February 2, 2004

Honorable Jim Nussle
Chairman
Committee on the Budget
U.S. House of Representatives
Washington, DC 20515

Dear Mr. Chairman,

CBO's baseline budgetary projections released in the Budget and Economic Outlook include $395 billion in outlays over 2004 to 2013 for the Medicare Prescription Drug, Improvement, and Modernization Act of 2003 (Public Law 108-173). That amount is identical to CBO's scoring of the bill when passed. In contrast, the Administration estimates that additional outlays resulting from that act will total $534 billion over the 2004-2013 period.

Of course, a complete comparison of the overall budgetary impact of the legislation must also consider the effect on revenues. CBO estimates that the revenue effects of the legislation are largely offsetting. The legislation reduces revenues by providing qualified taxpayers with health savings accounts. At the same time, it increases revenues, CBO estimates, as businesses reduce expenditures on nontaxable health benefits and increase them on taxable wages. The Administration has not released its estimated effects of the legislation on revenues. Those estimates could certainly differ from CBO's.

Because the new prescription drug program represents a major departure from what currently exists, there is a great deal of uncertainty about its budgetary impact and a wide range of possible outcomes. CBO's estimate was the result of extensive analyses of the pharmaceutical drug market, the Medicare program, and the likely responses of potential enrollees. To date, we have not received any additional data or studies that would lead us to reconsider our conclusions. Therefore, CBO believes its estimate is sound and has no reason, at present, to revise it.

CBO has consulted with the Administration to identify the major factors that account for the differences between the two estimates. Although such a comparison is complicated and we do not have complete detail on the key attributes, it appears that the difference derives from differing assumptions or estimates in a number of areas. Attached is a summary of those major differences. We will continue to work with the Administration to understand the differences in more detail.

I hope this information is helpful to you. The CBO staff contact for this analysis is Tom Bradley.

Sincerely,

Douglas Holtz-Eakin
Director
 

Enclosure
 

cc:	Honorable John M. Spratt Jr. Ranking Member

Identical letters sent to Honorable William "Bill" M. Thomas, Honorable W. J. "Billy" Tauzin, Honorable Don Nickles, and Honorable Charles E. Grassley
 
 

Note: CBO estimates that the revenue effects of H.R. 1 total less than $0.5 billion over 10 years. CBO presently has no specific information on the Administration's estimate of the revenue effects of H.R. 1.

The Administration's estimate of $534 billion in outlays through 2013 is $139 billion higher than CBO's $395 billion estimate.Almost all of that difference is attributable to our estimates of the Part D drug benefit and the Medicare Advantage program.

PART D drug benefit: The Administration's estimate is about $100 billion higher (2004-2013)

• Basic benefits account for about $32 billion (about 7 percent higher than CBO estimates).
  
  The Administration assumes a higher participation rate (94 percent of all Medicare enrollees, compared to CBO's assumption of 87 percent participation).

  Both agencies exclude about 5 percent who would be subject to secondary-payer rules. The Administration assumes that 94 percentage points out of the remaining 95 percent will participate.
  
  CBO excludes Medicare enrollees who decline Part B--we assume those beneficiaries generally will not participate because they have already showed that they will turn down a benefit with a 75 percent subsidy. CBO also excludes some beneficiaries who have generous prescription drug coverage through the Federal Employees Health Benefits program and other federal programs.

  The Administration estimates per capita costs that are about 4 percent higher throughout the 2006-2013 period than CBO estimates.
  
  The differences in the number of participants and the per capita costs each account for about half of the $32 billion difference in the cost of the basic benefit.



• The Administration's low-income subsidy is about $47 billion higher (a 24 percent difference).
  
  The Administration assumes higher participation (13 to 15 percent higher than CBO after the third year of the program). CBO estimated participation in the low-income subsidy based on experience with participation of Medicare beneficiaries in the Medicaid, QMB, and SLMB programs.
  
  The Administration assumes full participation immediately; CBO ramps up over three years.
  
  The Administration's per capita cost is higher than CBO's--initially 7 percent to 10 percent higher, with the difference shrinking to 4 percent by 2013.



• Savings in the Medicaid program account for about $18 billion of the difference. Medicaid savings will occur because creation of the Medicare drug benefit will end Medicaid's responsibility for providing prescription drugs to individuals who are eligible for both programs. However, those savings will be offset in part by additional Medicaid costs for newly enrolled people. CBO estimates that H.R. 1 will reduce federal Medicaid spending by $141 billion over the 2004-2013 period, compared to the Administration's estimate of $123 billion.
  
  CBO's estimate of net savings is higher largely because of differences in baseline projections for Medicaid spending on waiver programs to provide limited drug coverage to low-income Medicare beneficiaries who do not otherwise qualify for Medicaid. CBO's baseline projection of federal Medicaid spending for those waiver programs was $18 billion higher than the Administration's baseline projection.

Medicare Advantage: The Administration's estimate is $32 billion higher ($46 billion vs. $14 billion)

• The Administration assumes much higher participation (32 percent of Medicare beneficiaries enrolled in Medicare Advantage (MA) plans vs. CBO's estimate of 9 percent).⁽¹⁾
  
  Both estimates assume that many of the participants in MA plans are in areas where the payments to MA plans and beneficiaries (through premium rebates) would exceed what it would cost if those beneficiaries were in the fee-for-service (FFS) sector. Most of the additional participants in the Administration's estimate are in relatively low-cost, low-density areas where the payments to MA plans and beneficiaries would be substantially higher than the cost of those beneficiaries in the FFS sector.