New Research Finds Congressional Budget Office Has Underestimated Savings and Overestimated Costs from Health Policy Changes

August 26, 2009

Contact(s):

Over the last 30 years, the Congressional Budget Office (CBO), which assesses the costs of health reform and other legislation as it moves through Congress and is widely respected for its competence and integrity, has underestimated the amount of savings and overestimated the costs that major changes in the health care system would bring, says Jon Gabel in an op-ed published in today's New York Times.

Drawing on Commonwealth Fund-supported research, Gabel, a senior fellow at the National Opinion Research Center of the University of Chicago, analyzed CBO's forecasts of three major changes in the Medicare program relative to their ultimate outcomes. He found that in the early 1980s, CBO underestimated savings from reforms Congress made in the way Medicare paid hospitals by $11 billion. Under the new law, Medicare would pay hospitals a fixed amount per admission based on the patient's primary medical condition, instead of covering costs incurred during a patient's stay. This encouraged shorter stays, led to fewer diagnostic services, and reduced administrative costs. CBO predicted that by 1986 total spending would be $60 billion. Actual spending in 1986 was $49 billion.

Gabel also found that savings from the Balanced Budget Act of 1997, which changed the way skilled nursing facilities and home health services were reimbursed under Medicare, turned out to be 50 percent greater in 1998 and 113 percent greater in 1999 than the budget office forecast. And, CBO predicted that drug prices would rise following the Medicare Modernization Act of 2003, which added prescription drug benefits to Medicare, estimating that spending on the drug benefit would be $206 billion. Actual spending was nearly 40 percent less than that, Gabel found.

Gabel explains that when CBO analyzes initiatives aimed at reducing costs, it requires considerable evidence that similar previous policy changes have saved money. When there is a lack of historical examples, the "unknown" variable often becomes zero. The task for the budget office becomes even more challenging when it considers the impact of more than one change simultaneously—changes that might have synergies.

Gabel observes that underestimating savings that can come from cost-control initiatives in Medicare and throughout the health system could undermine efforts to pass health reform legislation. "As Congress now works on its greatest push for health care reform in generations, the budget office needs to revise the methods it uses to make predictions about costs," he says.