Texas Children's Medicaid and CHIP Enrollment: The Facts

Center for Public Policy Priorities
October 6, 2004

Newspapers across the state have published news coverage and opinion pieces offering various perspectives on the decline in enrollment in Texas' Children's Health Insurance Program (CHIP) in recent weeks. Reporting has made it clear that state elected officials disagree about the changes in CHIP adopted by the 2003 Texas Legislature. Beyond that, readers may not feel the picture is at all clear. Officials and other commentators have made statements and cited statistics which seem to call into question whether CHIP was "cut" or not, or have suggested that the drop in CHIP enrollment is not a problem because growth in the children's Medicaid program, for the state's poorest children, continues.

In response to potentially confusing and at times conflicting assertions presented of late, CPPP has produced this Policy Page, in Q & A format, to provide up to date information about children's Medicaid and CHIP (all from official state agency and legislative sources) against which competing assertions may be judged. While some recent statements are accurate, others are questionable, and all are best understood in context, rather than as isolated "sound bites."

Q: How much has CHIP enrollment dropped since the Legislature's changes took effect?

A: In September 2003 when the state's new budget cycle began, CHIP enrollment was 507,259; as of September 2004, enrollment had dropped to 355,528, for a drop of 151,731 children (30%).

Q : Is it true that "more children are getting health insurance coverage from the state of Texas in 2004 than any other year in history"?

A: No. The high point in combined coverage of children by Texas Medicaid and CHIP occurred in the summer of 2003, when combined enrollment in June and August reached nearly 2.15 million. In August 2004 (the latest month for which both Medicaid and CHIP enrollment data were available for this analysis), Texas' combined coverage of children by Medicaid and CHIP (2.11 million) was more than 11,000 lower than it was in August 2003.

Table 3 (at the end of this report) shows official Texas Health and Human Services Commission (HHSC) child Medicaid and CHIP monthly enrollment numbers from January 2002 through July 2004. Because CHIP enrollment has dropped faster than children's Medicaid has grown since September 2003 (when changes started to take effect), the total combined coverage so far remains lower than it was before the legislature's changes took effect.

Q: Is it true, as one opinion article said, that "today, CHIP and Medicaid serve more than 2.1 million Texas children. That's 1.1 million more Texas children getting health coverage than just five years ago"?

A: Yes. CHIP did not even begin in Texas until May 2000, when children's Medicaid covered just under 1 million children. At that time, HHSC estimated that there were about 600,000 uninsured Texas children below the federal poverty income level (FPL) but not enrolled in Medicaid, and that another 500,000 uninsured children were between the poverty line and twice that amount (200% FPL). The creation of CHIP in May 2000 for children above the poverty line, and the streamlining of children's Medicaid enrollment to make it more like CHIP that began January 2002 have resulted in health coverage for about 1.1 million more Texas children in poor and low-income families in 2004 compared to 2000. (See Table 3 at the end of this document for a history of enrollment.)

Of course, it is important to remember that Texas' child population grows every year; the State Data Center estimates that there are over 303,000 more children under age 18 in Texas in 2004 than in 2000.¹ This means that unless we have annual growth in the number of children who have insurance from some source-private or public-that is equal to the growth in the child population, we lose ground and Texas' percentage of uninsured children gets worse.

According to U.S. Census reports, coverage of children under employer-sponsored insurance has continued to decline for the last three years,² so there is no evidence that the drop in CHIP enrollment has been "made up" by private insurance.

Q : Is it true that on average more children will be covered on Texas CHIP and Medicaid combined in FISCAL YEAR 2004, than in FISCAL YEAR 2003? One article says "more than 200,000 more kids are in those two programs today than there were just two years ago" and another says, "According to the Health and Human Services Commission, an average of 102,000 more children will be on CHIP and Medicaid combined in Fiscal Year 2004 than in Fiscal Year 2003 (before the changes began). The commission expects the increases to continue into 2005."

A: Yes. Some media reports alluding to higher coverage rates in 2004 refer not to the most current month's enrollment and how it compares to a year ago, but instead compare the average combined number of children per month enrolled in Medicaid or CHIP for all 12 months of state fiscal year (FY) 2004 to the average for all of FY 2003 (Texas' fiscal years run from September through August). Medicaid data for August 2004 are now available, and it is true that the combined FY 2004 average is greater than for FY 2003.

This may seem counter-intuitive, given that CHIP enrollment dropped by more children in FY 2004 than children's Medicaid added in the same period. There are a couple of reasons for the lower FY 2003 average that are not too hard to understand. First, annual averages do not tell you if numbers are growing or dropping. For example; a year when enrollment starts at zero and grows to a million children has the very same average as a year that starts at a million children and drops to zero. So, even though combined coverage of children was dropping for much of FY 2004, the annual average cannot reflect that.

