LTC Bullet: The LTC Wars (shawsky)

Friday, February 24, 2017


LTC Comment: The self-styled conservative LTC Commission co-chair has declared war on government-financed long-term care proposing private sector solutions that mirror our own. Details follow.



LTC Comment: Mark Warshawsky co-chaired (with Bruce Chernoff of SCAN) the underfunded, short-lived and ultimately hapless Long-Term Care Commission, which was supposed to autopsy CLASS and replace it with something better. He has launched two ICBMs in a new assault on the latest government LTC financing takeover proposal.

We reported on the first missile last week in this “LTC Clipping”:

2/15/2017, “The Urban Institute Model of Financing Long-Term Services and Supports: A Critical Review,” by Mark Warshawsky, Health Affairs Blog

Quote: “Scoring in the LTSS area needs to be based on widely accepted facts (i.e., a fair reading of the literature and/or settled empirical findings), completely transparent (with all major assumptions disclosed and justified), and robust (to consider all types of policy interventions). Scoring should also be alternatively illustrated by conservative results and assumptions, to give a sense of the possible range of outcomes. Unfortunately, despite the great apparent effort of the Urban Institute, their model, as presently constituted, cannot serve this purpose.”

LTC Comment: This is a devastating refutation of the data and reasoning employed by advocates of a new, compulsory, payroll-financed government LTC program. See our earlier critique of the same work in “LTC Bullet: LTC at a Crossroads,” Friday, June 3, 2016.

If the theoretical and evidentiary underpinnings supporting a new government LTC program are defective, what would be a better intellectual foundation for a different, more market-based approach? That’s what Warshawsky offers in missile #2, “Improving the System of Financing Long-Term Services and Supports for Older Americans,” a working paper authored with Ross Marchand for their employer, the Mercatus Center at George Mason University.

That paper was the subject of last Wednesday’s Long-Term Care Discussion Group meeting. You can retrieve Warshawsky’s slides for that presentation as well as the paper itself here. The LTC Discussion Group dubs itself “an informal non-partisan networking group of long term care (LTC) policy stakeholders.” You can participate in most of their meetings by phone. It’s a great way to stay abreast of all matters related to LTC financing and service delivery. I’ve addressed the group eight times over the years and follow them closely from afar nowadays.

Back to The LTC Wars (shawsky). What follows are excerpts from Mark J. Warshawsky and Ross A. Marchand, “Improving the System of Financing Long-Term Services and Supports for Older Americans,” Mercatus working paper, Mercatus Center, George Mason University, Washington, DC, January 2017. We’ll comment on each excerpt.

Improving LTSS: Abstract: Medicaid currently pays for most of the long-term services and supports (LTSS) given to older Americans. With the aging of the population, these costs to state and federal governments will increase rapidly. We summarize the current Medicaid eligibility rules, which are commonly portrayed as allowing access only to low-income and low-asset populations, but in reality they allow covered households to own significant housing and retirement assets. Furthermore, we present new empirical information about the weak efforts of states in enforcing even the current porous rules. We then report on the extensive asset holdings, especially in housing and retirement assets, across the distribution of retired households. We find that liberal eligibility rules and uneven enforcement increase the costs of governments and discourage the retired households that can afford it from covering LTSS exposure through private insurance and assets. We conclude with targeted recommendations to reform Medicaid and improve the LTSS financing system, following up on some of the proposals made by members of the 2013 federal Commission on Long-Term Care.” (p. 2)

LTC Comment: Hear, hear! If you follow LTC Bullets and our many state and federal level reports, this should sound very familiar. Let’s look now at the details.

Improving LTSS: “Data are from the Health and Retirement Study (HRS 2012), the gold standard of data on the finances of older households, which is sponsored by the National Institute on Aging and is conducted by the University of Michigan, Ann Arbor.” (p. 3)

LTC Comment: Here we part company slightly. HRS and AHEAD, its companion survey focused on the oldest old, provide very dubious data on which to base conclusions about long-term care spending as we explained in “LTC Bullet: Behind AHEAD,” Friday, September 2, 2016. The HRS/AHEAD data are unreliable because respondents who provide them have conflicts of interest which invite deception and/or non-reporting. Besides, most assets covered in those surveys are exempt for Medicaid eligibility purposes, especially home equity and personal retirement accounts.

