Premium Sensitivity and Persistence of Health Plan Choice Among the Elderly

To reduce the growth of federal spending on healthcare, policymakers have considered changes to various federal health programs and in particular Medicare.  For example, premium support proposals would convert Medicare into a defined contribution benefit with or without the traditional Medicare program as a plan option.  The effects of such proposals on federal spending and health plan enrollment depend on how beneficiaries respond to financial incentives, and in particular, how beneficiaries’ choice of plan depends on health insurance premiums.

To help address these issues, this paper estimates Medicare beneficiaries’ price elasticity of demand for health insurance using premiums in the Medicare Advantage market as the central measure of price.   In most cases, when Medicare beneficiaries first enter the program they are given the opportunity to choose between the traditional fee-for-service benefit (FFS) or one of several private health insurance plans which are offered through the Medicare Advantage program.  In subsequent years, enrollees may elect to switch plans (including the choice of FFS), although the opportunity to switch is limited to once per calendar year.  A key question in this analysis is how price sensitivity varies with length of enrollment.  Previous evidence suggests the presence of inertia; enrollees seem likely to remain in the same plan over several years even though premiums or benefits may have changed (Strombom et al., 2002). This suggests beneficiaries may be more price sensitive in the first year of enrollment than in later years.

Earlier work on the price sensitivity of elderly beneficiaries has mostly focused on either the health plan choices within employer-based settings (Buchmueller, 2000; Buchmueller, 2006; Buchmueller, 2012) or utilized cross-sectional survey data (Dowd et al., 2003; Atherly et al., 2004).  This paper utilizes several unique sources of data to construct a panel of beneficiaries and their plan choices and to produce premium elasticity estimates over time. By linking together two types of Medicare administrative data for the years 2006 through 2011 – plan enrollment information and premium and benefit data for Medicare Advantage plans – I offer estimates that are more applicable than firm-specific studies, and that contain intertemporal premium variation that can be used to generate more reliable estimates than those found in the cross-sectional literature.

This paper develops a model of health plan choice which analyzes beneficiaries’ sensitivity to insurance premiums in both their initial choice year and in subsequent years.  I utilize a fixed effects model of health plan choice that exploits variation in health plan premiums and benefits over time to identify sensitivity to their changes.  A key component of these models examines whether lagged price changes may be a more important influence on plan choice than current prices, which may indicate inertia or persistence in beneficiaries’ choices of health plans.  Similar evidence has been found in Medicare’s prescription drug insurance program (Ericsson, 2012) and in employer-based health insurance settings (Handel, 2012).