MAY 2004

Issue Brief

How Asset Tests Block LowIncome Medicare Beneficiaries from Needed Benefits
Laura Summer and Lee Thompson Center on an Aging Society Georgetown University Health Policy Institute
For more information about this study, please contact: Laura Summer, M.P.H. Senior Researcher Health Policy Institute Georgetown University Tel 202-687-3595 Fax 202-687-3110 E-mail [email protected] or Barbara S. Cooper Senior Program Director The Commonwealth Fund Tel 212.606.3862 Fax 212.606.3500 E-mail bsc@ cmwf.org

ABSTRACT: The Medicare Savings Programs and Medicaid help elderly Medicare beneficiaries with their cost-sharing responsibilities and provide much-needed additional benefits. But 40 to 50 percent of eligible low-income Medicare beneficiaries are not enrolled in the Medicare Savings Programs. One persistent barrier is the use of asset tests, which greatly complicate the application process for applicants and program staff. Older people with low incomes generally have few assets and the income and assets of older people who qualify for public programs do not change substantially over time. Findings suggest that use of asset tests should be eliminated or asset limits should be increased; at a minimum, cost-of-living adjustments should be made. Findings also indicate that the renewal process should be simplified, and use of verification documents should be reduced or eliminated.

Additional copies of this (#727) and other Commonwealth Fund publications are available online at

www.cmwf.org
To learn about new Fund publications when they appear, visit the Funds website and register to receive e-mail alerts.

OVERVIEW The cost of health care can be a significant financial burden for older Americans with low incomes. The Medicare program is an important source of health care coverage for about 35 million people age 65 and older and close to 6 million adults with disabilities under age 65, but costs can be steep even with Medicare coverage. The Medicare Savings Programs and Medicaid provide important benefits to some of the neediest seniors and disabled. They fill in Medicare cost-sharing responsibilities and, in the case of Medicaid, provide much-needed additional benefits (Table 1). But 40 percent to 50 percent of eligible low-income Medicare beneficiaries are not enrolled in the Medicare Savings Programs.1 A number of factors contribute to the low enrollment rate. One persistent barrier is the use of asset tests, which greatly complicate the application

The Commonwealth Fund

Table 1. Financial Eligibility Criteria and Benefits for Medicaid and the Medicare Savings Programs* Program Full Medicaid Coverage** Qualified Medicare Beneficiary (QMB) Program Specified Low-Income Medicare Beneficiary (SLMB) Program Qualifying Individuals I (QI-I) Programs***** Countable Income Limits At or below 75 percent of the federal poverty level*** At or below 100 percent of the federal poverty level Countable Asset Limits $2,000 for an individual $3,000 for a couple $4,000 for an individual $6,000 for a couple Benefits Coverage for a broad range of health care services Medicaid pays all Medicare Part B premiums ($66.60 per month in 2003) and cost-sharing charges**** Medicaid pays Medicare Part B premiums ($66.60 per month in 2003) Medicaid pays Medicare Part B premiums ($66.60 per month in 2003)

Between 100120 percent of the federal poverty level Between 120135 percent of the federal poverty level

$4,000 for an individual $6,000 for a couple $4,000 for an individual $6,000 for a couple

* This chart shows standard income and asset eligibility criteria. In counting income or resources, however, states may also use methods that are less restrictive than those specified for the Medicaid, Qualified Medicare Beneficiary (QMB), Specified Low-Income Medicare Beneficiary (SLMB), and Qualifying Individuals (QI) programs. In those instances, income and asset limits are higher than those listed in the chart. ** The primary eligibility pathway for elderly individuals receiving full Medicaid coverage is through the Supplemental Security Income (SSI) program, which has income eligibility limits of approximately 75 percent of the federal poverty level. States have the option, however, of extending full Medicaid coverage to people age 65 and older at higher poverty levels. Specifically, they may cover elderly people whose incomes do not exceed 100 percent of the federal poverty level and whose resources do not exceed $2,000 for an individual or $3,000 for a couple. States may also use less restrictive methods to adjust income and asset limits. In 2001,16 states and the District of Columbia had income eligibility limits at 100 percent of the federal poverty level or higher. *** In 2003, the federal poverty level was $9,310 for individuals and $12,490 for couples. **** States are not required to pay for cost sharing if the Medicaid payment rates for a given service are substantially lower than the Medicare payment rates. ***** The QMB and SLMB programs are entitlement programs, but the QI program is not. Federal QI funding is capped each year and is due to expire September 30, 2004, unless Congress passes new legislation. Sources: Andy Schneider, Risa Elias, Rachel Garfield, David Rousseau, and Victoria Wachino (2002). The Medicaid Resource Book. The Kaiser Commission on Medicaid and the Uninsured; Henry J. Kaiser Family Foundation. State Health Facts Online. 50 State Comparisons: Medicaid Coverage Expansions for Medicare Beneficiaries (available at http://www.statehealthfacts.org).