Improving LTSS: “In particular, the commission did not assess the extent of state efforts to recover assets (housing and other assets) from the estates of Medicaid recipients of LTSS. It did not estimate the asset holdings of older Americans from which financing of LTSS could be found. It also did not look carefully into the variability in state rules regarding the exclusion of retirement assets, such as individual retirement accounts (IRAs) and 401(k) accounts, from assets that count for Medicaid eligibility—the ‘millionaires on Medicaid’ problem.” (p. 8)

LTC Comment: Right on! That was the LTC Commission’s biggest single failing. What does this new paper tell us regarding how Medicaid overlooks assets that could fund LTC privately?

Improving LTSS: “As we will see, despite its reputation, Medicaid for LTSS is not just a program for the poor.” (p. 8)

LTC Comment: Hallelujah. It’s so good to hear someone else state the obvious truth almost uniformly evaded in most of the LTC financing literature.

Improving LTSS: “But some people become eligible for Medicaid because of their spending on LTSS: They ‘spend down’ to Medicaid eligibility by spending nearly all their income and some of their assets on services. Because nearly all income must be spent before Medicaid begins to pay, rules protect some income and assets for community-resident spouses. In addition, some assets are excluded, thus enabling a Medicaid recipient to retain assets of substantial value. For example, the value of the family home is protected during the lifetime of the Medicaid recipient and spouse.” (p. 9)

LTC Comment: Medicaid LTC eligibility rules require most income to be spent down either on health or LTC services. But the rules do not require assets to be spent down on care. It seems a minor point, but it is very important because much, not to say most, of the peer-reviewed literature gets it wrong, saying people must spend down assets for care services. The only actual requirement is that assets spent or divested must receive fair market value. So there is no restriction whatsoever on purchasing exempt assets, including a home, car, prepaid burials, etc., as a means to spend down to Medicaid eligibility.

Improving LTSS: “Medicaid allows the recipient to exclude from countable assets the value of the primary residence—up to $536,000 (indexed) in 2013, although states can allow up to $802,000 (indexed) in 2013—as well as a car, personal and household items, burial funds, term life insurance, and some or all qualified retirement assets in most states.” (p. 9)

LTC Comment: This quote is imprecise. Medicaid does not exclude home values, but rather home equity. A home worth $1 million would be exempt if the owner had a $.5 million mortgage on it. Also, why cite 2013 home equity exemption levels? The latest exemption, effective January 1, 2017, varies from $560,000 to $840,000 depending on the state.

Improving LTSS: “Despite varied efforts by the states to recover resources from estates of deceased Medicaid beneficiaries since Medicaid’s creation in 1965, there are no systematic data sources on these collections over the subsequent five-decade period. The US Department of Health and Human Services (HHS) published state-by-state estimates of estate collections for fiscal years 1985 and 1993, but these data preceded the Omnibus Reconciliation Act of 1993, which required states to attempt to recover resources from estates.” (p. 12)

LTC Comment: Actually, the USDHHS IG published a comprehensive report in 1988 titled “Medicaid Estate Recoveries” of which I was the author. Based on that study, we projected that if every state in the country collected from estates at the same rate as Oregon, which had the most successful estate recovery program at the time, total annual recoveries could have been $589.2 million instead of the 1985 actual recoveries of $41.7 million. Most of the recommendations in that report, including longer and stronger transfer of assets restrictions and mandatory estate recoveries, became federal law with the passage of the Omnibus Budget Reconciliation Act of 1993.

Improving LTSS: “Had all states met these [Idaho] ‘best efforts’ rates, which were in the range of 1 percent to 4 percent (hardly strenuous), nearly $20 billion more—$24 billion instead of $4.4 billion—could have been recovered for governments nationally from estates of deceased Medicaid LTSS beneficiaries in the 2002–2011 period.” (pps. 21-22)

LTC Comment: We certainly agree having said the same thing over two years ago in “LTC Bullet: IG Report Reveals Medicaid Estate Recovery Weakness,” Friday, December 5, 2014. To wit:

Nevertheless, actual and potential dollar recoveries are nothing to sneeze at: For example, what if every state in the country recovered at the same rate as the most successful state, Idaho? Total recoveries would have been $2,845,253,843 instead of $497,905,382 based on percentage of nursing home expenditures recovered and $2,941,856,963 instead of $497,905,382 based on percentage of total LTC costs recovered. That’s nearly $2.5 billion [per year] in money now allowed to pass unencumbered to heirs. Those lost funds have the effect of converting Medicaid from a long-term care safety net program for the needy to free inheritance insurance for prosperous baby boomers.

