There was never any disagree­ment about covering the children who meet the statutory definition of a "targeted low-income child." Bills sponsored by Republicans in both the House and the Senate would have provided sufficient funding to cover the "currently eligible" SCHIP children at a much lower cost to the taxpayer.

There was nonetheless a great deal of confusion and disagreement even among Members who voted for the SCHIP legislation last year. It was a missed opportunity to engage honestly in a critical national discussion about who should be eligible for public assistance and whether there should be such a redistribution of wealth among the various states. The fiscal integrity and program integrity of Medic­aid and SCHIP were put at risk by Express Lane eli­gibility, expansion of presumptive eligibility, and at least eight special-interest earmarks.

Of the total population of approximately 80 mil­lion children in the United States, about 45 percent receive taxpayer-funded health benefits. For Con­gress to expand dependence on public assistance even more, aggravating the current problems with entitlement spending, would have profound impli­cations for the future of our health insurance system and a significant impact on America's social, eco­nomic, and political systems. One cannot escape these effects if more than half of America's children start out relying on government assistance.

Preparing for Next Time. Before legislation is sent back to the floor, the House and Senate should hold balanced hearings-with outside policy ex­perts who would thoroughly examine the alterna­tives as well as the pitfalls of last year's versions of the legislation. Only then should they begin to mark up legislation through regular order.

Last year, the SCHIP legislation went through dramatic changes between committee and floor action, leaving little time for Members of either party to digest differing explanations of policy. Locked into political gamesmanship rather than a commitment to bipartisan problem-solving, the debate lacked serious discussion about the real pur­pose, goals, and future of SCHIP and Medicaid, two government benefit programs that already carry a combined commitment of $1 trillion in spending on children over the next 10 years.

Setting a Clear Health Care Policy Is Crucial

During last year's debate, Congress picked bud­get numbers first, then clumsily backed into certain policies with some strange results. With the number of children enrolled in Medicaid and SCHIP already exceeding the number of children below 200 per­cent ($41,300 for a family of four in 2007) of the federal poverty level (FPL), Congress could not spend its budget target of $35 billion without expanding SCHIP eligibility into the middle class or federalizing even more of the cost of Medicaid. For several states, Medicaid is already the vehicle of choice for the coverage of children eligible for SCHIP, and policies in last year's legislation were further tilted toward Medicaid expansion.

Congress should refocus the program on the population for which it was originally designed- children in low-income working families above Medicaid eligibility-but not on children from higher-income families. This means setting a firm cap on eligibility based on the definition of a tar­geted low-income child that applies to both SCHIP and Medicaid; closing the income eligibility loop­holes that are the source of confusion, tensions, and jealousies among the states; and ending the state practice of expanding eligibility to any income level through a "backdoor" approach, including through Medicaid.

A glance back at last year's debate shows the need for clarity about how states were using SCHIP and about their policy goals. If one examines the growth in SCHIP and Medicaid enrollment-and the Con­gressional Budget Office (CBO) score tracking addi­tional expansions-then one must conclude that last year's legislation was only marginally about the children for whom SCHIP was created.

Eligibility for SCHIP. During last year's debate, nothing was more muddled than the question of who was supposed to benefit from the program.

During the original SCHIP debate in 1997, the goal of insuring 10 million children was widely shared. In federal fiscal year (FY) 1998, the number of "ever enrolled" children in Medicaid was 19.6 million, and the number "ever enrolled" in SCHIP was 660,000, for a combined total of 20.3 million children covered by Medicaid and SCHIP. In FY 2007, the number of children "ever enrolled" in Medicaid had increased to 29.3 million, and the number enrolled in SCHIP had increased to 7.1 mil­lion, for a combined total of 36.5 million children- an increase of 16 million children, or nearly 80 percent, from the 1998 lev­els.[1]  (See Chart 1.)

Chart 1

Yet, since 2007, as has been widely reported and as Representative Frank Pallone (D-NJ), chairman of the House Energy and Commerce Committee's Subcommittee on Health, declared, "Nine million children in this country have no health insurance." "In a country as wealthy and compassionate as ours," he continued, "no child should be left behind without health insurance, let alone 9 million."[2] 

In other words, Congress started with the goal of covering 10 million children, added 16 million, and was somehow left with 9 million children still with­out health insurance 10 years later.

Has the number of poor children increased in the United States? No. The number of children in fam­ilies with incomes below 200 percent of FPL actu­ally declined slightly from 30.6 million in 1997 to 30.2 million in 2006.[3]  The growth in the total num­ber of children occurred among the higher-income families, not the lower-income families, after the enactment of welfare reform in 1996 and the imple­mentation of pro-growth economic policies. The combined total of children on Medicaid and SCHIP (36 million) now exceeds the number of children living below 200 percent of FPL (30.2 million).

