Skip to content

2014 Conference Presentation

Evaluation United States

1 September 2015

Long-term care insurance for married individuals: insuring one without the other?

Norma B Coe, University of Chicago, United States
Tamara R Konetzka, University of Chicago, United States
Courtney van Houtven, University of Chicago, United States


Objective: Most research on long-term care insurance focuses on individual decision making. From other studies we know that married individuals are more likely to purchase long term care insurance (LTCI), compared to single individuals, and that one’s spouse’s characteristics also significantly influence an individual’s purchase. Although there is some support from qualitative work for the idea that couples consider long-term care planning and purchases of LTCI together, we know little about purchase patterns within a couple. In this paper, we describe the purchase patterns for LTCI among married couples and explore the potential reasons underlying these purchase patterns.

Data and Methods: We use data from eight waves of the HRS (1996–2010), a publicly available, nationally representative bi-annual survey of the near elderly in the U.S. Respondents were ages 51–61 when they entered the sample initially, old enough to have formed expectations and decisions about LTC planning and LTCI purchase. We include all HRS cohorts and the precursor to the HRS, the Assets and Health Dynamics Among the Oldest Old (AHEAD) cohort. We link the HRS to the restricted Cross-Wave Geographic Information (State) file to obtain respondents’ state of residence and to control for unobserved state attributes. We also use state-level variation Medicaid asset limits for community spouses to identify marginal purchasers among married couples. Finally, for couples who purchase policies for one member but not the other, we explore distal outcomes such as death and nursing home entry, to determine whether couples “got it right”. That is, among those who bought only for one member of the couple, was that member more likely to become disabled and need LTC?

Results: Whereas only 16% of the couples ever purchased LTCI while in the panel, their purchase patterns do not meet expected patterns. Considering a sample where at least one member of the couple meets LTCI purchase eligibility, just over 5% of couples buy for the woman only and 5% buy for the man only. The remaining 6% buy for both members of the couple. The slightly higher purchase rate for women is not statistically significantly different than the purchase rate for men. These patterns cannot be explained by differential spousal eligibility, or sequential timing of purchases. Analysis of the role of other characteristics is ongoing.

Policy Implications: Despite much attention from academics and policy makers, LTC remains one of the largest uninsured risks facing American elderly today. Understanding how couples make the decision whether to purchase is critical to inform efforts to expand LTCI markets, especially considering premiums will likely begin to differ by gender.

Audio   Slides

Skip to toolbar