The second, even more important factor making the 2003 average lower than 2004 is that children's Medicaid grew so much over that period (due to the Children's Medicaid Simplification policies put in place under SB 43), adding more than 251,000 children to Medicaid. CHIP enrollment basically held steady in FY 2003 (see Table 3 at the end of this document). Simply put, the FY 2003 annual average is low because of the much lower Medicaid enrollment in the early months of that year, even though the combined enrollment of children in the last month of FY 2003 was higher than it is today.

The source of the different numbers used in the two quotations above is an April 2004 HHSC CHIP Caseload Fact Sheet which estimates that "in FY 2004 an average of almost 102,000 more children per month will be enrolled in either Medicaid or CHIP…(and) in FY 2005 an average of about 216,000 more children per month will be enrolled in either Medicaid or CHIP…over FY 2003." While the HHSC data provided in Table 3 show that the increase in the annual combined average from FY 2003 to 2004 was about 65,500 rather than 102,000, this difference is likely due to the age of the Fact Sheet, and the extent to which it assumed higher Medicaid caseload growth in 2004 than actually materialized.

Q : Is it true that some children who left CHIP are now covered by Medicaid?

A: Yes, every month (since CHIP began) some children who try to renew their CHIP coverage get moved to Medicaid.

• HHSC does not currently report the number or percent of CHIP children who are transferred to Medicaid in monthly CHIP renewal data. However, the April 2004 HHSC CHIP Caseload Fact Sheet (cited above) states that about 24% of CHIP disenrollment is due to enrollment in Medicaid, though the sheet does not specify the time period for which this was calculated.
• Children are only moved to Medicaid if their family's income or resources have dropped (usually down below the poverty line). Thus, if all or even a majority of the drop in CHIP coverage was due to children going to Medicaid, it would be a sign that something very bad was happening among Texas' low-income families.
• The number of children added to Medicaid since September 2003 (the month when new policies started phasing in) is less than the CHIP enrollment drop over the same time period.
• The average monthly growth in Children's Medicaid since September 2003 has been much smaller than in FY 2002 or 2003-about half the growth rate of FY 2003 and one-third the rate of FY 2002. If Medicaid were "making up" for the decline in CHIP, that growth rate would have increased, not declined. (See graphic below, Average Monthly Enrollment Growth for Texas Children in Medicaid.)

Again, all the available data suggest that the percentage of children transferring to Medicaid in FY 2004 is no higher than it was in FY 2003; rather, it is probably lower than it was in the previous two years.

Q: Is it true that "more children are enrolled in both (children's Medicaid and CHIP) than experts anticipated"?

A: No. In the 2003 Legislative session, before changes were made to CHIP and children's Medicaid, state officials projected that the average combined child enrollment would have been almost 400,000 higher in 2004 than they now expect, and more than 575,000 higher in 2005 if the changes had not been made (see the table below).

Q: I have heard that most of the drop in CHIP is not because kids got cut off, but because their parents did not renew their coverage when they were supposed to. Is that true? Does that mean the drop in coverage is not the result of the Legislature's changes?

A: Yes, it is true that most (not all) of the drop in CHIP enrollment has been due to parents not renewing their children's coverage. No, this does not mean that the Legislature's changes are not the reason for the drop.

The changes to CHIP enrollment and eligibility adopted by the Legislature were designed to reduce enrollment in CHIP, and thus reduce the state budget. State CHIP officials and the Legislature knew that these changes-in particular, more frequent renewal and higher premiums-would drive down enrollment, because their own history with the CHIP program showed that, and because research in other states has shown that. HHSC officials knew that a very consistent percentage of children lost coverage at renewal time every month (on average, 25% each month). When instructed by the Legislature to propose policy changes that would result in substantial budget reductions, HHSC reasoned that if twice as many children were renewing their coverage each month (which is the effect of going from a 12-month coverage period to a 6-month term of coverage), twice as many children would also lose coverage due to a parent's failure to complete the renewal process. Unlike renewal trends, HHSC did not have prior experience with the impact increased CHIP premiums and co-payments would have on enrollment, but they did know how such increases have affected enrollment in other states.

Generally, policies that reduce enrollment "passively" (e.g., because a parent does not renew coverage, or stops paying when premiums increase) are considered more politically palatable that those that require an active intervention in enrollment. For example, the Legislature might have instead capped CHIP enrollment and started a waiting list for enrollment, or limited new enrollment to certain periods, or simply reduced the upper income limit for CHIP eligibility. Any of these might have been used to reduce enrollment by the same amount Texas has seen, and to reduce the state budget accordingly. While each approach has its own pros and cons, all are conscious and deliberate tactics, each designed and intended to reduce enrollment and spending.