The question is this: How could 16 million chil­dren be added to the rolls and 9 million remain uninsured? It seems implausible that increasing enrollment and spending $40 billion on SCHIP and even more on Medicaid would have resulted in a net reduction of only 1 million in the number of unin­sured children. There could have been various rea­sons, including "crowd-out" of private coverage from public program expansion, a continuing decline in private coverage, counting Medicaid chil­dren who were already enrolled as "uninsured," allowing higher-income children to be enrolled through special "income disregards," counting chil­dren from higher-income families, changing the length of time used in measuring the number of uninsured, deficient data, and counting non-citi­zens among the uninsured in the same way that cit­izens are counted.

The most plausible explanation is, of course, that there were not 9 million uninsured low-income chil­dren. Representative Pallone may have recognized this. Later in the debate, he stated that, "If we are going to find the approximately 6 million children who are eligible for SCHIP or Medicaid, but who are not enrolled, we would need at least $50 billion over the next five years."[4]  To add 9 million children to SCHIP, Congress would then have to expand SCHIP coverage to higher income levels and some­how make non-citizens eligible.As shown below, proponents of expanding SCHIP conceded that they would not cover even 6 million uninsured chil­dren eligible for SCHIP and Medicaid.

Nearly 40 percent of the uninsured children live in families with incomes above 200 percent of FPL, including 27 percent living in families with income above 250 percent of FPL.[5]  Also included in the 9 million figure are 1.5 million children who are inel­igible because they are not American citizens.[6] 

By September 2007, the goal of enrolling 9 mil­lion had quietly been cut by more than half, and congressional leaders spoke in favor of adding cov­erage for just 4 million children by 2012. Represen­tative John Dingell (D-MI), chairman of the House Committee on Energy and Commerce, explained that the SCHIP reauthorization legislation (H.R. 976) then under consideration "gives $35 billion to strengthen and improve children's health coverage. It protects 6 million children today covered by SCHIP. It adds an additional 4 million."[7]  House Majority Leader Steny Hoyer (D-MD) echoed this understanding: "Let's be clear; this fiscally responsi­ble legislation will ensure that some 10 million chil­dren will receive health insurance coverage. That's approximately 4 million more than are covered under the [State] Children's Health Insurance Pro­gram today."[8] 

Deciphering CBO Numbers. During last year's debate, much of the CBO score remained a mystery throughout the legislative process. The CBO pre­sented enrollment figures only for FY 2012 and combined spending from different sections of the legislation in such a way that the effects of individ­ual parts could not be clearly understood. This pre­vented a comparison of viable alternatives. The next time SCHIP is considered, congressional lead­ers should make sure that the CBO provides year-by-year enrollment figures, explains "take up" rates, identifies states that would expand eligibility, and disaggregates funding provisions so that Mem­bers understand how the legislation is channeling the spending.

Consider the question of who is counted among the 4 million children. The CBO score was critical to understanding that last's year's legislation, H.R. 3963, was only marginally about uninsured chil­dren currently eligible for SCHIP. The CBO pro­jected that by 2012, the legislation would reduce the number of uninsured individuals (some adults would still be on SCHIP) by 3.9 million.[9]  (See Table 1.) But only 800,000 were the result of additional enrollment within existing SCHIP eligibility groups.

The CBO further estimated that H.R. 3963 would also be used to enroll 2 million individuals who would forfeit private coverage for enrollment into Medicaid or SCHIP. (See Table 1.)

Table 1

Only about one-quarter of the total enrollment increase projected by the CBO (1.5 million of 5.8 million people[10] ) was attributed to uninsured, low-income children currently eligible for SCHIP. This included 700,000 already enrolled children that the CBO projected would lose coverage under the cur­rent baseline. So only 14 percent (800,000 newly enrolled, previously uninsured of 5.8 million total) would have gained coverage. The balance of enroll­ment was due to Medicaid, "crowd out" of private insurance, and SCHIP expansion into higher income levels of the population.

Members of Congress may not have been aware that, in terms of increasing coverage for people cur­rently eligible for SCHIP, they could have achieved the same results for a fraction of the cost. The CBO estimated that a Senate alternative, the Kids First Act, sponsored by Senator Trent Lott (R-MS), would have increased the enrollment of currently eligible but uninsured children by 700,000 at a net cost of $8 billion, compared to 800,000 at a stun­ning cost of $32.1 billion under H.R. 3963,[11]  the last version of the SCHIP legislation.

Table 2 compares only the costs that were appli­cable to both bills. H.R. 3963 contained another $3.3 billion in spending that Kids First did not include and that therefore has been omitted. Kids First would have saved $2.5 billion in Medicaid costs because the CBO assumes in its budget base­line, under current law, that states experiencing "shortfalls" in SCHIP funding would move popula­tions into Medicaid. Adding funds to SCHIP would prevent that from happening and would thereby reduce Medicaid spending relative to the baseline, thus "saving" money. Under H.R. 3963, Medicaid spending would have increased by $5.7 billion. Because of the SCHIP-Medicaid baseline interac­tion just described, the impact on Medicaid spend­ing, then, presumably would really be $8.2 billion ($8.2 billion - $2.5 billion = $5.7 billion). In addi­tion, the $2.5 billion in Performance Bonus pay­ments would come from SCHIP funds, but they would be used for increases in Medicaid enroll­ment. If it were good policy, which it is not, Con­gress could have considered other policies for increasing Medicaid enrollment that would have cost less.

