HMGT 435 WEEK 8 DISC 8

MichaelAMorrise_2020_Chapter24MedicaidCrow_HealthInsuranceThirdE.pdf

Home >Science homework help >HMGT 435 WEEK 8 DISC 8

CHAPTER

479

24MEDICAID, CROWD-OUT, AND LONG-TERM CARE INSURANCE

Medicaid is a joint federal–state program enacted along with Medicare in 1965. Unlike Medicare, however, Medicaid is a needs-based pro- gram designed to provide health insurance coverage to low-income

individuals and, in particular, pregnant women, children, the elderly, and people with disabilities. Substantial expansions of Medicaid in the mid- to late 1980s and again in 1996 extended coverage to children in households with higher incomes through the Children’s Health Insurance Program (CHIP). The Affordable Care Act (ACA) expanded Medicaid in 2014 for those aged 19–64 living in states that chose to expand eligibility. Nationally, Medicaid covered some 72.8 million people in 2018 (Centers for Medicare & Medicaid Services [CMS] 2019a). Because it is a joint federal–state program, Medicaid eligibility and coverage vary considerably from state to state.

This chapter provides a broad description of the Medicaid (and CHIP) programs, with respect to both eligibility and covered services. It also dis- cusses the ACA Medicaid expansion, building on our discussion in chapter 2, and identifies the economic incentives facing the states in the expansion decision. A key element in the Medicaid program is crowd-out—the extent to which Medicaid expansions have reduced coverage in the private insurance market. Finally, we discuss the private long-term care insurance market. This topic may seem surprising to include, but the long-term care insurance mar- ket has failed to develop largely because Medicaid has crowded out private coverage for many.

Medicaid Overview

Medicaid is not a single government-sponsored program. Rather, it is essen- tially an umbrella of programs, some that states must offer and others that states can opt to offer. There are 28 types of patients that must be covered, and 21 optional eligibility types. However, the programs essentially provide coverage for four groups of low-income people: (1) pregnant women and adults in families with children; (2) children; (3) the elderly; and (4) indi- viduals with disabilities. The federal government has established categories of services that must be covered by state Medicaid programs and also has

C o p y r i g h t 2 0 2 0 . A U P H A / H A P B o o k .

A l l r i g h t s r e s e r v e d . M a y n o t b e r e p r o d u c e d i n a n y f o r m w i t h o u t p e r m i s s i o n f r o m t h e p u b l i s h e r , e x c e p t f a i r u s e s p e r m i t t e d u n d e r U . S . o r a p p l i c a b l e c o p y r i g h t l a w .

EBSCO Publishing : eBook Collection (EBSCOhost) - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS AN: 2459482 ; Michael A. Morrisey.; Health Insurance, Third Edition Account: s4264928.main.eds

Health Insurance480

identified optional services that a state may choose to offer. Within these strictures, the states have considerable flexibility on the level of mandatory services they provide as well as whether they provide any of the optional services. In addition, the states have flexibility on the eligibility criteria they apply for the program.

Federal–State Funding The share of Medicaid medical spending that federal sources cover is deter- mined by the FMAP, the Federal Medical Assistance Percentage. It is com- puted as the relative per capita income of the state. The matching formula is

Federal share = 1.0 – [(State per capita income2 / Federal per capita income2) × 0.45].

Per capita incomes are the average of the preceding three years, and the federal share is constrained by law to be no less than 50 percent and no more than 83 percent. Thus, the federal share in a poor state, such as Mis- sissippi, is approximately 75 percent, while an affluent state, such as Con- necticut, has a federal match closer to 50 percent. The formula is designed so that the average state will get a 55 percent federal match. In contrast, the federal match on most administrative costs is limited to 50 percent, although some elements of Medicaid, such as fraud and abuse efforts, get a somewhat higher federal match. In 2019, the FMAP was highest in Mississippi, West Virginia, and New Mexico, with rates of 76.39 percent, 74.34 percent, and 72.71 percent, respectively. Fourteen states had rates of 50 percent.

Although the FMAP gives a larger federal match to poorer states, some have argued that the 50 percent minimum federal match is nonethe- less overly generous to wealthy states. Pauly and Grannemann (2009), for example, have argued that a better matching arrangement would seek to provide equal benefits and equal tax burden across the states. Their model would take into account both the number of poor persons and taxpayer incomes in each state. Thus, the greatest increases in federal matching, rela- tive to the current system, would come to states with many poor persons and many lower-income taxpayers; the greatest decreases would go to states with few poor people and many high-income taxpayers. New York would see an increase in its Medicaid match under this model, while Connecticut and New Jersey would see substantial reductions.

Eligibility Under the Categorically Needy Programs In general, eligibility for Medicaid is established by being a member of a covered group and having sufficiently low income. That is, individuals are eligible because they fall into a category that Congress has identified as

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 481

appropriate for coverage. Initially, income eligibility for Medicad was deter- mined by virtue of eligibility for welfare—that is, the program that at the time was titled Aid to Families with Dependent Children (AFDC). With the welfare reforms enacted during the Clinton administration, the direct link between welfare and Medicaid eligibility was severed, but Congress contin- ued to apply the rules for AFDC eligibility to Medicaid.

This approach changed with the enactment of the ACA. Under the law, low-income eligibility for Medicaid is determined by one’s modified adjusted gross income (MAGI). The ACA definition of MAGI differs from that used in the Internal Revenue Service tax code, so we will refer to it as the ACA-MAGI. For most people, ACA-MAGI is the adjusted gross income from one’s tax return (basically, wages, salaries, interest, dividends, and tax- able Social Security income). ACA-MAGI adds in nontaxed Social Security income and tax-exempt interest and some foreign earned income. Eligibility is then determined by household size and the federal poverty level (FPL). Exhibit 24.1 shows the applicable 2019 FPL for families with one to four members, along with calculations of incomes at various levels above the FPL. A range of percentages is included to facilitate the discussion of eligibility.