process. Most applicants are required to provide documents verifying their assets. In focus groups, older Americans note that the number of verification documents required presents difficulties, particularly if they cannot locate the documents or must copy them.2 Older applicants also have misperceptions about eligibility rules regarding assets.3 Some people do not know that certain assets, such as the home, are not counted in determining eligibility. As a result, they may not apply for benefits even though they are eligible. Others are reluctant to apply because they incorrectly believe that they may have to give up their homes or other assets to receive benefits.4 The use of asset tests also complicates program administration. Reviewing asset information is the most time-consuming task in the enrollment

process. Eligibility workers often have to help applicants identify and obtain the needed documents, and then must copy, review, and return the papers. Difficulties related to producing and reviewing asset verification documents occur both upon the initial application and later when beneficiaries must reenroll in the programs. Most states require annual reenrollment. This issue brief presents findings from an analysis of data on the income and assets of lowincome older Americans. Data were drawn from the 1998 and 2000 Health and Retirement Study (HRS) and from the 1993 and 1995 Study of Assets and Health Dynamics Among the Oldest Old (a companion study to the HRS). The authors examine the kinds and amounts of assets that lowincome Medicare beneficiaries possess, the extent to

How Asset Tests Block Low-Income Medicare Beneficiaries from Needed Benefits

which the financial circumstances of beneficiaries change from year to year, and the implications of these findings for Medicare Savings Programs enrollment and renewal processes. MORE THAN 7 MILLION QUALIFY FOR BENEFITS BASED ON INCOME Some 2.3 million elderly individuals have incomes at or below the minimum eligibility limits for full Medicaid benefits. A total of 4 million seniors are eligible for the Qualified Medicare Beneficiary (QMB) program. This group includes the 2.3 million people who qualify for full Medicaid coverage and an additional 1.7 million with incomes between 75 and 100 percent of the federal poverty level. About 3.1 million people qualify for the Specified Low-Income Medicare Beneficiary (SLMB) or Qualifying Individuals (QI) programs (Table 2). Older People with Low Incomes Generally Have Few Assets In determining eligibility for Medicaid and the Medicare Savings Programs, countable assets include items such as money in checking or savings accounts, bonds, stocks, or mutual funds. Part of the value of assets such as vehicles and life insurance is also counted, as is the value of any real estate other than the applicants home.

The median value of countable assets is just $300 for persons with incomes at 75 percent of poverty or less, and only $8,000 for persons with incomes between 100135 percent of poverty. Not surprisingly, many people who qualify for programs based on income also qualify based on assets. A substantial proportion of people age 65 and older who meet program income limits have no countable assets. Many others have assets valued at or below the program limits. For example, 66 percent of people with incomes below the poverty level also have assets below the limits for the QMB program, and about two-thirds of these people have no assets at all (Figure 1).5

Table 2. Number of People Age 65 and Older Who Meet the Income Criteria for Medicaid or the Medicare Savings Programs Full Medicaid (Income at or below 75 percent of the federal poverty level) All 65 and older 6574 7584 85 and older 2,323,372 1,134,156 798,536 390,680 QMB (Income at or below 100 percent of the federal poverty level)* 4,047,276 1,853,272 1,449,603 744,401 SLMB or QI (Income from 101135 percent of the federal poverty level) 3,054,879 1,274,321 1,254,516 526,042

* Includes individuals in the first category (< 75 percent). Source: Calculations by the Center on an Aging Society based on data from the 2000 Health and Retirement Study.

