LTC Bullet:  GAO on TOA Underwhelms

Wednesday, October 5, 2005 

Washington, DC-- 

LTC Comment:  The Government Accountability Office's new report on Medicaid asset transfers asks the wrong questions, uses the wrong data, and so provides few helpful answers.  More after the ***news.*** 


LTC Comment:  The Government Accountability Office's "letter report" titled "Medicaid:  Transfers of Assets by Elderly Individuals to Obtain Long-Term Care Coverage," is now available to the public at  For highlights only, see  

This report disappoints in many ways.  It asked the wrong questions.  It used the wrong methods.  It searched the wrong data.  And, consequently, it provides little new information of value. 

The important thing to remember about asset transfers to qualify for Medicaid is that they are only a small part of the Medicaid eligibility problem.  Most people qualify for Medicaid nursing home eligibility without manipulating their income or assets and without spending down significantly.  Medicaid has no hard limit on income (if medical expenses are high enough) and no hard limit on assets (if they are held in exempt form).  (For details on Medicaid nursing home eligibility, see "Aging America's Achilles' Heel:  Medicaid Long-Term Care" at   

For people with more than average income and assets, dozens of techniques that do not depend on transferring assets are elaborated in hundreds of books and articles and advertised on thousands of elder law websites.  Such non-asset-transfer methods are easily used to qualify for Medicaid LTC benefits without spending down.  Hence, the vast majority of people who seek to have the government pay for their long-term care need not bother about transferring assets.  Unfortunately, "asset transfers" became the buzz phrase encapsulating the much bigger problem of wide-open Medicaid eligibility exacerbated by widespread Medicaid planning.  In truth, only a small part of Medicaid planning involves actually transferring assets. 

Furthermore, only people with substantial assets, especially real property assets, are likely to transfer significant assets to qualify for Medicaid.  They are most likely to transfer real estate.  And, they are highly likely to make such transfers at least three years in advance of applying for Medicaid to avoid the "look-back" window.  GAO's approach to the transfer of assets question missed on each of these counts.  They reviewed a sample of the whole elderly population instead of focusing on Medicaid nursing home recipients likely to have transferred assets.  They looked only at cash transfers and ignored real property transfers.  And they made no effort whatsoever to examine asset transfers outside the three-year lookback, which is the critical issue in deciding whether or not Congress should lengthen the lookback period beyond three to five years. 

In summary, by asking the wrong questions, using the wrong methods, and examining the wrong data, GAO missed a wonderful opportunity to shed light on the transfer of assets problem. 

What should GAO have done? 

First, analyze the problem correctly and ask the right questions.  Eighty-two percent of people over the age of 62 own homes and 74 percent of these own their homes free and clear.  These older homeowners have nearly $2 trillion in home equity, close to half of which could be unlocked to pay for long-term care by means of home equity conversion according to the National Council on the Aging.  Yet, GAO's own 1989 study of Medicaid estate recoveries ( concluded that only 14 percent of people in nursing homes on Medicaid still own their homes.  There is no evidence that people are selling their homes or otherwise using their home equity for long-term care.  So what happened to all that home equity?  Was it transferred?  If so, was it transferred for purposes of qualifying for Medicaid or for other reasons that resulted in Medicaid eligibility anyway?  When did such transfers occur, within or outside the three-year (general transfers) or five-year (transfers to trusts) look-back windows?  If a house was transferred, were cash, securities or other assets transferred at the same time?  Those are the right questions to ask about Medicaid asset transfers. 

