| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||
Correspondence: Address correspondence to Jinkook Lee, Department of Consumer Sciences, Ohio State University, 1787 Neil Ave., Columbus, OH 43210. E-mail: lee.42{at}osu.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Medicaid • Medicaid estate planning • Family wealth transfer • Private transfer • Crowding out
Recently, academics and policy makers have suggested that some elders, anticipating nursing home entry, may transfer their assets to their children or other relatives in order to qualify for Medicaid (Bassett, 2004; Bassett & Lumsdaine, 2001; Curry, Gruman & Robison, 2001). The practice of legal and financial planning to satisfy the financial eligibility requirement for Medicaid coverage is called Medicaid estate planning. To contribute to policy making on the issue of Medicaid estate planning, we empirically examine the wealth-transfer behavior of new Medicaid recipients among elders.
Specifically, using all the available waves of the longitudinal Health and Retirement Study, we tracked individual elders who were not Medicaid recipients at the baseline interview in 1993 but became Medicaid recipients during the follow-up interviews. We then tracked the occurrence and amount of wealth transfers before they became Medicaid recipients, with a focus specifically on new recipients residing in nursing homes. Finally, we discuss the implications for Medicaid financing of nursing home care.
Medicaid Eligibility
As a means-tested entitlement program, Medicaid requires an individual to meet financial eligibility criteria. There are two major pathways for older and disabled people to qualify for full Medicaid benefits. One is to be categorically needy, defined as people who receive Supplemental Security Income (SSI). The other group consists of medically needy people who are not strictly poor enough to be Medicaid eligible but whose extensive health care expenditures are counted against their income; this standard is often used by elders residing in nursing home facilities and those with high prescription-drug costs (Kaiser Commission on Medicaid and the Uninsured, 2003). The eligibility criteria differ according to the pathways. The income limits range from 27% to 120% of the federal poverty line for noninstitutionalized individuals. For institutionalized elders, income limits are up to 300% of the SSI benefit. The assets limits excluding the primary residence range from $2,000 to $10,000 (O'Brien & Elias, 2004; Yelowitz, 2000).
Federal law mandates categorical coverage under these eligibility rules but also gives states broad optional authority to extend Medicaid eligibility to the medically needy population and beyond minimum standards. Such flexibility has created state-to-state variations in who is and how many are covered by Medicaid (AARP, 2000; Bruen, Wiener, & Thomas, 2003; Kaiser Commission on Medicaid and the Uninsured, 2003). Income limits vary across the states, ranging from $100 to $716 for individuals and from $217 to $968 for couples living in communities. Income and assets limits are more generous, moreover, when one spouse of a couple is in a nursing facility and the other remains at home. The spousal impoverishment provision of the Medicare Catastrophic Coverage Act (MCCA) of 1988 requires states to permit spouses at home to retain a much larger amount of assets and income (Sloan & Shayne, 1993). In 2001, income limits for the spouse ranged from $1,406 to $2,175 per month, and asset limits ranged from $17,856 to $89,280 (adjusted every year for inflation; Kaiser Commission on Medicaid and the Uninsured, 2003).
Medicaid Estate Planning
Hubbard and colleagues (1995) argued that social programs with means tests based on assets have a crowding-out effect. Social programs provide a safety net for those without assets. Arguably, when society provides such as safety net, individuals are less willing to save, knowing that they can rely on the program to meet their needs. Thus by providing health insurance for medically needy people without assets through the program, individuals are discouraged from accumulating private savings or purchasing private insurance policies. Some researchers view Medicaid estate planning as a special case of such crowding-out effects: Elders who wish to protect their assets for their children and family members transfer their assets and become Medicaid recipients (Curry et al., 2001; Walker, Gruman, & Robison, 1999).
Federal Medicaid law attempts to discourage Medicaid estate planning by imposing the following rule: If an elderly individual or the individual's spouse disposes of personal resources, including housing, for less than fair market value within 36 months of applying for Medicaid, then the individual is subject to a period of exclusion from coverage for nursing facility services or home- and community-based care. The 36-month "look-back" period is extended to 60 months in the case of transfers into a trust. The period of exclusion from Medicaid coverage is based on the amount of resources transferred, the average monthly cost of nursing facility care in the state, and the date on which the transfer was made. Because the exclusionary period runs from the date of transfer, it is possible that it will already have expired by the time an elder applies for Medicaid. For example, assume that 12 months before applying for Medicaid, an elderly individual transferred $25,000 in savings to her children and that, on average, the monthly cost for a non-Medicaid resident in a nursing facility is under $5,000 per month. Under the statutory formula, the amount transferred is divided by the average monthly cost to yield a number that represents the number of months of exclusion from coverage. In this case, she would be excluded for 5 months ($25,000 divided by $5,000). However, because the exclusion begins to run from the date of the transfer, and because in this case the transfer occurred 12 months before application, there would be no exclusion from coverage in this case (Kaiser Commission on Medicaid and the Uninsured, 2002).