The states have considerable flexibility in how generously they set the eligibility criteria of their Medicaid programs, and they can vary the income level for eligibility for each of the Medicaid programs they offer. Exhibit 24.2 demonstrates this variability. Under the expansions to Medicaid in the 1980s referred to as SOBRA, states have the option of providing differing levels of eligibility for children depending on the child’s age. Florida provides eligibil- ity up to 206 percent of the FPL for children aged 0–1, but only provides coverage for children aged 1–5 at family incomes up to 140 percent of the FPL. Under CHIP, states could add coverage for somewhat higher-income families. The “Children CHIP” row in exhibit 24.2 reflects the eligibility for this program if the state had a separate state program. Florida set its eligibility level at 210 percent of the FPL. NA it means that the state CHIP program

EXHIBIT 24.1 2019 Federal Poverty Level

Persons in Household 100% 133% 150% 200% 250% 300% 400%

1 $12,490 $16,612 $18,735 $24,980 $31,225 $37,470 $49,960

2 $16,910 $22,490 $25,365 $33,820 $42,275 $50,730 $67,640

3 $21,330 $28,369 $31,995 $42,660 $53,325 $63,990 $85,320

4 $25,750 $34,248 $38,625 $51,500 $64,375 $77,250 $103,000

Source: Assistant Secretary for Planning and Evaluation (2019).

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance482

was part of the Medicaid program, rather than a separate program. Note that those states with an NA have higher Medicaid eligibility levels for older children.

Adults in families with dependent children typically have much lower eligibility thresholds than children. The eligibility level is 28 percent in Flor- ida and lower in some other states. Traditionally, this was virtually the only category under which nonpregnant adults could get mandatory coverage. If a state did not expand its program under the ACA, these levels of eligibility still apply. The ACA expansion covers people aged 19–64 who have income below 133 percent of the FPL. The law ignores 5 percent of income, so the eligibility level is actually 138 percent. Note that a handful of states allow somewhat higher levels of eligibility.

Conspicuous by its absence in exhibit 24.2 is eligibility for the blind, disabled, and elderly. They are eligible because they can receive Supplemental Security Income (SSI) rather than being eligible based on the FPL. Medicaid benefits for these folks differ by their allowed income level, but eligibility ranges from approximately 100—200 percent of the FPL.

The categorically needy Medicaid program also provides coverage to people in nursing homes, hospitals, posthospital extended care, and inter- mediate care facilities or facilities for the intellectually disabled (if they have income below 300 percent of the SSI limit). There are asset limitations as well, but these typically exclude the home if a spouse or dependent is living there or if there is a reasonable expectation of a return to the home. If assets are above the allowed threshold, the individual or couple is required to “spend down” their assets to become eligible for Medicaid.

EXHIBIT 24.2 Medicaid

Program Income Eligibility Levels

for Selected States, 2019

(by Percentage of FPL)

Eligible Group California Florida Minnesota New York Ohio Texas

Children 0–1 261% 206% 283% 218% 206% 198%

Children 1–5 261 140 275 149 206 144

Children 6–18 261 133 275 149 206 133

Children on CHIP

NA 210 NA 350 NA 201

Pregnant women

208 191 278 194 200 198

Adult parents 109 28 133 133 90 15

People eligible under ACA expansion

133 [Did not participate

in expansion]

200 200 133 [Did not participate

in expansion]

Source: Data from CMS (2019b).

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 483

Eligibility Under the Medically Needy Programs Currently, 32 states and the District of Columbia have implemented an optional Medicaid medically needy program. These programs typically do one or both of the following: First, they often provide coverage for people aged 19–20 who otherwise would not be Medicaid eligible or who are eli- gible because the medically needy program has a higher income threshold. Second, and more generally, the medically needy programs allow individuals who have too much income but who also have high medical expenses to spend down to eligibility. These people become eligible because, after adjust- ing their income for their medical spending, they meet the income threshold. Given these eligibility criteria, those covered under the medically needy pro- grams tend to have higher healthcare spending than other Medicaid-covered groups (Kaiser Commission on Medicaid and the Uninsured 2013).

Eligibility for Certain Medicare-Eligible Groups (Dual Eligible) In addition to those who may be Medicaid-eligible because they reside in a nursing home or other medical facility, three groups of Medicare beneficiaries are eligible for Medicaid. In the world of Medicare and Medicaid, these ben- eficiaries are called dual eligible. The first, as mentioned earlier, are those who are eligible because they are also eligible for SSI. Medicaid pays their Part B and D premiums and the cost sharing associated with the use of covered Medi- care services. In addition, however, they also receive full Medicaid benefits not covered by Medicare. Second, Qualified Medicare Beneficiaries (QMBs) have incomes below 100 percent of the FPL and limited assets. Medicaid pays their Part B and Part D premiums and Medicare cost sharing. However, they are not eligible for other Medicaid services. Finally, Specified Low-Income Medicare Beneficiaries (SLMBs) have incomes between 100 and 120 percent of the FPL. Medicaid pays their Part B and Part D premiums only. Approximately 20 per- cent of Medicare beneficiaries are also covered by Medicaid in some fashion.

Medicaid Recipients and Expenditures

Exhibit 24.3 is the single best summary of the very complex Medicaid pro- gram. It presents the Congressional Budget Office’s (CBO 2019) estimates of the four principal groups covered by the program and provides a sense of the distribution of the enrollees and expenditures made on behalf of the 75 million people who were estimated to receive Medicaid services in 2019. More than 40 percent of those receiving services were children, but they only make up 20 percent of spending. In contrast, individuals with disabilities and the elderly constitute only 20 percent of enrollees but 46 percent of expenditures.

Enrollment and expenditures by category have changed dramatically over the last decade as a result of the ACA. In 2009, the number of covered adults

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance484

was less than half the number of children. This statistic largely reflects the enroll- ment of adults 19–64 in the 34 states that expanded their Medicaid programs.

This relationship is made clearer by examining the average expenditure per recipient in 2019 (see exhibit 24.4). Expenditures for children (includ- ing CHIP) and adults averaged $2,450 and $4,360, respectively. However, average expenditures on behalf of individuals with disabilities and the elderly were $13,470 and $7,950, respectively.