Second, make use of a method and review data that can answer the right questions.  Analyzing sample data on the elderly population in general as GAO did is useless.  Obviously people in nursing homes on Medicaid or on the cusp of institutionalization based on severe disability have relatively few assets and low incomes.  That's a methodological tautology.  And examining the wealth of the elderly population in general is meaningless.  The vast majority of such people qualify for Medicaid nursing home care based on the program's generous eligibility rules.  To find out how widespread the practice of Medicaid asset transfers actually is will require original research to examine a valid random sample of current Medicaid nursing home cases.  Submit such a sample to a comprehensive review including detailed interviews with the Medicaid recipients and/or their personal representatives, thoroughly study the Medicaid application and case file, pursue all leads regarding income, asset or property ownership, and examine county recorders' and assessors' records to identify real property ownership and transfers both in the current county of residence and in counties of prior residence.  A review of this kind will solidly identify how significant the Medicaid transfer of assets problem is.  Such a review is beyond the scope of routine eligibility determination, but it will provide a baseline with which to compare and from which to improve the eligibility process.  More importantly, it will indicate clearly how far back the Medicaid transfer of assets lookback period must go to insure that most personal wealth remains available for long-term care costs.  Answering that question is the key to saving Medicaid from financial ruin. 

Despite the foregoing criticism, the GAO report was not a total waste of effort.  Following are some passages of interest with page citations and our comments when necessary. 


"None of the nine states GAO contacted tracked or analyzed data on asset transfers or penalties applied. . . .  These states generally relied on applicants' self-reporting of transfers of assets, and officials from these states informed GAO that transfers not reported were difficult to identify."  (Highlights)    


"Because certain types of long-term care-particularly nursing home care-are costly (estimated by some to average over $70,000 a year for a private-pay patient), individuals who pay for an extended stay in a nursing home can quickly deplete their assets and subsequently qualify for Medicaid." (p. 7)   

LTC Comment:  This trite observation that people "can" quickly spend down for nursing home care begs the question "Do they?"  Answer:  there is no evidence of widespread nursing home spend down.  All but 10 to 15 percent of total nursing home costs are accounted for by direct or indirect government financing and personal income, not assets. 


"[W]e analyzed data from the 2002 Health and Retirement Study (HRS), a longitudinal national panel survey, sponsored by the National Institute on Aging and conducted every 2 years by the University of Michigan.  . . .  Our analysis includes data from all elderly households; we did not assess whether the respondents had or were seeking Medicaid coverage.  Because the HRS only addressed cash transfers-cash provided to relatives or other individuals-our analysis understates the extent and amount of all transfers by excluding transfers of property and other types of assets.  Since the HRS did not inquire about the reason for the cash transfers, no conclusions can be drawn regarding the extent to which the survey respondents transferred cash for purposes of establishing Medicaid eligibility for long-term care."  (p. 3)   

LTC Comment:  Quod erat demonstrandum


"Although most of the officials in the nine states we reviewed reported that some individuals transferred assets for purposes of qualifying for Medicaid coverage for long-term care, none of these systematically tracked or analyzed data that would provide information on the incidence of asset transfers and the extent to which penalties were applied in their states." (p. 5) 


"Overall, severely disabled elderly households-those reporting three or more limitations in ADLs-were less likely to transfer cash than nondisabled elderly households." (p. 13)  

LTC Comment:  That's no surprise.  What did they own and how much did they transfer before they came to need help and did they manipulate their wealth in contemplation of Medicaid eligibility?  Those are the relevant issues. 


"According to data from the 2002 HRS, total income for the nation's approximately 28 million elderly households was $1.1 trillion and total nonhousing resources were $6.6 trillion."  (p. 13) 

"About half of all elderly households had nonhousing resources of $50,000 or less, while almost 20 percent had nonhousing resources greater than $300,000."  (p. 14) 

LTC Comment:  So, GAO should have focused on understanding the 20 percent over $300,000.  Pareto said you get 80 percent of your results from 20 percent of your efforts.  Businesses have to prioritize to survive; government rarely does.  GAO didn't in this case. 


"Disabled elderly households—which are at higher risk of needing long-term care—had lower levels of assets than nondisabled elderly households. Generally, as the level of disability increased, the level of assets decreased.  Severely disabled elderly households, which made up about 6 percent of total elderly households, had significantly lower median income ($13,200) and median nonhousing resources ($3,200) compared with all elderly households ($24,200 and $51,500, respectively)."  (p. 15) 

LTC Comment:  Possible explanation, people jettison their wealth the closer they get to the nursing home door. 