Q: Why should I be concerned about the drop in CHIP coverage if it's almost all due to parents not renewing for their children? Does this mean that only the neediest children are now still enrolled in CHIP? Is it too much to ask parents to fill out forms every 6 months, and to mail in a premium ranging from $15-$25 a month?

A: Readers will of course have to arrive at their own answers to this subjective question. Some relevant facts and ethical considerations are provided below.

• Virtually 100% of the more than 151,000 decline in CHIP enrollment through September 2004 has been among children below 150% of FPL; in fact, enrollment between 150-200% FPL has not dropped at all. So, the changes made to CHIP have not ensured that the neediest children are served first.
• A large share of the drop in CHIP has been among families below poverty,⁴ who do not have to pay CHIP premiums. Because the data clearly reflect that these children are not all simply dropping back into Medicaid eligibility, more needs to be known about the cause of this decline among children in poverty.
• Texas has no data on the percentage of CHIP families below 150% FPL who do not have a checking account, though most experts believe the proportion is high. The group between 100-150% FPL, which went from paying $15 per year to $15 a month ($180 annual) has had the bulk of the decline in enrollment. No policies have yet been adopted to create easy ways for families without checking accounts to pay premiums.
• There have been so many changes to CHIP made at the same time that it is hard to know what is really causing parents to drop coverage: renewing twice a year instead of once; paying higher premiums; the elimination of dental and vision coverage, or a combination of all of these. A survey by experts at Texas CHIP's official quality monitor organization (the Institute for Child Health Policy, or ICHP) was conducted in spring and early summer 2004, and HHSC has promised to release the survey findings soon (release was expected in September 2004).
• Community-based organizations report that CHIP parents whose children need dental care and vision care are choosing to save $180 in premiums in order to pay for the dental care and eyeglasses no longer covered by the program.
• CHIP renewal and premium policies, while not generally considered onerous, do require significantly more time and paperwork from parents, compared to what parents with employer-sponsored insurance must do.
• Outreach and marketing from the state to parents has been dramatically curtailed since the 78th Legislature. Renewed outreach is badly needed to explain all of the changes to CHIP, to eliminate misinformation and confusion about the program, and to encourage new applications by families with young and newly uninsured children. HHSC does not currently report data on application rates for CHIP, so it is not clear to what degree the disappearance of outreach has contributed to the decline in CHIP enrollment.
• One op-ed stated that the "innocence of the children (who lost CHIP coverage) is irrelevant." CPPP disagrees. However strongly one may agree with the notion that a subsidized health insurance program should demand that parents exercise "responsible" behavior, one must also consider which is society's higher priority: protecting children's access to health care, or enforcing some desired level of parental responsibility? Put another way, should children suffer without care because their parents cannot or will not pay CHIP premiums? At minimum, Texas should reinstate outreach, create alternative payment sites for premiums, and reconsider the amounts of premiums charged to the lowest-income families.

Q: Is it true that Texas spent $1.7 million on CHIP in 1998, and that was increased to $563 million in 2003?

A: Yes, but two things should be noted. These figures represent total CHIP funding, including both the state dollars and the much larger federal share. First, Texas did not actually start covering children in CHIP until May 2000. The small figure quoted for CHIP funding in 1998 was actually enhanced federal matching dollars Texas earned for covering teenagers below poverty in the Medicaid program Second, the larger figure quoted for 2003 is roughly accurate, but it should be noted that the cuts to state budget CHIP funding did not take effect until 2004. State dollar (general revenue) CHIP funding for 2004 and 2005 in the budget was about $287 million, compared to $501 million for 2002-2003, a cut of 43%.

Q: Is it true that the state budget for 2004 and 2005 includes $2 billion more in spending on health and human services than in 2002 and 2003?

A: Yes, but the vast majority of this increase was due to increased federal funding for health and human services (HHS). In the budget as passed, federal funds increased by about $1 billion (a 4.4% biennial increase), while state funds for HHS were increased by only $232 million, or 1.5%. Subsequent restorations have added back state and federal funds in about the same relative proportions, so the growth in HHS funding remains well below double-digit medical inflation and caseload growth rates.

This is another area that may seem counter-intuitive. Can it be accurate to say that Medicaid and CHIP are cut, if it is also true that overall HHS spending has increased? Some have editorialized that you cannot say that programs have been cut if there has been any net increase in appropriations.