To date, only a few empirical studies have examined Medicaid estate planning, and conflicting results have been found. Further, most discussions are based on anecdotal evidence (Stum, 2001), the review of case files in a few states such as Massachusetts and Minnesota (General Accounting Office [GAO], 1993; Minnesota Department of Human Services, 1996), or interviews with state workers, attorneys, and financial planners (Burwell & Crown, 1995; Walker et al., 1999).
Some researchers argue that Medicaid estate planning is not prevalent. For example, Moffitt (1983) noted that, in a phenomenon known as welfare stigma, people tend to be averse to welfare programs despite their positive potential benefits. Norton (1995) found that some elders in nursing homes receive transfers from their children or even sell housing assets in order to avoid Medicaid eligibility. Sloan and Shayne (1993) suggested that transferring assets in anticipation of being institutionalized is not widespread and happens infrequently. Using the 4,600 disabled elders, age 65 or older, who reside in the community in the 1989 National Long Term Care Surveys (NLTCS), they conducted a simulation analysis of the extent of Medicaid eligibility. The results showed that most were either already financially eligible or would have been so immediately upon admission to a nursing home; in the simulation, about 19% were actually on Medicaid, 59% would have qualified for Medicaid immediately upon institutionalization, and another 6% would be eligible in less than 6 months. Elders eligible within 6 months of nursing home entry would seem to have too little wealth to warrant hiring an attorney to arrange an asset transfer. Thus, Sloan and Shayne concluded that only a small percentage of the disabled elders would have an incentive to transfer assets.
Wiener (1996) showed direct and indirect evidence that the number of people transferring assets for Medicaid eligibility is much smaller than is commonly thought. The only direct evidence was from a GAO (1993) study of applicants for Medicaid nursing home care in Massachusetts. Of the 403 Massachusetts Medicaid applicants reviewed, 49 applicants had transferred assets. Wiener also argued that based on data from the noninstitutionalized elderly population in the Survey of Income and Program Participation (SIPP), few elders have sufficient assets to make a transfer worthwhile. At the time of their admission to the nursing home, 75% of the SIPP sample had less than $50,000 in nonhousing assets, and almost half had less than $10,000.
Other researchers have offered estimates of the prevalence of Medicaid estate planning that vary greatly, ranging from 5% to 54% of Medicaid beneficiaries. Burwell and Crown (1995) estimated that 5% to 10% of unmarried applicants and 20% to 25% of married applicants transfer wealth to qualify for Medicaid. The estimates of the Minnesota Department of Human Services (1996) and GAO (1993) are 22% and 54%, respectively. Walker and colleagues (1999) also estimated the rate of Medicaid estate planning at 25% to 50%.
The amount of assets transferred also varies across studies. For example, GAO (1993) estimated that the average amount of assets converted from countable to nonaccountable was $5,618, and only 1.5% of applicants (6 out of 403) transferred more than $100,000. Transferred assets in the GAO study thus generally are not sufficient to pay for one year or even one month of nursing home coverage. On the other hand, Walker and colleagues (1999) reported that the average value of transferred assets exceeds $50,000. These studies were all based on review of a small number of case files in few states and anecdotal interviews with state workers, attorneys, and financial planners.
Recently, Bassett (2004) investigated the impact of the self-assessed probability of entering a nursing home on asset transfer. Using the first wave of Asset and Health Dynamics Among the Oldest Old (AHEAD) data, he found an association between a positive assessment and asset transfer but concluded that Medicaid-induced asset transfers are fairly small. He estimated that Medicaid-induced asset transfer is approximately $1 billion per year, about 3% of Medicaid nursing home expenditures in 1993. However, he could not provide evidence of the percentage of this estimated amount that resulted from Medicaid estate planning. On the other hand, Moses (2005) proposed that replacing the home-equity exemption with reverse mortgages could save the program $20 billion per year, based on the high rate of home ownership among elders and disabled in the Medicaid program.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Only the original AHEAD sample (6,047) was included in the current study, because those aged 70 or older are the main consumers of nursing home care, and the original AHEAD cohort allowed us to investigate changes in Medicaid eligibility status during a 10-year period, beginning after respondents reached age 70.