The Commonwealth Fund

Older people who qualify for programs based on income but fail the asset test are not wealthy. Among people age 65 and older who meet income but not asset eligibility limits, the median amount by which they exceed asset limits is $6,500. Almost one-third30 percentof people with incomes at or below the poverty level do not qualify for QMB program benefits solely because they have life insurance policies. In other words, the only

countable asset they have is a life insurance policy valued higher than $1,500. Seventy-two percent of the people in this group have life insurance policies that exceed the allowable limit by $8,500 or less. The asset test provides a deduction of $4,500 for a vehicles value. About half of the people who qualify for Medicaid and the Medicare Savings Program based on income but not assets have vehicles that exceed the limit by $5,500 or less (Table 3).

Table 3. Median Value of Selected Assets for People Age 65 and Older Who Are Eligible for Programs Based on Income, But Not Assets Full Medicaid (Income at or below 75 percent of the federal poverty level) Median amount by which life insurance policies exceed the $1,500 limit Median value of funds in checking or savings accounts Median amount by which vehicles exceed the $4,500 limit QMB (Income at or below 100 percent of the federal poverty level)* SLMB or QI (Income from 101135 percent of the federal poverty level)

$6,000

$8,500

$8,500

$5,500

$7,000

$7,500

$5,500

$5,500

$7,500

* Includes individuals in the first category (< 75 percent). Source: Calculations by the Center on an Aging Society based on data from the 2000 Health and Retirement Study.

Asset Tests and the New Medicare Prescription Drug Benefit The Medicare Prescription Drug, Improvement, and Modernization Act of 2003 will subsidize drug benefit premiums and related costs for some low-income Medicare beneficiaries when it takes effect in 2006. Beneficiaries will have to apply for this assistance, however, and eligibility will be determined based on evaluations of income and assets. The income and asset limits for drug benefit subsidies will be somewhat higher than those for the Medicare Savings Programs, and the asset limits will be indexed for inflation. The asset limits for applicants with incomes less than 135 percent of the federal poverty level (FPL) are $6,000 for individuals and $9,000 for couples who receive the largest subsidies. Other applicants in that income group who have assets valued somewhat higherfrom $6,000 to $10,000 for individuals and from $9,000 to $20,000 for couplesqualify for smaller subsidies. Applicants with incomes from 135 percent FPL to 150 percent FPL also may qualify for the smaller subsidies if they have assets valued at less than $10,000 for an individual and $20,000 for a couple. An estimated 5.6 million people living in the community will be eligible for drug benefit subsidies. This represents just over twothirds67 percentof those who would qualify based on income alone. Currently, Medicare beneficiaries with incomes less than 135 percent FPL qualify for subsidies for the Medicare drug discount card. There is no asset test for the drug discount card.

How Asset Tests Block Low-Income Medicare Beneficiaries from Needed Benefits

Only a small number of people who exceed asset limits have deferred compensation retirement plans in the form of defined contribution plans such as Individual Retirement Accounts (IRAs), Keoghs, or 401K-type plans. These people are at a disadvantage, however, compared to those whose deferred compensation is in the form of a defined benefit plan. The total value of savings in defined contribution plans is considered to be a financial asset. By contrast, payments from defined benefit plans are considered a source of countable income. These rules favor people with defined benefit plans for two reasons: 1) program income limits are higher relative to asset limits, and 2) income limits, but not asset limits, are adjusted for economic growth.6 The proportion of people who have defined contribution rather than defined benefit plans is growing for the population overall. ASSET LIMITS HAVE NOT CHANGED More people would qualify for benefits from the Medicaid and Medicare Savings Programs if asset limits were adjusted for economic growth. Asset limits for the programs have not changed since 1989, despite the fact that the cost of living has increased. If cost-of-living adjustments had occurred, about 100,000 additional people age 65 and older would be eligible to receive Medicaid benefits and another 180,000 would be eligible for the Medicare Savings Programs. This represents a very small increase in the number of people eligible for benefits based on both income and assets (Figure 2). The deductions allowed in counting the value of specific assets also are unchanged since 1989. If