It is unambiguously clear that state funding for CHIP has been cut by 43% (see answer to the previous question). Where Medicaid is concerned, however, increased spending can and does coincide with service, benefit, and/or eligibility cuts. There are several reasons for this. First, Medicaid, like Medicare, is an entitlement program. That means that once eligibility standards are established, the state must serve all persons who meet those standards, regardless of how much enrollment may increase, or how much drug prices and other health care inflation may increase. So, if enrollment growth and medical costs increase by more than the Legislature increases appropriations for the program, something "has to give." Reduced benefits, stricter eligibility standards, and provider payment cuts are the main options for reducing spending when that happens. Thus, even when total spending goes up, the program is "cut" from the perspective of the individual who is no longer eligible for Medicaid, the health care provider of uncompensated care to those now-uninsured persons, the providers whose rates are cut, or the clients who can no longer get benefits like hearing aids, eyeglasses, or mental health services from Medicaid.

Similar semantic arguments often arise at the Congressional level when Medicare benefits or payments are curtailed. Is the program being "cut" if Congress is appropriating more total funds? While it is valid and important to monitor, question, and try to control increasing health care costs, seniors and health care providers alike certainly view roll-backs in coverage or fees as cuts, even when program funding increases in the aggregate.

Q: One newspaper op-ed referred to CHIP as an "ill-conceived and failing program," saying that "despite billions of new dollars spent on Medicaid and CHIP" the number of the uninsured continues to rise. Is this true? Are these programs simply replacing private insurance?

A: Several facts should be considered in evaluating this claim.

• Focusing first on Medicaid and the mostly below-poverty income population it serves, it should be observed that very few Texans or other Americans below poverty have access to private insurance. U.S. Census data show that fewer than 15% of the poor get insurance through their own job or a family member's job.⁵ Thus, substitution of Medicaid for private coverage is almost unheard of among the working poor.
• The cost of private insurance has grown so much and so fast that it is unrealistic to expect a large percentage of either working poor or low-income workers (families between the poverty line and 200% of that income) or their employers to absorb the costs. New research reports that the average annual cost of family (worker, spouse and children) group health employer-sponsored insurance (ESI) for 2003 was about $10,000.⁶ The annual federal poverty-level income for a family of four in 2004 is $18,850. Clearly, a family in poverty cannot afford that premium-it would take more than half their income. Nor is it likely that many businesses will increase employee compensation by $5,000, let alone the $10,000 needed to cover the whole family.
• The maximum pre-tax income that a family of four can have and still qualify for CHIP coverage is $37,700, but their average actual take-home pay is about $29,200. Again, without a substantial public or employer subsidy, many families at this income level (and their employers) simply cannot afford private coverage. Just 32% of Americans between 100-150% of the poverty line have ESI, increasing to 49% of those between 150-200% of poverty (national average is 64%).
• Public programs have prevented a large increase in uninsured rates among poor and low-income Americans (those below 200% of poverty). Meanwhile, the share of the uninsured that is at higher incomes has grown. The latest census data indicate that ESI has continued to decline over the last several years, and the decline is affecting Americans at higher and higher incomes. In the meantime, public programs have grown slightly, but not nearly enough to offset the drop in private coverage, since CHIP and Medicaid are limited to persons with low and below-poverty incomes and much of the ESI coverage drop is among persons above 200% of poverty. In 1992, 39% of uninsured Americans under 65 had incomes above 200% of poverty, compared to 46% of the uninsured in 2002.⁷ In short, the decline in employer-sponsored insurance is causing the ranks of the uninsured to grow among higher-income families who would never qualify for public programs.
• Policies such as increasing access to Health Savings Accounts, and eliminating mandates that require group health plans to include certain benefits (e.g., maternity coverage, coverage of newborns and adopted children, mammography) have been suggested by conservative commentators as a preferred alternative to public programs. While these approaches may have useful applications for middle-and upper-income Americans, without a major subsidy component they are not likely to have a substantial impact on coverage of the working poor and low-income families of Texas, simply because the cost of coverage relative to income is so very high (and because poorer families have no disposable income to put into a Health Savings Account).
• Why public coverage does not pick up uninsured Texas adults: In Texas, while children may be covered under either Medicaid or CHIP up to 200% FPL, parents cannot make more than $188 per month (for a family of 3) to get public coverage. Working nine hours a week at minimum wage ($5.15) would make a parent with 2 children too well off to get Medicaid in Texas. An adult with no children (unless fully disabled or pregnant) cannot qualify for Medicaid at all, no matter how poor he or she is.
• Finally, most of Medicaid's total spending, and the greatest share of Medicaid cost growth, is related to costs of care for Americans with serious disabilities and the elderly poor. To suggest, in the face of this and all the information provided above, that increased spending on Medicaid and CHIP should have prevented the decline in ESI coverage for persons who make far too much to qualify for public benefits is illogical.

For more information about this Policy Page contact the Center for Public Policy Priorities. You may get in touch with Lynsey Kluever, Communications Director, or , Assistant Director, at (512) 320-0222 X102.

Related Information:

• PDF version of this page [pdf]
• Families talk about how CHIP cuts have affected them
• About CHIP Factsheet [pdf]