Measures
Medicaid
For each wave of interview, we recorded a binary variable of whether or not a respondent received Medicaid benefits at the time of the interview. Even though Medicaid eligibility is granted at the individual level, the resource criteria for determining the Medicaid eligibility are applied at the individual or couple level. Thus, we conducted analysis at either the individual or couple level, examining whether any elderly individual in the household was a beneficiary of Medicaid.
Wealth Transfer
Wealth transfer was the dollar amount of wealth that elders transferred during the 2-year period preceding the interview. Specifically, the HRS asked the following question: "Did you (or your husband/wife/partner) give financial help totaling $500 or more in the last two years (since the previous interview) to any of your children (or grandchildren) not counting shared housing or shared food?" The HRS defined financial help as "giving money, helping pay bills, or covering specific types of costs such as those for medical care or insurance, schooling, down payment for a home, rent, etc." The financial help could be considered support, a gift, or a loan. The earlier waves of the HRS had some variations in the wording of the survey questions. Specifically, the 1993 surveys asked the respondents about wealth transfers of $500 or more during the past 12-month period, and in the 1995 surveys, the threshold was $100 or more during the past 12-month period. Such variations in the wording may bring about some measurement errors. Regarding the amount of wealth transfers, we converted all dollar figures into 2002 dollars, using the current methods version of the Consumer Price Index for all urban consumers. (See Appendix for the verbatim.)
|
As the Medicaid program is a major funding source for expensive nursing home care and the prospect of spending one's assets on this care presumably motivates wealth transfer, this study also tracked those who were not in a nursing home at the baseline interview but were institutionalized before the follow-up interviews. Once we identified new nursing home residents, the investigation looked at who received Medicaid benefits and who had made asset transfers to children and grandchildren.
We have reported the descriptive statistics for the occurrence and the amount of wealth transfers and for changes in Medicaid status. Because the AHEAD–HRS is a multistage probability sample of households, we weighted the data, accounting for clustering and stratification.
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
Among these new Medicaid recipients, a small, but nontrivial, percentage of elders transferred wealth before receiving Medicaid. Among the new recipients in 1998 (n = 235), 37 elders (15.7%) transferred wealth to children. Among couples, 32.1% of the new Medicaid recipients transferred wealth to children. Similarly, in 2000, 39 elders, or 19.7 % of the new Medicaid recipients, transferred wealth to family, and among couples, 41.7% of the new recipients transferred wealth. Finally, in 2002, 19 elders transferred wealth, representing 13.5% of all new Medicaid recipients, and 6 of them were married. Altogether, 154 out of 861 (17.9%) new Medicaid recipients transferred wealth during the 10-year period. The average amount of transferred wealth was $8,507 for all, $6,687 for singles, and $10,140 for couples. Almost all (except 1 respondent) of those who transferred wealth gave financial gifts to children only once during the 10-year period.
The characteristics of the 154 who transferred wealth and the 707 who did not transfer wealth are shown in Table 2, showing significant differences. Women recipients were more likely not to transfer, whereas people who were married, White, more affluent, and better educated were more likely. The mean and median incomes of those who transferred were $30,100 and $17,200, respectively; about twice the $13,700 and $9,200 for those who did not. Half of the former had less than $20,000 of financial wealth while half of the latter had only $1,000 of financial wealth. The HRS defined financial wealth as consisting of checking and savings accounts, money market accounts, certificates of deposit, stocks and bonds, and individual retirement accounts; these were the potential sources of wealth transfer in this study. These findings mean that even though both those who transferred assets and those who did not had limited financial wealth before becoming Medicaid recipients, the former had more transferable resources. They also had about $20,000 more median and mean housing wealth than elders who did not transfer assets.
|
|
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
The results show that 16.4% of those who were not Medicaid recipients in the base year became Medicaid beneficiaries during the 10-year period. Among these new Medicaid recipients, a small but nontrivial number, 17.9%, transferred wealth to children before becoming eligible. Wealth transfers to children were more common and larger among elders who were married, not because transfers were made between spouses—the data refer only to transfers to children and grandchildren. The findings further show that the amount of familial wealth transfers is modest: On average, new Medicaid recipients transferred $4,112 to family during the 2 years prior to becoming Medicaid eligible in a nursing home. In the most recent phase of the study, transfers were infrequent and modest enough that half the recipients transferred $1,000 or less. Given the high cost of nursing home care, these elders would likely have become Medicaid eligible during the first month of nursing home residency even without the transfer.
