A central challenge for ageing Western societies is securing sustainable financing for long-term care (Araki et al. 2024). This is especially salient in the US, where roughly half of Americans reaching the age of 65 will eventually require some form of long-term care (Favreault and Dey 2015). Most older adults rely heavily on their own wealth – especially housing and financial assets – to cover care costs not met by Medicaid, which finances nearly half of US expenditures on long-term services and supports (LTSS). Because housing is the dominant non-pension asset for older Americans, wealth shocks stemming from fluctuations in home prices or stock market valuations can meaningfully influence whether individuals self-insure or seek private long-term care insurance (LTCI). Rising wealth may reduce the perceived need for private insurance, whereas negative shocks may increase vulnerability to high care costs and potential Medicaid eligibility.

Private LTCI remains a marginal product in the US, with only around 8% of adults holding coverage and participation continuing to decline over time (Costa-Font and Raut 2025a, 2025b), and it is far from clear that there is clear potential to expand further. Prior research attributes this underdevelopment to high premiums, policy complexity, low financial literacy, aggressive underwriting, instability among insurers selling policies, and limited impact of tax incentives or partnership programmes. Behavioural factors – such as personal caregiving experience – and informational interventions have also been shown to influence take-up. However, even when structural barriers are removed, many wealthier households may prefer to self-insure, especially in a system where the means-tested Medicaid programme serves as the default safety net for those who deplete their income and assets. Understanding how changes in wealth affect LTCI decisions is therefore key to assessing the future of LTSS financing.

Wealth and long-term care funding

Housing and financial wealth play a critical role in long-term care insurance decisions in ageing societies where traditional family-based caregiving is declining. Increases in housing or financial assets raise household wealth, influencing consumption, planning, and LTCI purchases. Older adults have a strong preference to age in place and most hold housing wealth, which is often drawn upon only at advanced ages or in response to health shocks, making unexpected housing price changes particularly influential for LTSS decisions. Because wealth and health are closely linked, shocks to housing assets can affect both functional impairment risk and care choices, potentially shifting demand between formal (paid) and informal (unpaid) LTSS, altering Medicaid eligibility, or prompting LTCI purchases to protect remaining assets. Overall, changes in either asset class can alter perceived lifetime resources, shaping LTCI behaviour through multiple channels. Illiquid housing wealth is posited to crowd out LTCI demand, whereas financial wealth does not (Davidoff 2008, 2010), however there is limited causal evidence to disentangle these effects.

Evidence from exogenous variation in housing prices and stock market wealth

Using 20 years of data from the Health and Retirement Study, in a recent paper (Costa-Font et al. 2025) we exploit exogenous variation in both housing prices and stock market wealth to measure how wealth shocks affect the probability of holding LTCI or enrolling in Medicaid. We find strong evidence that positive wealth shocks, particularly gains in liquid financial assets, significantly reduce demand for private LTCI.

Specifically, we find that a $100,000 increase in housing or financial assets lowers the probability of holding LTCI coverage by 1–3 percentage points (on a base of X), reflecting a substitution effect consistent with self-insurance. A $100,000 increase in housing wealth lowers the likelihood of LTCI coverage by 1.24 percentage points, while an equivalent gain in stock wealth reduces it by 3.22 percentage points.

In contrast, wealth shocks have no meaningful effect on Medicaid enrolment, suggesting that shifts in private coverage are driven by choices about self-insurance rather than changes in eligibility. These findings extend earlier work on housing wealth and LTCI (Davidoff 2008) by providing the first causal estimates on LTCI take uptake. The estimates show that both the composition and volatility of household wealth shape LTSS planning.

These effects are concentrated among homeowners and stockholders, while renters and non-stockholders show no significant response. Back-of-the-envelope calculations suggest that LTCI coverage could have declined by approximately 6.5% between 2016 and 2025, consistent with the observed mean coverage in 2016.

Financial affordability and health effects after wealth shocks

Two distinct pathways explain the effect: a direct financial affordability effect and an indirect effect via expected disability risk. The financial affordability pathway reflects how greater wealth enhances a household’s capacity to self-insure against potential long-term care expenses. Wealth increases access to better healthcare, nutrition, and living conditions (Cheng et al. 2018), boosts income from assets, and allows greater investment in pensions and annuities – all of which act as enhanced self-insurance. Increases in housing and total wealth also stimulate bequest motives, providing additional LTSS options through family transfers, as evidenced by a higher probability of leaving bequests of at least $10,000, reducing reliance on private LTCI.

The second pathway operates through health improvements. Wealth expansions are associated with better health (Schwandt 2018), reducing the anticipated need for LTSS and, consequently, the likelihood of purchasing LTCI. Increases in housing and total wealth also raise out-of-pocket medical spending, consistent with reduced reliance on insurance, though financial wealth alone shows no significant effect. Finally, while bequest motives are important, we find no significant impact of wealth on informal care provision from children, suggesting that the primary mechanisms linking wealth and LTCI are financial self-insurance and health-related reductions in care needs.

Welfare and policy implications

These results carry important welfare implications. By relying on housing and financial wealth as a self-insurance mechanism, households may face increased financial vulnerability and potential asset depletion in the absence of broad-based insurance protection. Reduced LTCI uptake may constrain access to LTSS, creating unmet needs.  However, to date it appears that the expansion of private LTCI has failed to pass the market test.

Our findings point to the fact that while wealthy individuals substitute private insurance with self-financing, this mechanism does not extend to public insurance or Medicaid. We find no significant evidence that a positive wealth shock affects Medicaid receipt, most likely because the sample of individuals that hold housing or financial wealth are unlikely to meet the Medicaid eligibility requirements. That is, individuals at risk of spending down to Medicaid tend to hold fewer assets, making them less responsive to wealth shocks.

References

Araki, S, J Barszczewski, K Killmeier and A Llena-Nozal (2024), “Affordability of long-term care systems in times of rapid population ageing”, VoxEU.org, 29 November.

Costa-Font, J, R  Frank and N Raut (2025), “The effects of wealth shocks on public and private long-term care insurance”, Journal of Health Economics.

Costa‐Font, J and N Raut (2025a), “Long‐Term Care Partnership Effects on Medicaid and Private Insurance”, Health Economics 34(6): 1171-1187

Costa-Font, J and N Raut (2025b), “Expanding private insurance for long-term care to reduce public spending”, VoxEU.org, 20 May.

Davidoff, T (2008), “Illiquid housing as self-insurance: The case of long-term care”, mimeo.

Davidoff, T (2010), “Home equity commitment and long-term care insurance demand”, Journal of Public Economics 94(1-2): 44-49.

Favreault, M and J Dey (2015), “Long-Term Services and Supports for Older Americans”, Risks and Financing Research Brief, Office of the Assistant Secretary for Planning and Evaluation.