they were adjusted for inflation, the deduction allowed for life insurance would have been $1,926 in 2000 and $2,030 in 2003. After adjustments for inflation, the deduction for vehicles would have been $5,777 in 2000 and $6,089 in 2003.7 A large proportion of beneficiaries are eligible year after year. Nearly 70 percent of people age 70 and older who were eligible for the QMB, SLMB, or QI benefits in 1993 were still eligible seven years later, and 64 percent still qualified for Medicaid (Figure 3).8 More than 90 percent of beneficiaries who exceed limits in subsequent years acquire either

The Commonwealth Fund

1. The use of asset tests can be eliminated. 2. Rules regarding asset tests can be changed. 3. The renewal process can be simplified. 4. The use of verification documents can be reduced or eliminated. Policy changes in these areas have the potential not only to reduce enrollment barriers when people apply for program benefits, but also to reduce administrative costs. Enrollment simplification strategies that reduce administrative costs may be particularly attractive to state Medicaid programs, which administer the Medicaid and Medicare Savings Programs for older Americans. Eliminating Asset Tests To qualify for program benefits, applicants must not exceed separate limits on their incomes and assets. Income and asset tests are used to help target program benefits so that limited resources assist those most in need.Yet the data show that among the older low-income population, income and assets are closely related for a substantial number of people. This suggests that income tests alone would identify those people who most need benefits.9

assets or incomes above the limitsbut not both. Among people who qualified for the QMB program in 1993 but not in subsequent years, a small proportion8 percenthad both income and assets above the limits (Figure 4). The pattern is similar for Medicaid and the other Medicare Savings Programs. For beneficiaries who in subsequent years exceed the limits, the amounts of extra countable income and assets are small (Table 4). RECOMMENDATIONS The findings suggest that four aspects of the enrollment process can be simplified for the older low-income population:

Table 4. Median Value of Extra Annual Income and Assets for People Age 70 and Older Who Were Eligible for Benefits in 1993 But Not in 2000 Income at or below 75 percent of the federal poverty level Median value of extra income Median amount of extra assets
* Includes individuals in the first category (< 75 percent). ** Includes individuals in the first two categories (< 75 percent and < 100 percent). The sample was too small to examine values for the population between 101135 percent of the federal poverty level. Source: Calculations by the Center on an Aging Society based on data from the 1993 AHEAD and the 2000 Health and Retirement Survey.

Income at or below 100 percent of the federal poverty level* $2,812 $6,000

Income at or below 135 percent of the federal poverty level** $1,800 $9,500

$2,220 $3,500

How Asset Tests Block Low-Income Medicare Beneficiaries from Needed Benefits

For example, two-thirds of older people who qualify for the QMB program based on income also qualify based on assetssome 2.7 million people. And even people who would qualify for Medicare Savings Programs based on incomes but not assets typically have minimal assets. The value of those assets could easily drop below program limits if the potential applicants had to use their modest savings to pay for home repairs, unexpected medical expenses, or similar items. One-third of people who would qualify for the QMB program based on incomes but not assets have extra countable assets valued at or below $4,700. In 2000, elderly Medicare beneficiaries spent an average of just over $3,000 on outof-pocket medical expenses.10 Despite the potential for increased program enrollment if asset tests were eliminated, four statesAlabama, Arizona, Delaware, and Mississippihave modified the tests to effectively eliminate them for Medicare Savings Programs. Connecticut and New York disregard all assets for the QI program.11 Arizona eliminated the asset test for Medicare Savings Programs in 2001 after conducting a fiscal impact study. The study found that savings on administrative costs related to documenting assets roughly equaled the costs of benefits for additional persons who would enroll in the programs.12 Nineteen states that offer Medicaid benefits for parents in low-income families do not use asset tests. Medicaid officials in those states report that asset tests require significant staff time and that few denials occur because of excess assets. They said that eliminating asset tests raises the productivity of eligibility workers, makes it easier to use automated eligibility determination systems, and reduces administrative costs.13 For example, before Oklahoma eliminated the Medicaid asset test for families, officials concluded that doing so would save about $1 million. The savings would come from the difference between the $3.5 million spent on administrative activities related to verifying assets and the $2.5 million spent on benefits for

