The Medicare Advantage (MA) program, formally Part C of Medicare, originated with the Tax Equity and Fiscal Responsibility Act (TEFRA), which authorized Medicare to contract with risk-based private health plans, or those plans that accept full responsibility (i.e., risk) for the costs of their enrollees’ care in exchange for a prospective, monthly, per-enrollee payment. This program has been called a variety of names over the past three decades (e.g., Medicare Advantage, Medicare+Choice). In this article, we refer to the program by both its current names, Part C and Medicare Advantage (MA). TEFRA was passed in 1982, and the rules to implement risk-based contracting were completed in 1985. Those beneficiaries who choose to enroll in an MA plan continue to pay, directly to Medicare, their required Part B premium for physicians’ services and, if they elect it, their Part D premium for drug coverage. In return, they receive health insurance for all services through (and may pay supplemental premiums to) their MA plan. For these beneficiaries, enrollment in an MA plan replaces not only traditional Medicare but also a Medicare supplemental insurance policy (i.e., Medigap). The MA plans themselves receive, directly from the Medicare program, a predetermined, monthly, risk-adjusted payment to cover each beneficiary’s care. (Later we will describe in detail the beneficiaries’ enrollment decisions and plan payment.)
Over the past twenty-five years the MA program has pursued two stated goals. The first is to expand Medicare beneficiaries’ choices to include private plans with coordinated care and more comprehensive benefits than those provided through traditional Medicare (TM) (MedPAC 2001, chap. 7). The second is to take advantage of efficiencies in managed care and save Medicare money (Prospective Payment Assessment Commission 1997, chap. 3).
Reducing Medicare program or on-budget spending is different from reducing economists’ notion of social cost, which reflects the opportunity cost of the resources actually used for medical care. Although we focus here primarily on the perspective of the budget, we believe that there is a third goal, which is related to the economists’ social cost concept: to minimize the inefficiencies induced by the inevitable errors in TM’s administered price system, by allowing the health plans and providers to negotiate prices or, in some cases, to integrate the finance and delivery functions. An example is a group or staff model Health Maintenance Organization (HMO).
On first glance, the two stated goals appear to be at odds with each other. An obvious way to increase access to MA plans would be for Medicare to increase the plan payments, thereby making it more attractive for plans to enter the Medicare market. But doing so would contradict efforts to save Medicare money. In principle, however, it is possible to attain both stated goals (as opposed to having no MA program). Traditional Medicare (TM), which consists of Part A (mainly hospital insurance), Part B (mainly physician services), and, as of 2006, Part D (prescription drug coverage) is on an unsustainable cost path, due in part to pricing errors that make certain services or sites of care either profitable (provide economic rents) or unprofitable (Ginsburg and Grossman 2005; Newhouse 2002) and in part to utilization induced by supplemental insurance coverage from former employers or purchased individually (MedPAC 2010b). Including the 16 percent of those Medicare beneficiaries also eligible for Medicaid, who cannot realistically afford the amount of cost sharing in TM, such supplemental coverage is held by the majority of beneficiaries (Atherly 2001; Christensen, Long, and Rodgers 1987; Dowd et al. 1992). Furthermore, TM’s fee-based physician payments, which are based on volume and hospital payments based on admissions, accommodate variations in provider practice patterns, in turn absolving providers from pressure to restrain overuse. Geographic variations in utilization and quality create business opportunities in places like south Florida (one of the highest per-capita TM spending regions in the United States), where private managed care plans ought to be able to expand choice to beneficiaries and to provide care that is just as good as or better than TM for less money (Dartmouth Medical School 1999; Fisher et al. 2003a, 2003b). If the MA program induced the plans to enter the right markets and the right beneficiaries to choose those plans, creating choice for the beneficiaries could save Medicare money and achieve both stated goals of the Part C program.
To satisfy both of these goals, however, Part C’s payment rules would need to thread a policy needle: The plans would have to be paid enough by Medicare and by the plan enrollees to make a profit sufficient to justify their participation and to offer care that satisfied regulatory requirements. At the same time, they would have to keep the beneficiaries’ premiums and cost sharing low enough or to offer enough additional services to attract beneficiaries from TM. In addition, Medicare would have to pay less than it would if the beneficiaries who enrolled remained in TM, or Medicare would not save money.
How have things gone? So far, not so well. Over more than twenty-five years, as Medicare policymakers have attempted to meet both the policy’s challenges and their own political objectives, Medicare has not been able to find payment rules that simultaneously expand beneficiaries’ choices and save Medicare program funds. As the 2010 debate on health reform showed, in the years since the 2003 Medicare Modernization Act (MMA) was enacted, MA plans have been generously paid, resulting in expanded choice and enrollment (achieving the first goal), but costing Medicare more money than TM, an estimated $14 billion more in 2009 (and thus failing on the second goal). As the June 2007 Report to Congress of the Medicare Payment Advisory Commission (MedPAC) explained, “Current MA payment policy is inconsistent with MedPAC’s principles of payment equity between MA and the traditional FFS program” (MedPAC 2007, xii). This “overpayment” to MA plans will be substantially reduced by the 2010 Patient Protection and Affordable Care Act (P.L. 111–148, or ACA), with new MA payment rules that freeze MA plan payments for 2011 and further reduce MA spending in 2012 and 2013. As a result, the Congressional Budget Office (CBO) projects that MA’s enrollments will fall each year until 2017 (CBO 2010).
Medicare’s policy failure is not new, and in this article we start at the beginning to put the current policy in context. In 1972 Congress first authorized capitation payments for services covered under Parts A and B. But no action was taken until 1976, when Medicare began to field demonstration projects that contracted with HMOs to provide care for Medicare beneficiaries in exchange for prospective payments. In the 1970s and early 1980s these demonstration HMOs provided some of the first evidence of managed care’s potential savings by reducing the number of Medicare beneficiaries’ inpatient hospitalizations by 8 percent over two years. But these demonstration projects also showed danger signs of favorable selection into managed care plans. That is, in the two years preceding enrollment in a Medicare demonstration HMO, the average demographically adjusted Medicare reimbursement per enrollee was 21 percent lower for those beneficiaries who enrolled in the Medicare demonstration HMOs than for those beneficiaries who did not (Eggers 1980; Eggers and Prihoda 1982; Langwell and Hadley 1989). Because the demographic adjustments were for age, gender, Medicaid eligibility, and institutionalization, this implied that younger, non-Medicaid eligible, and noninstitutionalized beneficiaries were signing up for managed care—in short, a healthier, less frail population than that remaining in TM.