Medicaid-Covered Services

Medicaid specifies certain mandatory services that state programs must cover (see Medicaid-Covered Services). There are also several optional benefits, any number of which a state may choose to include in its program. For example,

EXHIBIT 24.3 CBO

Estimates of Enrollment and

Expenditures in the Medicaid Program, 2019

Enrollment in millions 35

140

Children Adults Blind/disabled Aged Children Adults Blind/disabled Aged

28 31 122 122

120

100

Expenditures in billions of dollars

Source: Data from CBO (2019).

EXHIBIT 24.4 Average Federal

Expenditures per Person,

2019

16,000

14,000

12,000

10,000

8,000

6,000

4,000

2,000

2,450

4,360

13,470

7,950

AgedBlind/disabledAdults

In dollars

Children

Source: Data from CBO (2019).

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 485

dental services are required for children but optional for adults. Some states exclude adult dental coverage; others cover it under differing circumstances. The states, however, maintain considerable discretion about the level of benefits to provide within each of the mandated or optional services they offer. Alabama, for example, covers 16 inpatient hospital days, 3 nonemer- gency hospital outpatient visits, and 14 physician visits per calendar year as part of its mandatory benefits.

Medicaid-Covered Services

Mandatory Services

• Inpatient hospital services

• Outpatient hospital services

• Rural health clinic and federally qualified health center (FQHC) services

• Laboratory and X-ray services

• Nurse practitioners’ services

• Nursing facility (NF) services and home health services for individuals age 21+

• Early and periodic screening, diagnosis, and treatment (EPSDT) for individuals under 21

• Family planning services and supplies

• Physicians’ services and medical and surgical services of a dentist

• Nurse-midwife services

Optional Services

• Podiatrists’ services

• Optometrists’ services

• Chiropractors’ services

• Psychologists’ services

• Medical social workers’ services

• Nurse anesthetists’ services

• Private duty nursing

• Clinic services

• Dental services

(continued)

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance486

• Physical therapy

• Occupational therapy

• Speech, hearing, and language disorders

• Prescribed drugs

• Dentures

• Prosthetic devices

• Eyeglasses

• Diagnostic services

• Screening services

• Preventive services

• Rehabilitative services

• Intermediate care facilities/services for the intellectually disabled (ICF/MR)

• Inpatient psychiatric services for under age 21

• Christian Science nurses

• Christian Science sanatoriums

• Nursing facility (NF) services for under age 21

• Emergency hospital services

• Personal care services

• Transportation services

• Case management services

• Hospice care services

• Regulatory care services

• TB-related services

• Inpatient and NF services for 65+ in institutions for mental diseases (IMDs)

Source: Reprinted from CMS (2000).

Children’s Health Insurance Program

The Children’s Health Insurance Program (CHIP) was created as part of the 1997 Balanced Budget Act and was reauthorized in 2009, and under the ACA, and in 2018. It is currently authorized through 2028. In essence, CHIP provides federal matching funds for the provision of health insur- ance to children whose family income is up to 350 percent of the FPL, at the state’s discretion. Georgetown University Health Policy Institute (2013) reported that 17 states have eligibility levels of 300 percent of the

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 487

FPL or higher. The eligibility levels changed with the enactment of the ACA-MAGI eligibility standards that we discussed earlier in the chapter. The federal match on CHIP is 15 percentage points higher than for Med- icaid, within the range of 65–83 percent of total costs up to the dollar cap established for each state. If they choose to participate, the states have three ways to provide coverage: (1) they can expand their existing Medicaid pro- gram, (2) they can create a new separate program, or (3) they can develop a combined Medicaid–private program. As of 2015, seven states and the District of Columbia had expanded their Medicaid program, 13 had created a separate program, and the remainder had taken a combination approach (CMS 2015).

One of the reasons many states adopted a separate private program was a concern that Medicaid was stigmatizing. It was believed that parents with eligible children would be more likely to enroll their children in private pro- grams. If the states expanded their Medicaid program, the benefits offered had to be the same as those in their state program. However, if they took another, private, option, the benefits could be as follows (National Health Policy Forum 2004):

• Benchmark coverage—analogous to the Blue Cross Blue Shield coverage or the state employees benefit package available in the state

• Benchmark-equivalent coverage—coverage that was actuarially equal in value to the benchmark option

• The same as plans offered by Florida, New York, or Pennsylvania that were in place prior to the legislation

• A plan of their own creation approved by the CMS

The states may impose premium sharing and copays on services, but the payments may not exceed 5 percent of the family’s income.

In chapter 2, we reviewed the empirical evidence on the effects of the ACA on Medicaid enrollment. The pre-ACA literature examined the effects of Medicaid and CHIP expansions on enrollment and use of services by children. De La Mata (2012) used the 1997, 2002, and 2007 Panel Study of Income Dynamics and the Child Development Supplement. She takes advantage of the differing levels of the FPL eligibility adopted (and periodically changed) by the states to examine take-up and use of services. She found that the increases in eligibility thresholds raised take-up rates by 10–13 percentage points on average and 24–29 points at lower-income eligi- bility levels. She also found that preventive healthcare utilization increased by 12–14 percentage points at the lower-income levels, with no effects at higher levels. However, even among the low-income groups, she found no one- or five-year improvements in health. There were crowd-out effects, as well, that we discuss later.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance488

Copayments and Premiums Under Medicaid and the Children’s Health Insurance Program

In 2005, Congress gave states the authority to charge copayments up to 10 percent of the cost of services for those individuals with family income between 100 and 150 percent of the FPL. Those with higher incomes may be charged up to 20 percent. However, total cost sharing could not exceed 5 percent of the family’s income (CBO 2006). As a result of these limita- tions, few states have implemented cost sharing in the Medicaid population. Indeed, only three states require copayments for children’s Medicaid services, but 26 states do so in their CHIP program.

Beyond the RAND Health Insurance Experiment (chapter 8), little is known of the service-use response to higher copays among families with low- income children. Sen and colleagues (2012) examined the increase in copays in the Alabama CHIP over the 1999–2009 period. They found that the effects of copayment increases varied substantially depending on the service. Copayment increases from $0 to $3 or from $3 to $5, depending on income level, decreased brand-name prescription drug use among the continuously enrolled by 5 percentage points over a full year. A similar increase in physi- cian office visit copays reduced visits by 2.7 percentage points. However, a $5 increase in emergency department visit copays and a $1 increase in generic drug copays had no statistically significant effects.