additional persons who would qualify.14 Most states do not use asset tests as part of the Medicaid enrollment process for children.15 Reducing costs is especially crucial now because the new drug benefit subsidy program will most likely result in additional administrative costs and complexity for state Medicaid programs, which will be responsible for administering it in addition to the Medicare Savings programs. The Social Security Administration is also slated to play a role in administering the drug benefit program. There is likely to be some confusion among older low-income Medicare beneficiaries regarding availability of the two types of benefits, the need in some cases to apply separately for each one, and the different eligibility rules for the benefits. Plans for implementing the drug benefit subsidy program should include efforts to simplify the enrollment and renewal processes. It may be beneficial to consider aligning eligibility rules for the new drug benefit subsidy with those for the Medicare Savings Programs to facilitate enrollment in both. Changing Asset Rules If asset limits are retained to determine benefit eligibility, they should be raised. The data show that if asset limits were adjusted to account for economic growth, a relatively small number of people would be newly eligible for program benefits. They are among the people originally targeted to receive benefits. Limits on deductions for certain assets that have remained unchanged since 1989 also should be updated. For example, asset rules allow a deduction of $1,500 for life insurance policies. Older people typically have life insurance so that money will be available to pay for funeral and burial costs. In 1999, a basic adult funeral cost an average of $5,020. Burial costs an additional $2,000 or more.16 The $4,500 deduction allowed for vehicles is also outdated. The treatment of deferred compensation retirement savings also should be examined. Current rules favor defined benefit rather than defined contribution plans. Only a small proportion of people

The Commonwealth Fund

with low incomes have retirement plans, but for those who do the difference in how the types of plans are treated may affect whether they qualify for program benefits. The bias toward defined benefit plans could be eliminated if the savings in defined contribution plans were not counted as assets. Such a change would respond to changes in the broader market and could encourage modest saving for retirement. A number of states have adjusted the asset test for their Medicare Savings Programs. Some states exclude a certain amount of assets when determining eligibility, effectively increasing the asset limits. For example, Florida excludes the first $1,000 for each person. Maine excludes the first $8,000 for an individual and $12,000 for couples. Minnesota excludes the first $10,000 for individuals and $18,000 for couples. Some states disregard the value of one or more vehicles.These include Florida, Georgia, Kansas, Maine, Missouri, South Carolina, and Vermont. Higher values for life insurance are excluded in Florida ($2,500), Georgia ($5,000), and South Carolina ($5,000). Louisiana officials note that a new policy which allows a $10,000 exclusion for life insurance policies has simplified the application process for applicants and eligibility workers, thereby reducing the time required to process applications. Simplifying Renewals Periodic reviews are conducted for many public programs to ensure that participants continue to meet financial eligibility requirements. In most instances reviews are conducted annually for the Medicare Savings Programs. However, the data indicate that income and asset levels are unlikely to change among the low-income elderly population. The data also show that people who qualify for program benefits in one year, but not in subsequent years, do not have increased income or assets that exceed financial eligibility limits by large amounts. People whose benefits are discontinued may become eligible again and reapply within a short period of time if they face unexpected medical expenses or other costs.