It is also interesting that Kids First would have provided more funding for outreach ($1 billion) than would have been provided under H.R. 3963 ($300 million). On the surface, this suggests that part of the increase in spending in H.R. 3963 (both for SCHIP and for Medicaid) was not attributed to findingnew eligible but not enrolled children; rather, the federal government would spend more money keeping the same children enrolled. But it is unclear whether this is the case, because the CBO tables combine the effects of different provisions of H.R. 3963.

Congress should also ask what percentage of low-income children will still be uninsured if these new enrollment levels in SCHIP and Medicaid are achieved. Given the criticism of the Bush Admin­istration's August 17, 2007, directive that states should demonstrate that they have achieved a 95 percent coverage rate for low-income children before approving state requests to expand to higher income levels,[12]  Congress and the public should know whether the substantial increase in spending would achieve this policy goal.

Finally, Congress should ask the CBO to fully explain its Medicaid enrollment estimates. CBO estimated that in 2012, the average monthly net Medicaid enrollment would increase by 2.3 million individuals within current eligibility groups. As Chart 1 illustrates, Medicaid enrollment has slowed in recent years because of the success and maturity of SCHIP. Between 2004 and 2007, the average annual increase in the number of children ever enrolled in Medicaid for the three-year period was less than 300,000. There is a finite number of chil­dren who are currently eligible for Medicaid. Old models that assume that for every child enrolled in SCHIP, one or more will also be enrolled in Medic­aid cannot be sustained indefinitely. Given that the number of "ever enrolled" is a substantially higher figure than "average monthly enrollment," the CBO appears to credit H.R. 3963 with an aggressive enrollment rate. Congress should seek a clear understanding of which new policies would achieve such results and whether the same results could be achieved at a lower cost to the taxpayers.

table 2

Confusion over Eligibility. Eligibility policy was just as murky. During consideration of the SCHIP legislation, Congress was aware of a request from the state government of New York to expand eligibility to 400 percent of FPL ($82,600 for a fam­ily of four in 2007). Congressional leaders who favored SCHIP expansion inserted language into the first SCHIP bill, H.R. 976, to protect the ability of New York and New Jersey (350 percent of FPL; $72,275 for a family of four in 2007) to claim the SCHIP-enhanced match rate for their citizens while limiting other states to claiming the enhanced match for up to 300 percent of FPL ($61,950 for a family of four in 2007). With the addition of this language, the bill was passed by both the House and the Senate.

H.R. 976 was subsequently vetoed by President George W. Bush, and the veto was sustained on October 18, 2007. A new SCHIP bill, H.R. 3963, was introduced in the House on October 25. House and Senate SCHIP supporters had broadened dis­cussions to a small group of House Republican members in hopes of picking up additional votes that would make SCHIP veto-proof. A cap on income eligibility was one of the key discussion points that congressional leaders believed might win sufficient support from Republicans.

This time, the House and Senate negotiators dropped the explicit income-eligibility exception for New York State, so Representative Pallone assured Members that "if you go over 300 percent, okay, other than those that are already grandfa­thered into the program, you're no longer going to be able to cover those kids at that $82,000 or the other levels they suggested."[13]  Representative Gene Green (D-TX) echoed his comments: "The bill is clear on family income. Only the lowest-income children are covered with a prohibition on coverage of children above 300 percent. You can't go above 300 percent."[14] 

However, H.R. 3963 did not cap eligibility at 300 percent of FPL; it would merely have pro­vided the lower Medicaid match rate if a state expanded eligibility above 300 percent. As the legislation was written, states could also adopt "disregards" so that families with gross incomes above 300 percent of FPL could still qualify for the enhanced match rate. These disregards, as explained further below, are used to subtract income or expenses from a family's gross income in order to let them qualify for assistance.[15] 

Expanding Dependence: An Example from the 2007 SCHIP Debate

A colloquy between Representatives John Dingell (D-MI) and Michael Burgess (R-TX) shows how, through "income disregards," the states could have continued to expand eligibility even above 300 percent of FPL:[16]

Mr. Burgess: "Mr. Chairman, under the changes that have been made in regards to the income disregards in the bill, could a state in its current practice still allow a family to exclude from income $500 a year for child care expenses?"

Mr. Dingell: "The answer to the question is yes."

Mr. Burgess: "I thank the Chairman. Could a state allow a family to exclude from income $20,000 a year for housing expenses?"