CHIP programs are allowed to assess a premium on income groups above 100 percent of the FPL; in 2013, 30 states did so (Georgetown Uni- versity Health Policy Institute 2013). Evaluations in Alabama (Morrisey et al. 2012), Florida (Shenkman et al. 2002), New Hampshire and Kansas (Kenney et al. 2006/07), and Arizona (Kenney et al. 2007) suggest that a $50 per year increase in premiums along with copayment increase of $1–3 per visit were associated with about a 6 percent reduction in reenrollment. Larger reductions were found in the Kentucky and Georgia programs (Marton and Talbert 2010).

Managed care plays a significant role in Medicaid. By 2017, approxi- mately two-thirds of all Medicaid recipients—around 54 million individu- als—were in a managed care program (Medicaid and CHIP Payment and Access Commission 2020). The vast majority of these recipients are chil- dren and working-age adults; managed care is much less common among the disabled and elderly Medicaid populations. One reason for this is that many states require that children and pregnant women be enrolled in a comprehensive managed care plan such as a health maintenance organiza- tion (HMO). Some Medicaid enrollees are in limited-benefit plans focusing on, say, behavioral health, or in primary care case-management programs in which primary care physicians are responsible for directing the patient’s care—these are sometimes in addition to the HMOs.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 489

The Medicaid managed care market has some key differences from the private market. Draper, Hurley, and Short (2004) identified these features early on. Some managed care plans have specialized in the Medicaid market; Cen- tene and Molina are examples. Specialization often involves maintaining nar- rowerer networks than in the private sector. However, these narrower networks often have to include FQHCs and traditional Medicaid inpatient providers in the community. The plans also maintain relatively broad service offerings.

While commercial plans have partially moved away from utilization management, Medicaid managed care plans have not. Part of this reflects an inability to use anything other than nominal copays to limit moral hazard.

Finally, while Medicaid managed care plans have continued to use capitation more aggressively than have plans that focus on the private market, whether they have been able to selectively contract with providers is unclear. Early work by Leibowitz, Buchanan, and Mann (1992) demonstrated that Medicaid populations voluntarily enrolled in a managed care plan had sub- stantially lower Medicaid expenditures than did those who were assigned to a managed care plan or who were voluntarily in a fee-for-service arrange- ment. They concluded that apparent Medicaid cost savings from Medicaid managed care was, in fact, the result of favorable selection. If the savings are the result of favorable selection, then requiring all Medicaid-eligible people to participate in the managed care program will not save money. As we saw in chapter 10, the key to cost containment in managed care is selectively contracting on a price basis. It is not clear that Medicaid managed care plans have done this. Work by Duggan and Hayford (2011) is instructive. Using 1991 through 2003 data, they find that shifting Medicaid populations into managed care essentially had no effect on program costs, on average. As the authors say, “These results are consistent with recent research on managed care among the privately insured, which finds that HMOs and other forms of managed care achieve their savings largely through reduced prices rather than lower quantities” (Duggan and Hayford 2011, 1). They also found, however, that in states with low Medicaid reimbursement levels, Medicaid managed care tended to increase spending, while it lowered spending in states with relatively more generous reimbursement levels.

Medicaid Crowd-Out

Crowd-out exists when a public program such as Medicaid causes people to drop private coverage and shift to the public program. We alluded to crowd- out in chapter 16 when we discussed higher out-of-pocket premiums for employer-sponsored family coverage in states with generous CHIP programs. The argument was that some families substitute public coverage for private. Here we examine the phenomenon directly.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance490

Cutler and Gruber (1997) undertook some of the best work on the issue of Medicaid crowd-out. They used the US Census Bureau’s Current Population Survey (CPS) to examine the effects of the SOBRA expansions in coverage for children aged 7–19 between 1987 and 1992. They found that the decline in private coverage as a result of the expansions was roughly 50 percent. That is, for every two children who gained Medicaid coverage, one gave up private coverage.

Other ways to measure crowd-out are sometimes employed. However, as Cutler and Gruber showed, these measures understate the extent of crowd- out. For example, using the same CPS data, they estimated that 22 percent of the expansion in Medicaid over the period was offset by reductions in private coverage. The flaw in this approach is that Medicaid enrollment may have increased for other reasons besides the SOBRA expansion. Crowd-out also has been measured as the proportion of those with private coverage who lost it as a result of the expansion. Cutler and Gruber estimated this to be approxi- mately 15 percent. However, private coverage may change for any number of reasons besides the substitution effect.

Children’s Health Insurance Program and Crowd-Out

LoSasso and Buchmueller (2004) examined the effects of the CHIP expan- sion on coverage and crowd-out. Using much the same methods as Cutler and Gruber, they found that approximately 9 percent of the children meeting the income eligibility criteria gained insurance coverage through CHIP. They also concluded that, if anything, the straight Medicaid expansions were more effective than the separate programs. Thus, they concluded that the stigma of Medicaid was not an issue, or at least no different under the separate pro- grams. They argued that the growth in coverage was more likely the result of explicit outreach programs that CHIP used.

However, more than 46 percent of those who gained CHIP coverage gave up private coverage. The extent of crowd-out was about the same as that found in the earlier Medicaid SOBRA expansion. Not surprisingly, eligible children in families with higher incomes were more likely to move from pri- vate coverage to CHIP. They are the ones more likely to have had employer- sponsored coverage. Thus, one obvious way to reduce crowd-out is to target the program to only low-income families. Gruber and Simon (2008) revisited the crowd-out estimates using 1996–2002 data. They found crowd-out rates of about 60 percent. More recently, De La Mata (2012) and Gresenz and colleagues (2012) have examined Medicaid and CHIP expansions. They too find substantial crowd-out of private coverage, each with estimates at least as large as those of LoSasso and Buchmueller.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 491

Another way to reduce crowd-out is to impose waiting periods before coverage takes effect. LoSasso and Buchmueller (2004) found that a five- month waiting period essentially eliminated crowd-out, but it also reduced the take-up rate by 3.7 percent. Work by others suggests that premium shar- ing can also be an effective means of reducing crowd-out (Davidson, Blewett, and Call 2004). The CHIP out-of-pocket premium reduces the gains from dropping private coverage.