Given these findings, it may be prudent to simplify the renewal process for the older lowincome population. For example, instead of requiring participants to reapply for benefits, states can ask participants to sign a postcard or form stating that their financial circumstances have not changed and they wish to continue receiving benefits. States are currently required to use Income and Eligibility Verification Systems (IEVS) to confirm information about applicants incomes. Thus, a method already exists to ensure that enrollment errors do not occur at renewal. A few states use an ex parte renewal process in which Medicaid program staff use electronic data systems to verify that renewal should occur. For example, data regarding Food Stamp program participation can provide current information about income and assets. Automatic renewals help program participants by eliminating some of the barriers often associated with the renewal process, such as the need to complete application forms, provide supporting documents, or visit the Medicaid office. In addition, automatic renewalwhich is associated with continuous enrollmentmay increase the likelihood that participants will receive uninterrupted health care services. Given that so many participants remain eligible from year to year, longer eligibility periods also may be warranted. Some state-funded pharmacy assistance programs already use longer time frames. Participants in South Carolinas SilverCard Program must reapply for benefits every two years. Pennsylvanias Pharmaceutical Contract for the Elderly (PACE) Program and New Jerseys Pharmaceutical Assistance to the Aged and Disabled (PAAD) Program require that higher-income enrollees reapply every year, but lower-income enrollees have a two-year eligibility period.17 Finally, a simpler renewal process can reduce administrative costs. If eligibility workers do not have to review documents or contact enrollees to obtain missing information, they will spend considerably less time on each renewal. And regardless of the type of renewal process used, the cost of

How Asset Tests Block Low-Income Medicare Beneficiaries from Needed Benefits

conducting renewals will be cut in half if they occur every two years instead of every year. Reducing or Eliminating the Use of Verification Documents Reviews of documents to verify income and assets are common during the enrollment and renewal processes for Medicaid and the Medicare Savings Programs. However, the findings regarding the financial circumstances of the low-income elderly indicate that these reviews may not be necessary. Applicants sign documents, under penalty of perjury, stating that the information they provide is correct. States are required to use Income and Eligibility Verification Systems to confirm that income information provided by applicants is correct. This verification occurs regardless of whether applicants are asked to provide documents or whether they make self-declarations about their assets or incomes. Case studies of Medicare Savings Programs in five states indicate that states that use collateral verification systems do not report increases in errors or fraud.18 And a letter from the Centers for Medicare and Medicaid Services reports a direct link in one state between extensive renewal requirements and a significant number of denials and terminations of applicants who did not return verification information but were otherwise eligible.19 Under federal law, applicants for Medicaid must only show documents verifying immigration status. Some states have already reduced requirements for verification documents. Income verification documents are not required for the Medicare Savings Programs upon initial application in 12 states and at renewal in 11 states. There are no requirements for documents to verify assets at the initial application in 17 states and renewal in 16 states.20 Requiring fewer or no verification documents also may reduce administrative costs. There is some evidence of this from changes to Medicaid programs for children and families. Officials in Michigan report that after they eliminated verification documents in favor of self-declaration of income for the childrens Medicaid program and

the State Childrens Health Insurance Program (CHIP), each caseworker processed an average of four more applications daily. At the same time, audits of reported income showed that self-declaration did not lead to higher error rates.21 A study of options to simplify the enrollment process for families receiving benefits from Californias MediCal program indicates that allowing self-certification could result in savings because the administrative savings would be greater than the costs of new enrollments.22 SUMMARY AND CONCLUSIONS Income is generally a good predictor of the value of assets for older people with low incomes. Most people with incomes that meet eligibility standards for Medicaid and the Medicare Savings Programs also meet eligibility standards for assets. These findings suggest that it may be unnecessary to use asset tests to determine financial eligibility for these programs. If asset tests are used, they should be adjusted to ensure that people who need benefits receive them. In particular, the limits and the allowable deductions for particular assets should be adjusted to reflect economic growth. Rules regarding retirement plans also should be revised so that all types of deferred compensation retirement plans are treated similarly. The incomes and assets of older people with low incomes do not change substantially over time. This suggests that the benefits renewal process can be simplified and that longer eligibility periods can be used. In addition, the close relationship between income and assets and the fact that financial circumstances are unlikely to change significantly over time for this population suggests that requirements for documents verifying financial information can be reduced or eliminated. A number of states have already simplified the enrollment and renewal processes for Medicaid and the Medicare Savings Programs. Such policies have the potential to ease the enrollment process for applicants, ensure that beneficiaries stay enrolled, and reduce administrative costs for Medicaid programs.