Improving LTSS: “Because a plurality of states counts both the applicant’s retirement assets and the spouse’s assets, and other states count at least some of the retirement assets of the relevant populations, the resulting weighted average is more than half (71.3 percent). However, stated another way, almost one-third of retirement assets are not counted toward Medicaid eligibility despite the policy intent that these tax-qualified assets be used not for bequest but for all types of spending in retirement. Moreover, most states have exemptions that allow seniors with access to well-funded retirement accounts to exclude said assets.” (p. 27) Consequently: ‘Vermont seniors with sizeable retirement assets can qualify for LTSS Medicaid at the expense of New Hampshire seniors a few miles away, who are barred from having those same retirement assets if they enroll in Medicaid.’” (p. 28)

LTC Comment: Because of the transition from mostly “defined benefit” to mostly “defined contribution” retirement savings plans, aging Americans now have trillions of dollars set aside in tax-deferred retirement accounts. The intent of Congress is clear that such assets “are uncounted as long as the [Medicaid] AR [applicant recipient] is receiving periodic interest and principal payments,” as we reported in “Medi-Cal Long-Term Care: Safety Net or Hammock?” (p. 21) based on Social Security Administration, POMS, “SI 01120.210 Retirement Funds,” In other words, SSI regulations do not allow counting tax-sheltered retirement assets as long as those assets are being tapped periodically, as tax law requires beginning by age 70 and ½. Warshawsky and Marchand make an important contribution to the literature on this topic by reporting how different states treat these assets differently, resulting in their conclusion that overall more than two-thirds of these retirement assets are counted toward Medicaid eligibility.

Warshawsky has publicly expressed befuddlement at this outcome. How can different states interpret and enforce the law so differently causing inequities like the Vermont/New Hampshire example the paper cites. I can answer that question based on my experience three decades ago as a Medicaid State Representative for the Health Care Financing Administration (CMS’s predecessor). The reality is that state Medicaid programs flout federal law and regulations frequently and with impunity. Often federal Regional Offices do not hold states to the letter or even the spirit of the law. CMS did not enforce the OBRA ’93 estate recovery requirement on states. The main takeaway here is that although this paper concludes that 71.3 percent of retirement assets are counted toward determining Medicaid LTC eligibility, the federal law and regulations actually exclude nearly 100 percent and that is a condition that needs to be corrected if Medicaid is to maximize benefits for the poor and stop crowding out personal responsibility and long-term care planning among the affluent.

Improving LTSS: “To advocates of the federal provision of social insurance for LTSS, the current Medicaid program is inadequate. Furthermore, these advocates claim that failures in private insurance provision, coupled with the lack of retirement savings, make market solutions for the provision of LTSS untenable. As we have seen, such claims are not supported by the data.” (p. 35)

LTC Comment: Yes, thank you, exactly what we’ve been saying for so long.

Improving LTSS: “[E]mpirical evidence on the net worth of retired households clearly indicates widespread and significant holdings of housing and retirement assets. Those holdings are in precisely the asset classes that Medicaid rules and state administrations either always or sometimes exempt from consideration in determining eligibility (Warshawsky and Zohrabyan 2016). Lax eligibility criteria and administration, and weak estate recovery procedures and efforts, have led middle- and upper-income older Americans to seek Medicaid enrollment.” (p. 35)

LTC Comment: Yes again.

Improving LTSS: “To ensure that the Medicaid resources are allocated to individuals with insufficient financial means to pay for their long-term care, eligibility rules regarding retirement assets must be tightened across the country.” (p. 36)

LTC Comment: Of course.

Improving LTSS: “Additionally, the federal government should require states to narrow the ‘primary residence’ exclusion in examining applicants. . . . As we have seen in one study, older Americans vary home ownership depending on the rigidity of Medicaid policies. New federal rules requiring states to reject applicants with home equity interest exceeding $100,000 would diminish this strategic behavior and would ensure that program enrollees are those with legitimate financial need.” (p. 37)

LTC Comment: We’ve recommended an even lower home equity exemption of $50,000. Ironically, the current U.S.A. home equity exemption of up to $860,000 far exceeds the comparable exemption of 23,500 British pounds (approximately $29,500) in the UK’s socialized health care system. What’s more, the Brits limit their exemption to all assets, not just home equity. So much for the idea that ours is an evil, dog-eat-dog, force ‘em-into-impoverishment system.