Medicaid Expansion Under the Affordable Care Act

The ACA required the states to expand their Medicaid programs or incur the complete loss of federal matching of their existing Medicaid program. The Supreme Court held in 2012 that this was the equivalent of “hold- ing a gun to the head” of the states and was unconstitutional. As a result, the states have the option of expanding their Medicaid programs to cover US citizens and long-term legal residents aged 19–64 who have household incomes below 138 percent of the FPL. (The legislation calls for coverage below 133 percent of the FPL, but there is also a provision for ignoring the first 5 percent—thus, 138 percent.) As of 2019, 36 states and the District of Columbia have expanded coverage.

The legislation gave the states considerable incentive to adopt an expan- sion. In the first three years, 2014 through 2016, the federal government would pay 100 percent of the claims costs associated with the expansion. The federal matching share declined to 95 percent in 2017, 94 percent in 2018, 93 percent in 2019, and 90 percent thereafter. States that chose not to expand lost the early years of full federal funding and would get only the 90 percent match afterward. (See Expanding the Alabama Medicaid Program for an example of the estimated economic implications of one state’s Medicaid expansion.)

Expanding the Alabama Medicaid Program

Becker and I (2012) undertook an analysis for Alabama in late 2012. The num- bers of eligible people over time were derived from the US Census Bureau’s American Community Survey. We adjusted these data to reflect census esti- mates of changes in population from 2014 to 2020 as well as improvement in the economy and the effects of this improvement on private insurance coverage. This process yielded an estimate of approximately 522,000 eligible people in 2014, declining to 480,000 by 2020. It is unlikely that everyone

(continued)

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance492

eligible will take coverage, however. We used take-up estimates from the Urban Institute for our intermediate scenario. It assumed 75 percent of the uninsured would take Medicaid expansion coverage if available, 60 percent of those eligible but currently with private nongroup coverage, and 25 percent of those eligible currently with private group coverage. This yields estimates of approximately 293,000 likely to take up Medicaid coverage each year.

Estimates of Medicaid spending for this age group are not readily available. As we saw earlier in this chapter, Medicaid spending is concentrated among children, the elderly, and the disabled. We used estimates of health- care spending by those aged 19–64 with private coverage in the Southeast, drawn from the Medical Expenditure Panel Survey (MEPS). The newly covered under Medicaid will have coverage similar to that in the exchanges and, as we saw in chapter 8, those with health insurance use more care than those without coverage. Thus, the insured population is a reasonable proxy. We rolled these costs forward with CMS’s estimate of the trend in real health spending. To these numbers we added the estimated administrative costs. Under our intermediate scenario, these costs range from $1.76 billion in 2014 to $1.85 billion in 2020. (Costs per person increase but enrollment declines with the improving economy, so the value is fairly constant.)

The expansion brings new federal dollars into the state. This is direct spending. Hospitals, physicians, pharmacies, and their employees spend these dollars on gasoline, food, clothing, and so on. Much of this spending leaves the state. However, some stays and is spent again and again. Regional economists have developed input–output models that estimate the size of this indirect spending. For a state such as Alabama without a diverse economy, the indirect multiplier is relatively small. The model we used implied a multiplier of just less than 0.7. Each new dollar generated another 70 cents in spending. This yields new spending of $2.91 billion in 2014, dropping to $2.76 billion in 2015. (It drops because the federal matching share drops over time.)

Finally, the new revenue generates tax revenue. We used an estimate of state tax burden from the Federation of Tax Administrators. This estimate yields an estimate of tax revenue accruing to state and local governments ranging from $250 million in 2014 to $237 million in 2020.

Thus, our intermediate scenario found that over the 2014–2020 period, approximately 293,000 people would be covered each year. The federal gov- ernment would pay approximately $11.7 billion, and the state would have to pay $771 million. The expansion would generate some $19.8 billion in new income. After paying its share, the state, through its various taxing entities, would have some $935 million in net tax revenues.

(continued)

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 493

The states do not get a completely free ride, even in the first three years. Under the legislation, they must absorb their traditional share of the administrative costs associated with coverage for the newly eligible. Most administrative costs are shared 50/50 by the federal government and the state, although certain special programs, such as fraud and abuse control, get a higher federal match. Overall, the states pay about 45 percent of the administrative costs.

Any economic consideration of the expansion depends on six key factors:

1. the number of people likely to be eligible, 2. the take-up rates of the uninsured and people who are eligible but

privately insured, 3. the claims costs associated with the expansion, 4. the federal matching rates over time, 5. the magnitude of direct and indirect new federal spending in the state,

and 6. the effects of this new spending on state tax revenues.

As we noted briefly in chapters 2 and 3, much of the reduction in the number of uninsured in the United States over the last half decade has been the result of the ACA Medicaid expansion. Rigorous early work by Frean, Gruber, and Sommers (2017) used American Community Sur- vey data over the 2012–2015 period and concluded that 60 percent of the expansion in coverage was the result of Medicaid expansions, with the remaining 40 percent stemming from the individual mandate and its related subsidies.