10

The Commonwealth Fund

NOTES
1

Under current law, states may opt to disregard all assets, effectively eliminating asset tests. Stephanie Maxwell, Marilyn Moon, and Misha Segal (2001). Growth in Medicare and Out-of-Pocket Spending: Impact on Vulnerable Beneficiaries. The Commonwealth Fund. National Association of State Medicaid Directors. Aged, Blind, and Disabled Medicaid Eligibility Survey. Available at http://www.nasmd.org/eligibility/ results7.asp (accessed November 3, 2003). Also, personal communication with members of the New York Medicare Savings Program Coalition, October 2003. Kim Glaun (2002). Medicaid Programs to Assist LowIncome Medicare Beneficiaries: Medicare Savings Programs Case Study Findings. Kaiser Commission on Medicaid and the Uninsured. Vernon Smith, Eileen Ellis, and Christina Chang (2001). Eliminating the Medicaid Asset Test for Families: A Review of State Experiences. Kaiser Commission on Medicaid and the Uninsured. Smith, Ellis, and Chang (2001). Center on Budget and Policy Priorities. Start Healthy Stay Healthy Campaign: State by State Tables. Available at http://ww.cbpp.org/shsh/tables.htm (accessed November 3, 2003). AARP. Grief and LossFuneral Arrangements and Memorial Services. Available at http://www.aarp.org/ griefandloss/articles/73_a.html (accessed November 3, 2003). Stephen Crystal, Thomas Trail, Kimberley Fox, and Joel Cantor (2003). Enrolling Eligible Persons in Pharmacy Assistance Programs: How States Do It. The Commonwealth Fund. Glaun (2002). Timothy W. Westmoreland (2000). Letter to State Quality Control Directors from Timothy W. Westmoreland, Director, Center for Medicaid and State Operations. Available at http://ww.cms.hhs.gov/states/letters/ smd91200.asp (accessed July 2, 2002). Laura Summer and Emily Ihara (forthcoming). Simplifying Medicaid Enrollment for the Elderly and Individuals with Disabilities. AARP.

Mark Nadel, Lisa Alecxih, Rene Parent, and James Sears (2000). Medicare Premium Buy-in Programs: Results of SSA Demonstration Projects. Social Security Bulletin, vol. 63, no. 3. Michael Perry, Susan Kannel, and Adrianne Dulio (2002). Barriers to Medicaid Enrollment for Low-Income Seniors: Focus Group Findings. The Kaiser Commission on Medicaid and the Uninsured. Jennifer Stuber, Kathleen Maloy, Sara Rosenbaum, and Karen Jones (2000). Beyond Stigma: What Barriers Actually Affect the Decisions of Low-Income Families to Enroll in Medicaid? Center for Health Services Research and Policy, George Washington University. Perry, Kannel, and Dulio (2002). It should be noted that estimates of the number of people who qualify for programs are conservative because they are based on standard income and asset eligibility rules. Substantial numbers of beneficiaries live in states that use more liberal income and asset rules. For example, calculations by the Center on an Aging Society based on data from the Henry J. Kaiser Family Foundation found that about 42 percent of Medicare beneficiaries live in states that have increased the income limits for full Medicaid coverage for individuals age 65 and older to at least 100 percent of the federal poverty level. Also, a number of states provide QMB, SLMB, and QI benefits to people with assets above the standard limits. See Laura Summer and Robert Friedland (2002). The Role of the Asset Test in Targeting Benefits for Medicare Savings Programs. The Commonwealth Fund. Marilyn Moon, Robert Friedland, and Lee Shirey (2002). Medicare Beneficiaries and Their Assets: Implications for Low-Income Programs. Henry J. Kaiser Family Foundation. Calculations by the Center on an Aging Society based on 2003 GDP in the third quarter. As noted earlier, changes in survey methods over time may affect the number of people who appear to be eligible each year, with increases in numbers and values for particular assets reported in subsequent years. Thus, the proportion of people who remained eligible in 2000 may be somewhat higher than is reported here.

10

11

12

13

14

15

16

17

18

19

20

How Asset Tests Block Low-Income Medicare Beneficiaries from Needed Benefits
21

11

Laura Cox (2001). Allowing Families to Self-Report Income: A Promising Strategy for Simplifying Enrollment in Childrens Health Coverage Programs. Center on Budget and Policy Priorities.