Improving LTSS: “Preventing Medicaid funds from being distributed to households with significant assets also requires a more robust estate recovery scheme. Despite the growth in recovery programs over the past 15 years, recoveries represent a negligible percentage of both LTSS expenditures and the estimated net worth of program enrollees. With the best efforts of the most vigorous state, almost six times the current estate recoveries could be collected in aggregate by states.” (p. 37)

LTC Comment: Absolutely, we make the same point and elaborate on how to achieve it in “Maximizing NonTax Revenue from MaineCare Estate Recoveries” (2013); “Medicaid Estate Recoveries in Maine: Planning to Increase Non-Tax Revenue and Program Fairness” (1993); Medicaid Estate Recoveries: National Program Inspection (1988); The Medicaid Estate Recoveries Study--Volume I: Estate Recoveries in the Medicaid Program (1985) and in these LTC Bullets: LTC Bullet: Medicaid Estate Recovery, Wednesday November 8, 2000; LTC Bullet: Medicaid Estate Recoveries Clarified by HCFA, Wednesday March 7, 2001; LTC Bullet: The Critical Role of Medicaid Estate Recoveries, Friday, September 30, 2005; LTC Bullet: Medicaid Estate Recover. . .up, Thursday, July 5, 2007; LTC Bullet: How Estate Recovery Protects the Poor AND the Affluent, Wednesday, July 1, 2009; LTC Bullet: How Medicaid LTC Sprung a Leak, Monday, September 14, 2009; LTC Bullet: Center Tackles Medicaid Estate Recoveries, Friday, April 26, 2013; LTC Bullet: The Role of Estate Recoveries in LTC Financing, Friday, June 7, 2013; LTC Bullet: Free LTC Loan With No Pay Back Required, Friday, August 22, 2014; LTC Bullet: Holding CMS’s Feet to the Fire, Friday, February 6, 2015; LTC Bullet: Real ID vs. Estate Recoveries: A Decade of Divergence, Friday, February 26, 2016.

Improving LTSS: “Specifically, the federal government should enforce the existing requirement on states to automatically impose liens on the housing properties of beneficiaries. Imposing these liens in all cases would significantly improve recovery rates. (pps. 37-38)

LTC Comment: This is incorrect. There should be a mandatory lien on real property of Medicaid LTC recipients, but there is none. States may impose liens, but many do not.  According to “States may impose liens for Medicaid benefits incorrectly paid pursuant to a court judgment. States may also impose liens on real property during the lifetime of a Medicaid enrollee who is permanently institutionalized, except when one of the following individuals resides in the home: the spouse, child under age 21, blind or disabled child of any age, or sibling who has an equity interest in the home. The states must remove the lien when the Medicaid enrollee is discharged from the facility and returns home.” (“Estate Recovery and Liens” at

Improving LTSS: “As a penalty for state noncompliance with the housing lien procedures and to counteract the weak state incentives to collect, the federal government should decrease Medicaid matching rates for LTSS expenditures for states performing inadequately.” (p. 38)

LTC Comment: With the correction that estate recovery, not imposition of a lien, is the federal requirement, states definitely should be penalized for failure to conduct estate recovery effectively and efficiently. This should have been done all along and does not require Congressional action to lower federal financial participation for deficient states. Federal Medicaid Regional Offices have financial auditors whose job it is to evaluate whether state Medicaid programs are complying with federal law and regulations and to lodge claims for reimbursement of federal funds in cases of non-compliance. The fact that states like Texas, Michigan, Georgia and New Mexico failed to implement estate recoveries for many years after such programs were mandated by OBRA ’93 is the result of CMS’s Regional and Central Offices’ failure to enforce the law.

Improving LTSS: “Data show that even at the median of the older population, substantial assets exist to pay for retirement expenses, including LTSS. With updated eligibility-determination rules and processes, systematic asset tracking, and enhanced estate recovery programs, the Medicaid LTSS program can achieve financial sustainability and simultaneously reduce the crowding out of private alternatives such as LTCI. Federal mandates requiring states to undertake these reforms can result in substantial future savings for taxpayers while strengthening a key part of the social safety net for those who truly need it.” (p. 39) 

LTC Comment: Bottom line: we don’t need a new compulsory social insurance program for long-term care. What we need to is give Medicaid back to the needy and incentivize everyone else, with carrots and sticks, to plan early and save, invest or insure for long-term care.