Obviously, there is a strong economic case for expanding the Medicaid program. However, there are at least three strong reasons not to undertake the expansion. First, for all of the potential benefits, the state may not be able or willing to obtain the $771 million it needs for its share of the program. Even when there is a large return on investment, one must still find the resources to make the investment. Second, there is real concern that the federal govern- ment may not be able to keep its promise to continue to fund the expansion at 90 percent of the claims cost into the future. States are concerned that once the expansion occurs, it cannot be undone and they will be saddled with the future costs of a program. Finally, there is the concern that the Medicaid expansion itself is unaffordable at the federal level, and the state should not encourage reckless spending.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance494

Long-Term Care Insurance and Medicaid

Long-term care insurance would seem to be the product that ideally fits the demand for insurance that we discussed in chapter 4. Nursing home care, assisted living, and home health services are expensive. A year in a nursing home can cost $80,000–$100,000. An assisted living facility may charge $4,000 per month, and a home health aide’s visit can cost $10 to $30 per hour. The probability of using such services is certainly in the moderate range. Roughly two-thirds of 65-year-olds in the United States will eventu- ally spend some time in a nursing home, and 22 percent of men and 12 per- cent of women will spend more than three years in one. Yet, Li and Jensen (2011) report that data from the National Institute on Aging’s Health and Retirement Study suggest that only about 4 percent of those older than 50 have a policy. Finkelstein and McGarry (2006) report that only 10 percent of those aged 65 and older have a policy.

Long-term care insurance policies are relatively expensive. Recent online quotes can easily range from $2,000 to $4,000 for a 60-year-old couple with preferred health. A typical policy may cover $150 per day in expenses and three years of care. Policies typically have an elimination period of up to 180 days. These are initial days in a nursing home that are not cov- ered by the policy; as such, they are analogous to a deductible. There appears to be no published research on the effects of coverage options on the use of services. As you see from this summary, long-term care policies are manually underwritten (see chapter 6). The factors used depend on the company and state insurance regulations. Typically, the premiums will be based on age, location, and health status.

A number of arguments have been advanced for why the long-term care insurance market has not developed more significantly (see, e.g., Brown, Goda, and McGarry [2012]). These arguments have focused on both the demand and supply sides of the market.

On the demand side, it is often argued that people misperceive their likelihood of using long-term care services, that they believe that Medi- care covers such care, or that their family will take care of them. In fact, if a person lives long enough, the probability of spending time in a nursing home increases dramatically. Liang and colleagues (1996) estimated that, on average, someone alive at 65 will spend 14 percent of her remaining life in a nursing home. By age 85, this proportion rises to 50 percent, and by age 90, to slightly more than 70 percent. Medicare does provide some long- term care services: up to 100 days of skilled nursing home care per spell of illness and substantial amounts of home health and hospice care. However, most nursing home care is not delivered in skilled facilities but in facilities that provide much less intense levels of care. Moreover, while spouses or

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 495

family members commonly provide caregiver services, the earlier Liang and colleagues estimates belie their ability to substitute out of commercial care- giving services fully.

On the supply side, the arguments are that insurers lack reliable infor- mation on the extent of moral hazard and adverse selection. The information on moral hazard in the long-term care services market is indeed weak. As we saw in chapter 8, very few studies of private nursing home demand have been undertaken, but those that have been suggest substantial price sensi- tivity, with elasticity estimates in excess of 1.0 in absolute value. Thus, we would expect either substantial use of various forms of copayment (such as the elimination periods noted earlier) or high premiums designed to accom- modate this increased usage.

Finkelstein and McGarry (2006) examined the extent of adverse selection in the long-term care insurance market. Using data from the 1995 Asset and Health Dynamics Among the Oldest Old study, they showed that there is adverse selection. After controlling for the factors used by insurers in predicting the use of nursing homes, respondents’ self-reported probability of being in a nursing home within five years was still predictive of both the use of nursing homes and of the purchase of long-term care insurance. This finding implies asymmetric information and the presence of adverse selection. However, Finkelstein and McGarry also found that individual preferences for bearing risk also mattered—and in the opposite direction. Those who were more cautious—that is, more risk averse—were more likely to buy the coverage but less likely to use it. The net effect was that both types of people tended to buy long-term care coverage, and the aggregate effect was the appearance of no adverse selection.

However, the primary reason for the lack of a large private market in long-term care insurance has nothing to do with these factors; it is again crowd-out. In this case, people do not buy coverage because they already have it—Medicaid.

As we noted earlier, an older person can become eligible for Medicaid nursing home services in a variety of ways. In each case, there are income and asset limitations on eligibility. The binding constraint is usually thought to be the asset limitation. There are many anecdotes of people impoverishing themselves either by spending their assets on nursing home care until they become eligible for Medicaid or by transferring their assets to family mem- bers or friends in the years prior to their eligibility for Medicaid. There was concern that this effort also served to impoverish the community-dwelling spouse of someone entering a nursing home. For this reason, the Medicare Catastrophic Coverage Act of 1988 (MCCA) liberalized the income and asset rules when there was a community-dwelling spouse. Most important, it excluded the homestead from the asset considerations when there was such

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance496

a spouse, and it allowed the spouse to keep all income in their name and a portion of the income that was in the spouse’s name.

Sloan and Shayne (1993) examined the extent of such impoverishment before and after MCCA, and by extension, the extent to which people had Medicaid coverage for long-term care. They used the National Long-Term Care Survey and information on state Medicaid policies in 1987 and in 1991. They then simulated the extent to which people with disabilities would have to spend down their assets to be eligible for nursing home care. It is worth noting that their definition of disabilities is relatively modest: those with one or more limitations in activities of daily living. The authors argued convinc- ingly that individuals with disabilities are the relevant population. Those seniors in better health are likely to be able to legally transfer assets before their health deteriorates sufficiently to be disabled.

Sloan and Shayne’s findings, shown in exhibit 24.5, clearly show that, post-MCCA, nearly 78 percent of those at risk of entering a nursing home were already on Medicaid or were immediately eligible. Another 5 percent were eligible within six months. The detail of the table is also instructive. The MCCA had a relatively small but important impact on eligibility. It increased the percentage of people immediately eligible or eligible within six months by 8.7 percentage points. Virtually all of this increase came from increased

EXHIBIT 24.5 Percentage of Disabled

Elderly Eligible for Medicaid on Admission to a Nursing Home

Before Medicare Catastrophic Coverage Act of 1988 (%)

After Medicare Catastrophic Coverage Act of 1988 (%)

Single Married All Single Married All

Already on Medicaid

24.0 10.1 18.7 24.0 10.1 18.7

Immediately eligible

48.5 43.3 46.4 48.9 75.1 59.0

Eligible in 1– 6 months

7.8 11.3 9.2 7.8 1.3 5.3

Eligible in 6– 30 months

7.2 13.4 9.6 7.5 3.1 5.8

Eligible in 30–120 months

4.7 9.1 6.4 4.6 1.6 3.8

Not eligible in 120 months

7.8 12.8 9.7 7.2 7.7 7.4

100 100 100 100 100 100

Source: Sloan and Shayne (1993), “Long-Term Care, Medicaid, and Impoverishment of the Elderly,” Milbank Quarterly 71 (4): 575–99, Table 2. Reprinted with permission.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 497

eligibility among married people. This result is as we would expect because the key feature of the MCCA was to protect the income and assets of a community-dwelling spouse.