Mr. Dingell: "That would be a matter to be determined by the state in which the transaction and the events occurred."

Mr. Burgess: "I am not a lawyer, but if I were a lawyer and ask for a 'yes' or 'no' answer I would assume that's a 'yes.'"

Mr. Dingell: "Well it's a 'yes' if the state so decides. It's a 'no' if they [sic] decide not."

Mr. Burgess: "Further then, if the Chair will indulge me, could a state allow for a family to exclude from income $10,000 for transportation expenses?"

Mr. Dingell: "Again, the response is that that is up to the state, and there is nothing in the legislation to preclude that."

Mr. Burgess: "So the answer would be a 'yes' if to transportation expenses. If the Chairman would, then, could a state allow a family to exclude from income $10,000 a year for clothing expenses?"

Mr. Dingell: "Again, the answer is if that is so determined by the states, the answer is yes."

Mr. Burgess: "So the state income disregards, now, are up to $40,500, if I am doing my math correctly?"

Welfare Reform in Reverse. The House collo­quy cited here shows that the income-eligibility "cap" on SCHIP eligibility was a fiction. Through the magic of so-called income disregards, a family of four making $102,450 could subtract $40,500 and appear on paper to meet the eligibility cutoff of 300 percent of FPL. Historically, income- and work-related disregards have been used in welfare pro­grams as incentives for people to return to work and escape dependence on government. In the SCHIP debate, however, the world was turned upside down to allow families at higher income levels to appear to be low-income and thereby become eligible for pub­lic assistance.

Even worse, the legislation did not apply the lower match rate to an expansion through Medicaid. Section 115 of H.R. 3963 provided open-ended funding to a state at the enhanced SCHIP match rate at any income level. An eligibility cap should be applied equally if a state chooses to run its SCHIP program as a Medicaid expansion. Allowing states to circumvent eligibility caps by expanding Medicaid would render the policy meaningless.

Clarity in determining who should be eligible for SCHIP is critical in terms of public support as well. Most Americans do not support expanding SCHIP to the levels promoted by H.R. 3963. According to an October 2007 National Public Radio/Kaiser Family Foundation/Harvard School of Public Health Survey, "66 percent of Americans say they support having families of four who make $40,000 per year (roughly two times the poverty level) be eligible but support drops substantially-to 32 percent-at $60,000."[17]  Given such sentiment, important SCHIP policy should be clear in order to assure that Members of Congress and the public alike are fully informed.

Congress Should Not Favor Wealthy States

Offering the enhanced SCHIP match rate for families higher up on the income scale is likely to be attractive only to a minority of states, of which all but two are outside of the South and Plains. Of the 13 states that have expanded to 300 percent or that have pending requests to do so, only two (Missouri and Oklahoma) are from the South or Plains states. Missouri offsets the public cost of expansion to this higher income level by requiring the maximum cost-sharing allowable under federal law: 5 percent of the family's income ($3,097.50 for a family of four with income of $61,950). Tennessee had already learned hard lessons about overextending public benefits to the middle class and has lowered eligibility for children post-SCHIP.

All three SCHIP versions passed by the House (H.R. 3162, H.R. 976, and H.R. 3963) contained provisions that encouraged states to overspend their allotments and adopt more liberal policies related to eligibility, cost-sharing, and the "crowd out" effect. These provisions also were likely to benefit only a minority of states because most would not expand coverage to higher income levels.

Under the original SCHIP legislation, the federal government provides states an "enhanced" match rate for a "targeted low-income child." (Nationally, the federal government provides 70 percent of SCHIP funding, compared to 57 percent in Medic­aid.) This definition would have allowed states to expand eligibility to 200 percent of FPL, or 50 per­centage points above a state's Medicaid threshold as of March 31, 1997.[18] 

Sixteen states have exceeded the definition of a "targeted low-income child." These expansions could occur beyond the statutory definition, as explained by the Congressional Research Service: "[B]y using existing flexibility to define what 'counts' as income, any state can raise its effective SCHIP eligibility level above 200 percent of FPL through the use of income disregards."[19] 

In other words, SCHIP can be expanded further by redefining who is "poor." By doing so, SCHIP would coax middle-class families into dependence on public assistance.

Under the two bills passed by Congress (H.R. 976 and H.R. 3963) but vetoed by President Bush last year, the enhanced federal match rate would have been provided to states that expand SCHIP eli­gibility to 300 percent of FPL, and the regular Med­icaid match rate would have been provided to states that expand above 300 FPL.[20]  There was, however, little discussion about which states would benefit from an expansion of SCHIP.

The expansion of SCHIP (as well as Medicaid) tends to occur in wealthier states, so reauthorization that allowed the enhanced match rate for higher-income families favored those states, resulting in families in lower-income states subsidizing higher-income families in wealthier states.[21]  For purposes of state comparisons, the relative wealth of states is based on personal income per capita. Based on 2006 data, Connecticut ranks first in "personal income per capita" at $50,762, while Mississippi ranks last at $27,028-a difference of $23,734.[22]  (See Chart 2.)

