You can search for articles in back issues of Contingencies from July/August 2000 to March/April 2009 using the search box to the right. Simply type in subject words, author's name, or article title and click search. To search for articles from May/June 2009 to the present, go to the current digital issue of the magazine and use the search function on the left of the top navigation bar.
Medicare Financing and the 2004 Technical Panel
By Edwin C. Hustead
THE FINANCING CRISIS FACING SOCIAL SECURITY seems to have faded from the news lately, but an even more serious crisis facing Medicare has barely even touched the nation’s radar screen. The average income from earmarked taxes under the hospital part of Medicare is half what is needed. The hospital fund is projected to begin to decline in 2012 and will be exhausted in 2020. The physician and the new prescription drug parts of Medicare appear to be well funded but only because the government contribution from general revenue and beneficiary premiums is set to rise in line with health care costs. Many believe that the premiums will eventually be too large for either the government or the retirees to pay.
In 2004, I was privileged to be asked to co-chair a technical panel on Medicare. Technical panels are appointed every few years by the board of trustees of the Medicare trust funds. The panelists are usually economists or actuaries, and the 2004 panel had seven members, including three members of the Academy of Actuaries.
Technical panels have the authority to review and comment on any of the assumptions and models used by the Medicare actuaries. The Medicare actuaries work for the Centers for Medicare and Medicaid Services (CMS) of the Department of Health and Human Services (HHS). We were asked to pay particular attention to two sets of assumptions and models. The first set covered the assumptions used for the long-term projections. The second included the assumptions used for the new Part D plan for prescription drugs.
There are three separate parts to Medicare that are funded through two trust funds. Part A of Medicare pays for hospital and related benefits through the hospital insurance (HI) trust fund. Part B of Medicare pays for physician and related expenses. Part D is the new prescription drug program. Parts B and D are funded through the supplementary medical insurance (SMI) trust fund.
The trustees report annually on the operations and projections of the two trust funds over 75 years. The reports use a large number of assumptions, such as the expected growth in health care costs and participation by those eligible. Given the large number of assumptions and the degree of uncertainty, the trustees show projections based on three sets of assumptions that produce low, intermediate, and high costs. The intermediate set of assumptions is the most widely used and the set I’ll use here.
The panel’s main focus in this part of our assignment was on the long-range growth rate used to project Medicare expenditures. This had also been a major question for the last panel that met in 2000. Before that panel, the trustees had assumed that, in the long term, Medicare expenses would grow at the same rate as the gross domestic product (GDP). The 2000 panel had recommended that the long-term assumption for all Medicare expenses should be GDP plus 1 percent a year. The trustees adopted that assumption.
Our panel had four more years of data on medical expenses to work with and improved models to both disaggregate and project medical expenses within the context of the total economy. Through 2002, the growth in the national health expenditure (NHE) accounts exceeded the growth in the GDP by an average of 2.6 percent a year since 1945, and an average of 1.8 percent a year since 1990. Medicare is a large part of the NHE, but the difference, if any, between Medicare and the total NHE introduces an additional layer of uncertainty.
Similar to the 2000 panel, the 2004 panel noted seven major contributors to the past growth that may or may not continue in the future. Some of these, such as the aging of the population, are at least partially recognized in the Medicare projections. Others, such as supplier-induced demand and defensive medicine, may not contribute as much to future growth as they have in the past.
However, after extensive review of this new information, we weren’t able to come up with a more supportable recommendation than that of the 2000 panel, which was that the NHE will probably continue to exceed the GDP over the next 75 years, but the difference between the two is very difficult to estimate. We agreed that the 1 percent differential introduced by the trustees in response to the 2000 panel recommendations was reasonable. We also recommended that the trustees should continue to support analyses of the differences and that they eventually develop a more precise method for setting the long-range assumptions.
The trustees report contains a number of short-term and long-term measures of the difference between income and expenditure for the HI fund. The long-term projections end after 75 years. In their 2004 report, the trustees added an “infinite horizon” measure. Our panel found that the 75-year measures and the shorter-term measures the trustees used provide a complete and adequate presentation of Medicare’s financial state. Our recommendation was that the trustees not expand the presentation of the infinite-horizon measure beyond the minor role that it played in the 2004 report.
Our other major task was to review the models and assumptions used to project the cost of the new prescription drug program. Prescription drugs had been added to Medicare in 2003, to begin in 2006, and the 2004 report was the first that included an estimate of the cost of that program. Some of the effects of the new program are becoming known this year, but the true cost of the program will only begin to surface with data on operations in late 2006. As a consequence, the 2006 trustees report will largely rely on estimates based on other sectors of the economy.
We agreed that most of the methods and assumptions the trustees used were reasonable. We did recommend that two of the assumptions be modified. The first assumption was that 99 percent of those eligible for Part D would elect to participate in Part D. We believed that, at least initially, the participation rate would be lower, with around 90 percent participating in the first year.