"For example, giving away assets as a gift generally results in the imposition of a penalty period, but giving away assets valued at less than the average monthly private-pay rate for nursing home care may not, depending, in part, on whether the state imposes partial-month penalties."  (p. 20) 

"Because the penalty period begins at approximately the date of asset transfer, individuals that meet Medicaid income eligibility requirements can give away about half of their resources and use their remaining resources to pay privately for long-term care during which time any penalty period would expire.  This is often referred to as the 'half a loaf' strategy because it preserves at least half of the individual’s resources.  (pps.  22-23) 

LTC Comment:  GAO's discussion of the partial-month penalty, half-a-loaf strategy, and penalty date issues was weak.  We hope to tackle those issues in a policy analysis paper in the future. 


"Elder law attorneys explained that once individuals are Medicaid-eligible, they and their families will have limited means.  Therefore, they [the Medicaid planners] advise these individuals to update, renovate, repair, or replace old or deteriorating items such as homes and cars to reduce the need for maintenance and repairs in the future.  No penalty is associated with paying a debt or making a purchase as long as the individual receives something of roughly the same value in return."  (p. 21) 

LTC Comment:  Medicaid planners also maintain comprehensive lists of items people can buy to convert countable assets into non-countable property.  GAO was clearly captured by the Medicaid planners they interviewed whose sanctimonious altruism masks egregious self-dealing. 


"In addition, if the community spouse’s income is less than the state’s minimum needs allowance, the state can choose to make up the shortfall by (1) transferring income from the institutionalized spouse or (2) allowing the community spouse to keep resources above the resource allowance so that the additional funds can be invested to generate more income.  Under the latter approach, the community spouse may be able to retain a significant amount of resources in order to yield the allowable amount of income.  For example, a community spouse might ask to retain a savings account with $300,000 and an annual interest rate of 2 percent that would yield an additional $500 in income per month."  (p. 21) 

LTC Comment:  This commonplace practice allows people to retain hundreds of thousands of dollars over the community spouse resource allowance intended by Congress to be no more than $95,100. 


"None of the nine states we reviewed systematically tracked or analyzed data that would provide information on the incidence of asset transfers and the extent to which penalties were applied in their states. Nationwide, all states requested information about applicants’ assets, including transfers of assets, through Medicaid application forms, interviews to determine Medicaid eligibility, or both. The nine states we reviewed generally relied on applicants’ self-reporting of financial information and varied in the amount of documentation they required and in the extent to which they verified the assets reported. According to officials in these states, transfers that were not reported by applicants were difficult to identify." (p. 25) 

"Although officials from the nine states reviewed reported that some individuals transferred assets for purposes of qualifying for Medicaid, these states did not systematically track and analyze data on the incidence of asset transfers or associated penalties.  As a result, the states could not quantify the number of people who transferred assets, the assets transferred, or the penalties applied as a result of transfers for less than fair market value." (p. 26) 


"CMS noted that the Medicaid program will only be sustainable if its resources are not drained to provide health care assistance to those with substantial ability to contribute to the costs of their own care.  CMS acknowledged, however, the difficulty of gathering data on the extent and cost of asset transfers to the Medicaid program.  In particular, CMS commented that the law is complex and that the techniques individuals and attorneys devise to divest assets are ever-changing.  CMS reiterated the President’s budget proposal to tighten existing rules related to asset transfers, and associated estimated savings, which we had noted in the draft report.  CMS further noted one limitation to our analysis that we had disclosed in the draft report—that the HRS only addressed cash transfers provided to relatives or other individuals.  CMS commented that it believes that substantial amounts of assets are sheltered by individuals who transfer homes, stocks and bonds, and other noncash property.  We agree with CMS’s view that information on such noncash transfers would be valuable, but as we noted in the draft report the HRS does not include such data.  (p. 34) 

LTC Comment:  Criticism of the GAO report by the Centers for Medicare and Medicaid Services (CMS) pretty much hit the nail on the head.