Chart 2

Table 3 shows the 16 states that exceed their eli­gibility thresholds as of April 2008. Ten states have expanded eligibility to 300 percent, and four have pending proposals to do so (California is in both categories, so there is a total of only 13 states).[23] 

Table 3

Of the 10 "richest" states, seven (California, Con­necticut, Maryland, Massachusetts, New Hamp­shire, New Jersey, and New York) have already expanded or are seeking to expand SCHIP eligibility to 300 percent of FPL. Of the 13 states that have already expanded to 300 percent of FPL or are seek­ing to do so (California, Connecticut, Hawaii, Mary­land, Massachusetts, Missouri, New Hampshire, New Jersey, New York, Ohio, Oklahoma, Pennsyl­vania, and Vermont), nine rank among the top 20 "richest" states. None of the "poorest" states have expanded to more than 250 percent of FPL.

Expansion of eligibility in "richer" states is some­times justified because of their higher cost of living. Wealthier states may argue that health insurance, like other goods and services, is more expensive for their citizens. Connecticut, Massachusetts, and New Jersey, for example, can accurately point out that their residents face some of the highest health insurance premiums in the country.

But "affordability" in the cost of health insurance should be viewed as a combination of income and the price of health insurance itself. When expressed as a percentage of income, the amount that Connect­icut, Massachusetts, and New Jersey residents actu­ally pay for their health insurance coverage is less than families in poorer states pay. This is illustrated in Table 4 which compares the average median fam­ily income for a family of four to the average total family premium for private-sector employer-spon­sored health insurance in the 10 highest-income states and the 10 lowest-income states.

In other words, the question is: If a family of four earning a state's median family income paid the average health insurance premium in that state, what is the cost as a percentage of the family's income?[24] 

For the purposes of illustration, this paper recog­nizes that the employee ultimately pays the full cost of insurance as he forfeits the full wages necessary to purchase it. According to the federal Agency on Healthcare Research and Quality (AHRQ), the aver­age total (employer and employee shares) family premium per enrolled private-sector employee was $11,381 in 2006.[25] 

In comparing the states by median family income of a family of four in 2006, New Jersey was the high­est at $94,441 and New Mexico the lowest at $52,034, a difference of $42,407.[26]  Paying the aver­age total family premium cost a family in New Jersey $12,233 compared to $11,279 for a family in New Mexico, a difference of $945. However, as a percent­age of income, the family in New Mexico would have spent 21.7 percent of its income on coverage, while the family in New Jersey would have spent 13 per­cent of its income on coverage. It is therefore no great mystery as to why the rate of non-insurance is greater in New Mexico than in New Jersey.

Table 4

The average cost of health insurance among the 10 states with the highest median family income for a family of four is $11,775, while the average cost in the 10 states with the lowest median family income is $10,076, a difference of just $1,069. But the dif­ference in the median income for a family of four between the 10 "richest" states and the 10 "poorest" states in 2006 was $29,389.

If the cost of group health insurance is broadly comparable across the country[27]  while median family income varies as much as $40,000, why would taxpayers in "poorer" states be asked to subsidize higher-income populations in the six wealthiest states?

Let us look closer at two states, New York and Vermont, as examples for why Congress should examine what is really occurring with these cross subsidies. New York became a lightening rod during last year's SCHIP debate as it submitted a state plan amendment to expand eligibility to 400 percent of FPL ($82,600 for a family of four), but New York was already allowing families at any income level to "buy in" to its children's health program. The aver­age cost of health plans available through the state is $154 per member per month, or $1,848 per year. For two children, the annual cost would be $3,696. New York's share of SCHIP expenditures is 35 per­cent, which would mean paying out $1,293 per year for two children on behalf of one family. If the state contributed just this amount without any fed­eral funds being involved, the family's cost would be reduced to $2,403. Is it not reasonable to expect that a family making $82,600 can afford a premium of less than $7 per day?

In the case of Vermont, the state currently requires its families with income up to 300 percent of FPL ($61,950 for family of four) to pay a maxi­mum of $80 per month, regardless of family size, in premiums. This is 1.5 percent of family income. There are no other types of cost-sharing such as deductibles or copayments involved. Last year, the state decided to cut this modest premium in half. In effect, Vermont intends to give back to these fami­lies, who were already paying the current premium, a subsidy of $480 per year.

If health care coverage is already affordable for individuals and families, federal taxpayers should not be required to pay for additional subsidies to cover them. Such a congressional policy sends a demoralizing message to working families who use their own resources to purchase private insurance for their children. This suggests that the real-but thus far tacit-assumption in the recent debate about expanding SCHIP was not about "affordable" health insurance for "poor" children, but rather about increasing dependence on government and the redistribution of income through this venue of public assistance.