The upshot of all of this for the purchase of long-term care insurance is clear. The reason most people do not buy long-term care insurance is that they already have it. Based on the Sloan and Shayne estimates, more than 80 percent of those likely to use a nursing home will be eligible for Medicaid immediately or within six months of entry.

More recently, Brown, Coe, and Finkelstein (2007) examined the crowd- out effects of Medicaid on private long-term care insurance. They concluded that, in the presence of Medicaid and controlling for other relevant factors, between 66 and 90 percent of people would not buy long-term care coverage. The authors showed that there is some sensitivity to the asset threshold that Medicaid imposes to determine eligibility. A $10,000 decrease in qualifying assets would increase private long-term care coverage by 1.1 percentage points.

Summary

• Medicaid is a joint federal–state, need-based program to provide medical services to low-income populations, in particular pregnant women, children, the elderly, and individuals with disabilities.

• The states exercise considerable flexibility with the criteria for Medicaid eligibility, the generosity of services, and the inclusion of optional services.

• While children and adults make up nearly 75 percent of the enrolled Medicaid population, the elderly and those with disabilities have per person costs some seven times higher.

• Medicaid expansions in the 1980s and 1990s have sought to cover people with somewhat higher family incomes. There is evidence of substantial private insurance crowd-out as a result of these expansions, typically on the order of one person giving up private coverage for every two gaining public-sector benefits.

• The Medicaid expansion under the ACA is estimated to have accounted for 60 percent of the reduction in the number of uninsured over the first two years of the program. From an individual state’s perspective, the expansion brings substantial new federal dollars into the state. States, however, are concerned that the federal government will not continue to pay for the program over time.

• Under current laws, long-term care insurance is unlikely to be a major market. Estimates suggest that nearly 80 percent of likely nursing home residents are already on Medicaid or will be immediately eligible on entering a nursing home.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance498

Discussion Questions

1. Under what conditions would a Medicaid managed care program likely be successful in restraining Medicaid costs?

2. What effect would the ACA have on enrollment in CHIP? What effect would an expanding economy have on CHIP enrollment? Why?

3. In 2006, Congress enacted legislation that prevents persons with a home valued at more than $500,000 from qualifying for Medicaid; states can raise this limit to $750,000. (Homesteads of whatever value are exempt when a community-dwelling spouse or other dependent is living there.) In addition, Medicaid will look back for five years instead of three in determining whether assets have been transferred in anticipation of Medicaid eligibility. What effects do you expect this action to have on Medicaid eligibility and the demand for long-term care insurance?

4. Should a state expand its Medicaid program as allowed under the ACA? Discuss the advantages and disadvantages of such an expansion.

5. Suppose you are a reasonably wealthy individual. Under what circumstances would you buy long-term care insurance rather than investing in other assets?

For the Interested Reader

De La Mata, D. 2012. “The Effect of Medicaid Eligibility on Coverage, Utilization, and Children’s Health.” Health Economics 21 (9): 1061–79.

Finkelstein, A., and K. McGarry. 2006. “Private Information and Its Effect on Mar- ket Equilibrium: New Evidence from Long-Term Care Insurance.” American Economic Review 96 (4): 938–58.

Gruber, J., and K. Simon. 2008. “Crowd-Out 10 Years Later: Have Recent Public Insurance Expansions Crowded Out Private Health Insurance?” Journal of Health Economics 27 (2): 201–17.

References

Assistant Secretary for Planning and Evaluation. 2019. “2019 Poverty Guidelines.” US Department of Health & Human Services. Accessed December 10. https://aspe.hhs.gov/2019-poverty-guidelines.

Becker, D. J., and M. A. Morrisey. 2012. “An Economic Evaluation of Medicaid Expansion in Alabama Under the Affordable Care Act.” University of Ala- bama at Birmingham School of Public Health. Published November 5. www.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 499

soph.uab.edu/files/faculty/mmorrisey/Becker-Morrisey%20Study%20of%20 Alabama%20Medicaid%20Expansion%202012.pdf.

Brown, J. R., N. B. Coe, and A. Finkelstein. 2007. “Medicaid Crowd-out of Private Long-Term Care Insurance Demand: Evidence from the Health and Retire- ment Survey.” Tax Policy and the Economy 21: 1–34.

Brown, J. R., G. S. Goda, and K. McGarry. 2012. “Long-Term Care Insurance Demand Limited by Beliefs About Needs, Concerns About Insurers, and Care Available from Family.” Health Affairs 31 (6): 1294–302.

Centers for Medicare & Medicaid Services (CMS). 2019a. “April 2019 Medicaid & CHIP Enrollment Data Highlights.” Accessed December 26. www.medicaid. gov/medicaid/program-information/medicaid-and-chip-enrollment-data/ report-highlights/index.html.

———. 2019b. “Medicaid, Children’s Health Insurance Program, and Basic Health Program Eligibility Levels.” Accessed December 10. www.medicaid.gov/ medicaid/program-information/medicaid-and-chip-eligibility-levels/index. html.

———. 2015. “CHIP State Program Information.” Accessed December 26, 2019. www.medicaid.gov/chip/state-program-information/index.html.

———. 2000. A Profile of Medicaid: Chartbook 2000. Washington, DC: Centers for Medicare & Medicaid Services.

Congressional Budget Office (CBO). 2019. “Medicaid—CBO’s May 2019 Base- line.” Accessed December 10. www.cbo.gov/system/files/2019-05/51301- 2019-05-medicaid.pdf.

