VoxEU Column Health Economics

Promoting long-term care insurance

Limited insurance for long-term care threatens the sustainability of publicly funded social assistance programmes such as Medicaid in the US. This column looks at the effectiveness of a programme that encourages middle-income people to save for possible long-term care expenses. The evidence so far indicates that although this programme has indeed increased insurance applications, it has not increased insurance uptake.

Incentivising long-term care insurance

Barely 14% of Americans over the age of 50 purchase private insurance to protect against the costs of long-term care needs (Norton 2000, Costa-Font 2012, Frank 2012). As a result, either individuals (or their families) bear the (financial or personal) costs of long-term care, or they turn to Medicaid, the federal-state programme for low-income people, to finance their long-term care expenses. The latter option is straining the financial sustainability of Medicaid as a programme (as it has been designed to date).  

To reduce the financial burden of long-term care on Medicaid, both the federal and state governments have developed strategies that attempt to shift such costs away from Medicaid. These include point-of-purchase incentives, such as state and federal tax deductions, for purchasing long-term care insurance. However, analyses of these strategies indicate limited returns of state tax deductions on the dollar (Goda 2011). Alternatively, insurance design has been experimenting with incentives targeting the point of use. One version of this strategy includes long-term care partnership insurance contracts, which, in their most common format, allow individuals to purchase long-term care insurance that protects any assets up to the value of insurance coverage so that once exceeded, they become eligible for Medicaid coverage. This strategy is known as the Long-Term Care Partnership Programme. Its design attempted to provide an incentive for middle-income people to save more for possible long-term care expenses in their older years and to reduce the use of ‘spend down’ strategies to qualify for Medicaid. The programme was initially developed in four states (California, Connecticut, Indiana, and New York) and its extension to the majority of US states has taken place only since 2005 (Meiners 1992, Bergquist et al. 2015a).

Figure 1. Evolution of the total number of contracts, 2000-2008

Evidence from partnerships

In a recent paper we examine the effects of purchases and applications of long-term care insurance during the period 2000-2008 (both included), and trends in such purchases and applications after the introduction of the Long-Term Care Partnership Programme (Bergquist et al. 2015b). We adopt a flexible strategy using a triple interaction model so we can separate the pre-existing trends in the market for long-term care insurance from the effect of the programme.

Our findings indicate some interesting results, namely that the Partnership Programme expanded insurance applications but not insurance purchases (except for the year of implementation), as Figure 1 reveals.

These findings suggest two implications. First, insurance underwriting of potential insurance consumers is limiting the expansion of insurance uptake. That is, insurance applications are declined for being ‘too risky’ despite consumers being willing to pay the premium. Second, we find a short-term substitution between traditional insurance contracts and the Long-Term Care Partnership Programme contracts when they were launched. Other explanations for our results include poor targeting of Partnership Programme policies to middle-class individuals, along with poor informational and marketing campaigns for the programmes (Bergquist et al. 2015a).

Policy implications

The expansion of private long-term care insurance is subject to a number of barriers. In Europe, social insurance in some countries makes private insurance complementary or supplementary. In the US, even when public insurance is limited to poor people, insurance underwriting limits the potential purchases of private insurance to the middle class who are left without much alternative but to either spend down their wealth or self-insure. In addition, amongst those who manage to get insurance coverage, substantial increases in premiums often cause individuals to allow their insurance contract to lapse. Hence, incentives ought to be more targeted to the middle class, and policies should be put in place to reduce the incentive of insurance companies to only provide coverage to the ‘best risks’.


Bergquist S, J Costa Font, and K Swartz (2015a), “Partnership program for long-term care insurance: the right model for addressing uncertainties with the future?” Aging and Society: forthcoming.

Bergquist S, J Costa-Font and K Swartz, (2015b), “Long Term Care Partnerships: Are they 'Fit for Purpose'?”, CESifo Working Paper Series 5155, CESifo Group Munich.

Costa-Font, J (2012), “Addressing the incompleteness of long-term care insurance”, VoxEU.org, 9 June.

Costa‐Font J, C Courbage and K Swartz, (2015), “Financing Long‐Term Care: Ex Ante, Ex Post or Both?”, Health Economics, 24, 45-57

Frank, R G (2012), “Long-term care financing in the United States: sources and institutions”, Applied Economic Perspectives and Policy, 34(2), 333-345. doi: 10.1093/aepp/pps016   

Goda, G S (2011), “The impact of state tax subsidies for private long-term care insurance on coverage and Medicaid expenditures”, Journal of Public Economics,

Meiners, M R, H L McKay,  and K J Mahoney(2002), “Partnership insurance”, Journal of Aging & Social Policy, 14(3-4), 75-93..

Norton, E C (2000), “Long-term care”, In A J Culyer & J P Newhouse (Eds.), Handbook of Health Economics (Vol. 1): Elsevier Science B.V.

420 Reads