These findings imply that Medicaid estate planning is practiced by elders in some cases. Unlike previous studies, the data used in this study are representative of the U.S. population, providing evidence of Medicaid estate planning at the national level. However, this study is limited in the sense that it did not explicitly consider potential differences by state although Medicaid-eligibility criteria and benefit levels differ across states. Future studies should consider these differences and control for other contributing factors to the changes in eligibility status. Another limitation is that although we examined elders who were institutionalized subsequent to the baseline interview, the original samples include only community elders. Medicaid estate planning is usually an effort to protect assets before incurring the cost of nursing home care, but an unknown number of nursing home residents may attempt to transfer wealth after they have become private-pay patients. This may be true especially of elders who enter as short-stay (Medicare) residents but who discover that they are unable to return to the community. Third, these results may underestimate the prevalence or amount of wealth transfer for Medicaid estate planning in that we considered outright gifts of cash as the only form of asset transfer in this study. Medicaid recipients will tend to underreport wealth transfers that occurred prior to applying for Medicaid. In addition, elder law attorneys have devised a wide variety of sophisticated asset-sheltering instruments, including irrevocable annuities, life estates, and "spousal refusal" testaments in the practice of Medicaid estate planning. These instruments, in fact, make up of the bulk of Medicaid estate planning activity and are not captured by the AHEAD survey.
Policy Implications
The findings of this research are similar to those of most previous studies: Although asset transfer to secure Medicaid eligibility does occur among elderly people in the community and 9% to 15% of new Medicaid-eligible nursing home residents have in fact transferred assets, the mean amount transferred by institutionalized recipients was only about $4,000—less than the cost of one month of nursing home care in many markets. The policy implications are similar as well to those drawn by most previous researchers: Though it may be inequitable for more affluent elders to transfer assets to children in order to qualify for Medicaid nursing home benefits, the absolute dollar losses to the program—at least from this kind of estate planning—are small.
Two other findings relate to this general point. First, the occurrence and amount of wealth transfers by couples prior to Medicaid eligibility are greater than those by single elders—reaching more than 40% and a mean of nearly $15,000 in various waves. The cumulative mean transferred by singles is about two thirds of that transferred by couples, but married elders become Medicaid eligible in far fewer numbers, so the fiscal impact of the transfers is limited.
Secondly, asset transfers by community-based elders appear to be more common among those who do not enter a nursing home than among those who do. This is perhaps surprising, given that the extraordinary cost of nursing home care is thought to be the impetus for wealth transfer by middle-class elderly. It is possible that these community-based Medicaid-eligible elders transferred assets in anticipation of nursing home entry that did not occur (Bassett, 2004). In any event, these findings suggest an interesting and important question for further study.
Policy discourse is, at this writing, focused on limiting Medicaid expenditures by restricting wealth transfers to achieve eligibility for Medicaid nursing home benefits. The administration, Congress, and the National Governors Association all propose to reduce the Medicaid budget by making it harder for people to divest their wealth to family members prior to becoming Medicaid recipients. A policy to limit asset transfer is appealing to many for at least two reasons. First, there is surely some transfer of wealth by affluent elderly to secure Medicaid nursing home benefits, and second, this seems fundamentally unfair—a usurpation of public monies by elders who do not need them. This research, with its admitted limitation to self-reported transfers, however, finds what other studies have found—that wealth transfers are neither very common nor, on average, very large. Especially among new recipients in nursing homes, reducing asset transfer is at best a minor tactic in any strategy to control Medicaid costs.
There is, of course, something to be said for fairness. Affluent elders should not deplete Medicaid funds meant for their less affluent peers, especially when, as might be expected, more affluent elderly divest more resources more often. The current long-term-care system does not provide any supports for the elders with nursing home care needs who are not qualified for means-tested public benefits but cannot afford private insurance policies. From the perspective of the system as a whole, federal policies to further limit the transfer of wealth seem both unlikely to reduce Medicaid spending significantly and neglectful of the real crisis in long-term-care financing.
|
Footnotes |
|---|
2 Department of Family Studies, University of Kentucky, Lexington.
3 School of Public Health, Ohio State University, Columbus.
Decision Editor: Linda S. Noelker, PhD
Received for publication September 27, 2004. Accepted for publication September 21, 2005.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
| ||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