———. 2006. Medicaid Spending Growth and Options for Controlling Costs. 109th Congress. Testimony of Donald B. Marron, Acting Director, CBO. Published July 13. www.cbo.gov/sites/default/files/109th-congress-2005-2006/ reports/07-13-medicaid.pdf.

Cutler, D. M., and J. Gruber. 1997. “Medicaid and Private Insurance: Evidence and Implications.” Health Affairs 16 (1): 194–200.

Davidson, G., L. A. Blewett, and K. T. Call. 2004. Public Program Crowd-Out of Private Coverage: What Are the Issues? Published June 1. www.shadac.org/ sites/default/files/Old_files/CrowdOut_Report_Jun04.pdf.

De La Mata, D. 2012. “The Effect of Medicaid Eligibility on Coverage, Utilization, and Children’s Health.” Health Economics 21 (9): 1061–79.

Draper, D. A., R. E. Hurley, and A. C. Short. 2004. “Medicaid Managed Care: The Last Bastion of the HMO?” Health Affairs 23 (2): 155–68.

Duggan, M., and T. Hayford. 2011. “Has the Shift to Managed Care Reduced Medicaid Expenditures? Evidence from State and Local-Level Mandates.” National Bureau of Economic Research. Accessed May 23, 2013. www.nber. org/papers/w17236.

Finkelstein, A., and K. McGarry. 2006. “Private Information and Its Effect on Mar- ket Equilibrium: New Evidence from Long-Term Care Insurance.” American Economic Review 96 (4): 938–58.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Health Insurance500

Frean, M., J. Gruber, and B. D. Sommers. 2017. “Premium Subsidies, the Mandate, and Medicaid Expansion: Coverage Effects of the Affordable Care Act.” Jour- nal of Health Economics 53: 72–86.

Georgetown University Health Policy Institute. 2013. “Medicaid and CHIP Programs.” https://ccf.georgetown.edu/facts-statistics/Medicaid-chip- program.

Gresenz, C. R., S. E. Edgington, M. Laugesen, and J. J. Escarce. 2012. “Take-Up of Public Insurance and Crowd-Out of Private Insurance Under Recent CHIP Expansions to Higher Income Children.” Health Services Research 47 (5): 1999– 2011.

Gruber, J., and K. Simon. 2008. “Crowd-Out 10 Years Later: Have Recent Public Insurance Expansions Crowded Out Private Health Insurance?” Journal of Health Economics 27 (2): 201–17.

Kaiser Commission on Medicaid and the Uninsured. 2013. “The Medicaid Program at a Glance.” Published March 4. www.kff.org/medicaid/7235.cfm.

Kenney, G., R. A. Allison, J. F. Costich, J. Marton, and J. McFeeters. 2006/2007. “Effects of Premium Increases on Enrollment in SCHIP: Findings from Three States.” Inquiry 43 (4): 378–92.

Kenney, G., J. Marton, J. McFeeters, and J. Costich. 2007. “Assessing Potential Enrollment and Budgetary Effects of SCHIP Premiums: Findings from Ari- zona and Kentucky.” Health Services Research 42 (6, part II): 2354–72.

Leibowitz, A., J. L. Buchanan, and J. Mann. 1992. “A Randomized Trial to Evaluate the Effectiveness of a Medicaid HMO.” Journal of Health Economics 11 (3): 235–57.

Li, Y., and G. A. Jensen. 2011. “The Impact of Private Long Term Care Insurance on the Use of Long-Term Care.” Inquiry 48 (1): 34–50.

Liang, J., X. Liu, E. Tu, and N. Whitelaw. 1996. “Probabilities and Lifetime Dura- tions of Short-Stay Hospital and Nursing Home Utilization in the U.S. 1985.” Medical Care 34 (10): 1018–36.

LoSasso, A. T., and T. C. Buchmueller. 2004. “The Effect of State Children’s Health Insurance Program on Health Insurance Coverage.” Journal of Health Eco- nomics 23 (6): 1059–82.

Marton, J., and J. C. Talbert. 2010. “CHIP Premiums, Health Status, and the Insur- ance Coverage of Children.” Inquiry 47 (3): 199–214.

Medicaid and CHIP Payment and Access Commission (MACPAC). 2020. “Exhibit 29. Percentage of Medicaid Enrollees in Managed Care by State.” Accessed February 11. www.macpac.gov/publication/percentage-of-medicaid- enrollees-in-managed-care-by-state.

Morrisey, M. A., J. Blackburn, B. Sen, D. Becker, M. L. Kilgore, C. Caldwell, and N. Menachemi. 2012. “The Effects of Premium Changes on ALL Kids, Alabama’s CHIP Program.” Medicare & Medicaid Research Review 2 (3): E1–E17.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use

Chapter 24: Medicaid, Crowd-Out, and Long-Term Care Insurance 501

National Health Policy Forum. 2004. The Basics: SCHIP. Washington, DC: George Washington University.

Pauly, M. V., and T. W. Grannemann. 2009. “Equal-Burden-for-Equal-Benefit Medicaid Matching Rates.” American Enterprise Institute. Published July 6. www.aei.org/research-products/working-paper/equal-burden-for-equal- benefit-medicaid-matching-rates.

Sen, B., J. Blackburn, M. A. Morrisey, M. L. Kilgore, D. J. Becker, C. Cadwell, and N. Menachemi. 2012. “Did Copayment Changes Reduce Health Services Utilization Among CHIP Enrollees? Evidence from Alabama.” Health Ser- vices Research 47 (4): 1603–20.

Shenkman, E., B. Vogel, J. M. Boyett, and R. Naff. 2002. “Disenrollment and Reen- rollment Patterns in a SCHIP.” Health Care Financing Review 23 (3): 47–63.

Sloan, F. A., and M. W. Shayne. 1993. “Long-Term Care, Medicaid, and Impover- ishment of the Elderly.” Milbank Quarterly 71 (4): 575–99.

EBSCOhost - printed on 2/6/2023 8:43 AM via UNIVERSITY OF MARYLAND GLOBAL CAMPUS. All use subject to https://www.ebsco.com/terms-of-use