22

Lisa Chimento, Moira Forbes, Joel Menges, Anna Theisen, and Nalini Pande (2003). Simplifying MediCal Enrollment: Options for the Assets Test. Medi-Cal Policy Institute.

DATA AND METHODS

Analyses of the Number of People Eligible for Programs A nationally representative survey was used to examine the incomes and asset holdings of elderly Medicare beneficiaries. The Health and Retirement Study (HRS) contains detailed information on incomes and assets for 19,580 people representing the population over age 50 in 2000. It also includes information about the spouses of respondents, regardless of the spouses age. HRS data were used to calculate countable income and assets for individuals age 65 and older and their spouses. Program rules about counting spousal assets are complex. For this analysis, the assumption is that all assets are jointly held. Countable income was calculated using both earned and unearned income. Sources of income included in countable income are earnings, veterans benefits, Social Security benefits, pensions, unemployment compensation, workers compensation, annuity income, IRA withdrawals, alimony, lump sum payments, and income from assets such as rental property, a business or farm, stocks, and bank accounts, among others. Adjustments were made for sources of income that are excluded when determining eligibility.These include the first $20 of any monthly income, the first $65 of monthly earned income, and half of the remaining earnings. Countable assets include real estate other than the main home, vehicles, life insurance, IRAs or Keoghs, stocks or mutual funds, bonds, amounts in checking or savings accounts or money market funds, CDs or treasury bills, trusts, and other assets. Exclusions from countable assets include the value of one automobile up to $4,500, household goods and property, burial funds up to $1,500, and the cash surrender value of a life insurance policy up to $1,500. Longitudinal Analyses Two nationally representative surveys were used to conduct the longitudinal analyses.The Study of Assets and Health Dynamics Among the Oldest Old (AHEAD) contains detailed information on incomes and assets for 8,222 people representing the population age 70 and older in 1993. Information about the same population was collected for the 1995 AHEAD survey. The AHEAD and HRS surveys were combined in 1998. Income and asset information for the population included in the 1993 AHEAD survey is available from the 1998 and 2000 HRS surveys. Examining longitudinal data on older persons raises the issue of attrition, particularly due to death. Data from four years or waves1993, 1995, 1998, and 2000were used in the analyses. Only people who were alive in all four waves and participated in each wave of data collection were included in the population studied. Data were first collected from a representative sample of people age 70 and older in 1993, also known as the baseline year. This cohort of older persons was then re-interviewed in 1995, 1998, and 2000. Differences in data collection methods over the studys period may affect the number of people who appear to be eligible each year. Specifically, the later surveys asked more detailed questions about the types and amounts of assets people possessed. Thus, there may be some under-reporting regarding countable assets in 1993 relative to subsequent years.

ABOUT THE AUTHORS Laura Summer, M.P.H., is a senior researcher at Georgetown Universitys Health Policy Institute, where she conducts research on the design and operation of publicly-financed health care programs and on access to health and long-term care services. Ms. Summer also serves as the deputy director for the Institutes Center on an Aging Society, an organization that studies the impact of demographic changes on public and private institutions and families of all ages. She earned her master of public health degree from the University of Michigan. Ms. Summer can be reached at [email protected]. Lee Thompson, M.S., is a research analyst at the Health Assistance Partnership, part of Families USA. At the time this brief was prepared, she was a research associate at the Center on an Aging Society, located at the Health Policy Institute at Georgetown University. While at the Center, her research focused primarily on issues related to chronic conditions and long-term care financing. Thompson received a master of science in community development and applied economics from the University of Vermont in 1998. ACKNOWLEDGMENTS The authors would like to thank Robert Friedland of the Center on an Aging Society for his review of this paper. Thanks also to Barbara Cooper at The Commonwealth Fund for her helpful suggestions and support.

The Commonwealth Fund is a private foundation supporting independent research on health and social issues. The views presented here are those of the authors and should not be attributed to The Commonwealth Fund or its directors, officers, or staff.