The New York and Vermont examples expose the grave risks associated with getting America's middle class hooked on public assistance. Health benefits become tools for political control of the financing and delivery of care. While some families (those to whom eligibility has been expanded) would benefit, other families would be worse off because the cost for them in a shrinking private market with fewer younger and healthier members of the insurance pool would likely increase, resulting in another round of employers and families dropping or for­feiting private coverage.

This is one of the consequences of the "crowd out" of private coverage that accompanies public program expansion. As the CBO explained in recent congressional testimony:

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.[28] 

There is general recognition that, since the cost of health insurance has increased considerably faster than wages have increased, both the strain on family budgets and the strain on the national economy demand solutions. At the same time, the presidential candidates tacitly assume that there is plenty of money in the health care sector of the economy because their various proposals include "savings from reducing inefficiencies in the system."[29] 

No one credibly argues that American health care suffers from a lack of money, even though there is general agreement that health insurance should be more "affordable" for lower-income persons. Beyond that, even if one assumes that families in New Jersey who are paying 13 percent of median family income for the average premium may be pay­ing "too much," it does not logically follow that the right answer is to enroll them in SCHIP or some other public program. In any case, SCHIP expan­sion does not make health care more affordable; it merely redistributes the costs and, in the process, cre­ates new inequities among working families.

The states and the federal government should not expand SCHIP to higher-income levels. Instead, they should convert the assistance provided to fam­ilies at lower income levels into subsidies for the private health insurance market. One routine objec­tion to health savings accounts (HSAs) is that they entice healthy members out of the health insurance pools. Yet that is precisely what Medicaid and SCHIP have accomplished. Returning tens of mil­lions of healthy individuals to the private health insurance market would reverse the negative effects of "crowd out" from public program expansion and help lower the cost of health insurance for every­one. Public policies should put more people into the insurance pools, not take them out.

As shown in Table 4, the costs of health insur­ance in the group markets across the country are generally comparable. Contrary to what some in Congress suggest, there is no need to create some type of federal government entity to regulate the costs of health insurance; expanding the insurance pool itself will help to lower the cost, especially if states, as regulators of health insurance, open mar­kets and encourage greater competition-including competition across state lines.

There are, and likely always will be, individuals and families who need help paying for health insur­ance, but fairness requires a more equitable approach than letting wealthier states bankroll fam­ilies at any income level with federal dollars. States should provide individuals with access to affordable private health insurance through properly rede­signed insurance markets that are not smothered by unnecessary regulation, while federal financial assistance should reduce their reliance on current means-tested programs like Medicaid and SCHIP.

While Congress should demand that states adopt policies that prevent "crowd out," it should also encourage the use of premium assistance within Medicaid and SCHIP. To address the real problems faced by middle-class families who are struggling with the cost of health insurance, Congress may wish to look beyond SCHIP and provide these fam­ilies with a health care tax credit. This would not only enable families who do not get health insur­ance through their workplace to buy it, but also allow everyone to keep it by expanding the insur­ance pool.

Congress may wish to consider a tax credit for families with income at least as low as 150 percent of FPL ($30,975 for a family of four in 2007) that allows them to buy their own health insurance. This would also provide relief to all states by diverting families from public assistance provided through Medicaid and SCHIP.

Play It Straight with SCHIP Funding

All three versions of last year's SCHIP legislationwere designed to spend a pre-set level of money. The result: The money was rewarding unsound pol­icymaking at the state level. Again, such policies tend to benefit a minority of wealthier states that need it the least.

In the next debate, Congress should jettison the kind of funding gimmicks embodied in last year's bills that caused so much confusion and that, if enacted, would have been additional vehicles for manipulation by state officials. Specifically, Con­gress should return financing to true allotments and drop the Contingency Fund, which removes Congress from financial decision-making, and the Performance Bonus, which provided SCHIP dollars for increased Medicaid caseloads. Congress should also eliminate Express Lane eligibility, which would give new authority to different local agencies to determine eligibility and thereby weaken program integrity; eliminate the expansion of presumptive eligibility, which increases the potential for abuse by providers; eliminate the special-interest ear­marks for specific states and providers; and refrain from creating a budget cliff in 2013. Funding should be straightforward, maintaining the capped allotments that reflect reasonable growth rates, and include an updated allotment formula.

Congress missed an opportunitylast year to show real support for premium assistance. While appearing to promote premium assistance, Con­gress actually took a big step backwards. The provi­sions on premium assistance would have provided states with less flexibility than they currently pos­sess and would have hobbled efforts to employ the premium-assistance strategy.

Prevent State Bailouts. The preservation of state allotments would preserve the vital policy link with the bipartisan origins of SCHIP. The central features of the bipartisan compromise that created SCHIP were the enhanced federal match rates and program flexibility balanced by capped allotments. The federal government provided the states with a higher match rate and greater discretion to set poli­cies than they had under Medicaid-with the assumption that states would use their flexibility to set sound policies because funding was subject to capped allotments.

