The costs of plan choice and inertia in the Medicare Advantage market

Medicare Advantage (MA) consumers generally have a wide array of plans to choose from and, without their action, are reenrolled in the same plan each enrollment period. Passive reenrollment and other characteristics of MA may discourage enrollees from continually reconsidering their plan choices, potentially resulting in considerable plan choice inertia.  In fact, about half of MA beneficiaries remain in the same plan over a 5-year time period.  While researchers studying the Medicare prescription drug market have shown that there are significant financial consequences of plan choices on the premiums and out-of-pocket costs that beneficiaries pay, we are not aware of research on these effects in the MA market.

Our analysis relies on a unique link between several administrative databases that the Centers for Medicare and Medicaid Services (CMS) collect. The underpinning of our analysis is a 2 percent random sample of Medicare enrollment files. We link this sample to MA plan characteristics in the Medicare Advantage Bid Pricing Tool, which includes data on premiums and rebated benefits, and the Medicare Plan Finder Out-of-Pocket Cost Estimates, which includes CMS-derived estimates of the average expected out-of-pocket medical costs (OOPC) for each plan.

Using these data, we compare the plans beneficiaries chose to other plans available in their county and assess the degree to which beneficiaries might have reduced their payments by choosing differently.  Our key analyses are regressions of the difference between beneficiary costs associated with their chosen plan and the costs of the plan which would have minimized their total expected spending.  We consider differences across three dimensions: 1) enrollee premiums; 2) expected OOPC; and 3) the actuarial value of any supplemental benefits.  To highlight the cost of inaction, we show how these potential savings vary by whether beneficiaries switched plans between years and by the length of time the beneficiary has been enrolled in the same MA plan.  We control for beneficiary health risk, which reduces the bias from beneficiaries in relatively worse health status potentially choosing more expensive plans on dimensions which are unobservable in the data, such as wider provider networks.

Our results suggest that the average MA enrollee in 2013 could have reduced their spending for premiums and expected OOPC by $716 dollars per year by choosing a plan with a lower premium and a comparatively small increase in cost-sharing for medical services.  Beneficiaries who have enrolled in their plans for 6 or more years were most at risk of spending these higher amounts, paying $869 more than they would have spent in the lowest cost plan, compared to $520 for beneficiaries in their first year of enrollment. We find that, over time, beneficiaries pay more for premiums without consequent reductions in expected OOPC, suggesting that they may not actively and regularly compare plans with resulting inertia in plan selections. However, we cannot conclusively show that these choices are non-optimal, particularly because beneficiaries may be willing to pay more for access to specific providers or to reduce their financial risk, or they may face high switching costs.