For a time, the allotments, with redistribution of unexpended allotments to shortfall states, were generally successful. Then, rather than fixing the allotment formulas, Congress abandoned fiscal dis­cipline and began to bail out shortfall states in each of the past three years (in the Deficit Reduction Act of 2005, the National Institutes of Health Reform Act of 2006, and the Medicare, Medicaid, and SCHIP Extension Act of 2007) with little regard for why the shortfalls occurred.

Although there was sufficient funding in the aggregate, individual states have faced shortfalls. Some of these shortages were appropriately attrib­uted to flaws in the allocation formula, but a large part of the shortfall was caused by the states that had expanded their SCHIP program to include chil­dren at higher income levels and, in some cases, adults. In FY 2006, states faced a $454 million shortfall, which was covered by the federal govern­ment through redistribution of unexpended funds and supplemental appropriations.[30]  In 2007, states faced shortfalls of $813 million, which were also eventually covered by the federal government through redistribution of unexpended funds and supplemental appropriations.[31] 

Among the states that received nearly $1.3 bil­lion in additional federal funds in 2006 and 2007 because of budget shortfalls, seven (Illinois, Mary­land, Massachusetts, Minnesota, Missouri, New Jer­sey, and Rhode Island) received 79 percent of the funds and were covering children at higher income levels, adults, or, in some cases, both. The forecast for FY 2008 shows shortfalls of $1.2 billion, which are again covered by additional funding.[32] 

These states could forecast their shortfalls in 2006, 2007, and 2008, but instead of modifying their programs, they pressured Congress-i.e., the American taxpayer-for additional funding.

Last year, Congress purposefully set the new allotment levels far beyond projected need in order to disguise funding for the Contingency Fund and Performance Bonus payments, neither of which existed in the original Title XXI legislation. As pre­viously discussed, individual states have overspent their budgets in recent years, and Congress has pro­vided supplemental appropriations to cover the shortfalls.

To avoid the need for congressional review and action in the future should shortfalls occur again even after substantial increases in funding, the vari­ous versions of SCHIP reauthorization included a Contingency Fund. The idea of a Contingency Fund, however, is inconsistent with the original design of capped allot­ments and would send states the clear message that they are free to overspend because more money will always be available. This also relieves Congress of its appropriate oversight and decision-making obligations. By setting federal spending on autopilot, Congress abdicates its responsibili­ties and commits the federal govern­ment to fund any decisions made by the states. This creates a dangerous imbalance in the federal-state part­nership. Therefore, the Contingency Fund should be dropped.

Performance Bonus Uses SCHIP to Federalize Cost of Medicaid. Last year's SCHIP legislation would have created a new Performance Bonus funded with an initial $3 billion appropriation and supple­mented with unexpended allotments. Although the specific design went through a number of incarnations in the legislative process, all shared the same central feature: federal­izing the cost of Medicaid. Under the final version in H.R. 3963, the federal government would, in effect, pay at least an 81 percent match rate for cer­tain additional Medicaid enrollees-compared to the national average of 57 percent.

Moreover, the Performance Bonus formula pro­vided a disproportionate benefit to wealthier states. As shown in Table 5, a wealthier state with a 50/50 Medicaid match rate would receive $906.25 per additional enrollee, compared to $725 per addi­tional enrollee for a poorer state with a 60/40 Med­icaid match rate.

Table 4

The real purpose of the proposed SCHIP Perfor­mance Bonus was to reimburse states for growth in their Medicaid programs. This could result from recruiting "new" enrollees, but it also could reward states for simply "retaining" current enrollees. Because caseloads fluctuate on a monthly basis, a state would not necessarily need to engage in any real outreach efforts to find "new" children; it could still qualify for the bonus by simply keeping the same children on Medicaid for a longer period of time.

However, in order to qualify for a Performance Bonus payment under Section 104 of H.R. 3963, states would have been required to adopt at least five of eight "enrollment and retention" provisions. A state that did not adopt at least five of these pro­visions would not qualify for a Performance Bonus regardless of its success in enrolling eligible but pre­viously uninsured children into Medicaid. The requirements were:

  • 12-month "continuous eligibility" for both Medic­aid and SCHIP children. This would allow indi­viduals to qualify for a full year regardless of any increase in family income.

  • Liberalization of asset requirements. States have authority under current law to impose eligibility limitations based on a family's assets as well as income. A state, for example, could insist that even though a family meets the income test, it could be disqualified if the family also had sub­stantial assets that could be used for the cost of insurance or care. This provision would have required states to eliminate any such asset test for children under their SCHIP and Medicaid pro­grams. Alternatively, the state would have to allow a parent or guardian to make a declaration regarding the family's assets without requiring the family to provide documentation.

  • Elimination of the in-person interview require­ment. States have authority under current law to adopt their own procedures for determining eli­gibility including requiring, if they so choose, a face-to-face interview with the applicant. This provision would have prevented the state from requiring applicants to appear in person.

  • Use of a simple joint application and information verification process for Medicaid and SCHIP cov­erage. As states established their SCHIP pro­grams, they also chose whether to operate their programs as Medicaid expansions or separate programs. They also made decisions about how to "market" SCHIP. States have the authority under current law to design their own SCHIP applications and have made decisions about whether or not their applications included the information needed to perform a determination for Medicaid as well. This provision would require states to use the same application and information verification process for both Medic­aid and SCHIP.

  • Automatic renewal. Under current law, states have the authority to determine their procedures for renewing eligibility. The state would automat­ically renew a recipient's eligibility "unless the state is provided other information." This was a major shift in personal responsibility from the family to the state and a major break in long­standing public policy that requires individuals to be "re-determined" periodically to know whether the individual still meets eligibility criteria.

  • Presumptive eligibility of children for both Medic­aid and SCHIP. This meant beginning benefits before determining whether the individual is even eligible for the public programs. It would also have allowed ineligible individuals such as non-citizens to receive government benefits.

  • Use of Express Lane agencies. Under current law, states have the authority to establish their own procedures for taking applications for SCHIP. Under this provision, states would open up the application process to include additional local agencies such as schools and would accept infor­mation from such agencies that would not be under direct state control. Express Lane eligibil­ity has been pushed by various advocacy groups since the 1990s. It has blown a big hole in the eli­gibility determination process that would allow non-citizens to join the programs, threatening the integrity of the nation's largest means-tested program-Medicaid.

  • Premium assistance subsidies. Under this provi­sion, states that adopted premium assistance for the purchase of employer-based coverage under newly proposed rules in Section 311 of H.R. 3963 would be recognized for purposes of the Performance Bonus. In fact, Section 311 pro­vided states with less flexibility than they cur­rently possess.

These recruitment and retention provisions are not new to SCHIP, and over the course of the past 10 years, states have carefully considered whether to adopt them. Some advocates see these provisions as removing barriers to coverage. Others view them as putting taxpayer dollars at risk. Seven of these eight proposed provisions (all except premium assis­tance) would have weakened the integrity of the eli­gibility process because they involve some form of forfeiting information or relinquishing control. In order to qualify for Performance Bonus funds, states would have been required to change their eligibility determination processes for Medicaid as well as SCHIP. Given that the federal government would eventually proscribe procedures through regula­tion, and given the history of Medicaid litigation related to the eligibility determination processes, these requirements represent greater federalization of the SCHIP program.

Meanwhile, a state that increased Medicaid enrollment through normal "outreach" to needy persons but retained its regular eligibility policies and procedures would not qualify for the Perfor­mance Bonus. In other words, there would be no bonus unless a state were willing to put the integrity of its Medicaid program at risk. Combining the Per­formance Bonus with Express Lane increases the risk of covering ineligible individuals, including non-citizens, by rewarding states with even more federal dollars.

Throwing Kids Off the Cliff. In a mockery of fiscally sound budget rules, all three versions of last year's SCHIP legislation contained a budget gim­mick: It was designed to avoid the fiscal conse­quences of facing an honest 10-year score of the cost of the program as estimated by the CBO. By manipulating the manner in which the federal fund­ing would be provided in 2012, under Sections 101 and 108 of H.R. 3963, the projected costof SCHIP in 2013 dropped by 80 percent. Of course, if this were to happen in real life, all of the children who had been added to the program would lose their coverage. Moreover, even children who are covered under the assumptions of current law, with no change in eligibility, would also lose their coverage.

Unmasking this scheme, Representative Jim McCrery (R-LA) secured data from the CBO that showed SCHIP enrollment dropping by 6.5 million children in the second five-year budget window.[33]  This accounting gimmick was particularly offensive to fiscally responsible Members of Congress from both parties who had pledged to take action to restore integrity to the budget process and reduce the federal deficit. In the next debate, Congress should refrain from such practices.


Congressional leaders seemeddetermined last year to spend a pre-set level of money to the extent of rewarding unsound policies. SCHIP needs a fresh approach. Congress can return SCHIP to its original focus on uninsured low-income children by setting a firm cap on eligibility that applies to both SCHIP and Medicaid and by restoring fiscal discipline.

Instead of expanding SCHIP, policymakers should promote tax relief for working families to enable them to buy private insurance, including employer coverage, and inject the oxygen of healthy members and resources needed to improve and strengthen private insurance pools. In turn, such action will help to make insurance more affordable for everyone.

Dennis G. Smith is Senior Fellow in the Center for Health Policy Studies at The Heritage Foundation.


About the Author

Dennis G. Smith Visiting Fellow
Center for Health Policy Studies