The second recommendation concerned the expected actions by employers in response to Part D. Part D expenses will be substantially lower if employers keep their post-retirement prescription drug benefits for retirees eligible for Medicare Part D. To encourage employers to keep their plans, the Part D trust fund will pay around $600 a year for each participant eligible for Part D. Employers who don’t maintain their current plan and take the $600 have three other choices: drop the prescription drug coverage; wrap around Part D as they do around A and B; or become a Part D plan. Each of these choices has different financial consequences for the SMI fund.
The trustees assumed that 75 percent of employers with post-retirement coverage would continue the coverage and that almost all of the other 25 percent would drop their post-age-65 coverage of prescription drugs for retirees and their spouses. They assumed that the 25 percent wouldn’t “select against” Medicare, so they would have the same expenditure characteristics as the 75 percent who maintained their post-retirement coverage.
The panel had two observations about the trustees’ assumptions about employer actions in response to Part D. First, based on evolving information, we believed that a significant number of employers will either wrap around Part D or set up Part D plans. Second, we believed that employers will tend to choose the option that’s most advantageous to them and, therefore, less advantageous for Medicare. We didn’t offer a specific assumption but recommended that the CMS actuaries consider developing information about employer reactions.
CMS assumed that the proportion of beneficiaries with employer retiree coverage would not change in the future. The panel believed that the trend of employers dropping coverage will continue and that, therefore, the proportion of beneficiaries with the coverage will decline in the future.
Table 1 summarizes the primary Part D assumptions and our recommendations on those assumptions.
There has been tremendous activity since the technical pPanel was released in January of 2005, and the structure and design of the Part D plans is much clearer than it was in January. These activities, and the initial results of the open season elections, will help the trustees improve their estimates of the Part D plan in the 2006 report.
One major concern had been the extent to which plan sponsors would develop and offer Part D plans. CMS would have been required to offer two fallback plans, sponsored by the government, in regions in which there were no bids. In fact, 10 sponsors are offering plans throughout the United States, and there are between 25 and 50 different plans available in every region. This does not include the Medicare Advantage plans that provide prescription drug coverage and are available in local areas. The question at this point has turned from what will happen if there aren’t enough plans to concern that the wide range of offerings will result in confusion among the Medicare beneficiaries and a failure to select a plan.
This type of confusion kept down participation in the discount drug card program offered to beneficiaries in an earlier phase of the Part D program. It’s likely to result in even lower enrollment than the panel expected, at least in 2006.
Perhaps as a result of the intense competition in the Part D market, the average cost of a Part D plan will be lower than expected. The average premium for the new program was determined to be $32.20, about $5.00 less than expected, and many plans are available for less than $10 a month.
Employers appear to be taking a “wait and see” attitude before deciding on the best approach to integrating with Part D plans. Most employers with post-age-65 health insurance will maintain those plans in 2006 and accept the federal reimbursement for maintaining the plans. After the Part D plans go into effect, many employers will review their programs to determine if it might be more effective, and less costly, to adopt one of the other possible approaches.
Outlook for Medicare
As we can see from the current debate over the future of Social Security, it’s difficult to express the financing of Social Security in a few numbers, and different numbers can be used to support widely different views about the future of Social Security. The problem of which numbers best show the future of Medicare is much more difficult than for Social Security because there are two distinct parts of Medicare with very different funding requirements.
Part A of Medicare is funded in the same manner as Social Security—through equal contributions by employers and employees. Benefits are paid from the HI trust fund to anyone who is disabled or over age 65 and has met the minimum coverage as a contributor with no premium required after retirement.
Parts B and D of Medicare are funded through beneficiary and government contributions made after the beneficiaries become eligible for reimbursement from Parts B and D.
The operations of the HI trust fund can be projected and costs and benefits compared in the same manner as the Social Security trust fund. Future contributions and benefits can be compared over short- and long-term periods and the unfunded obligation of the HI program can be determined.
The income and outlays of the SMI trust fund can’t so easily be compared. The beneficiary and government contributions are automatically increased to match the costs. So in one sense, the benefits will always be fully funded and there is no unfunded obligation because the premiums will automatically rise to cover the cost. However, the premiums will eventually be so large, as a percentage of retiree income, that there is a real concern that the contributions won’t be affordable for either the government or the beneficiaries in the long run.
The immediate concern is for Part A, which covers the hospital bills. As with Social Security, Part A is financed through employer/ employee taxes. The 75-year outlook for Part A is much worse than for Social Security. The average income under the intermediate assumptions is 3.39 percent of payroll, or a little more than half of the expenditures of 6.48 percent.
Current Medicare legislation requires that the trustees report to Congress when they project that the “excess general funding” will exceed 45 percent within seven years. The “excess general funding” is the difference between program outlays for Parts A, B, and D, and dedicated financing sources. In 2005, the trustees determined that the excess is projected to be more than 45 percent in 2012. But the arcane details of this provision won’t require the trustees to formally report concerns about the excess general funding point until 2007, and even then the only requirement is that the president provide a plan to deal with this difference. The excess general revenue funding provision does not require Congress to take any action to correct the problem. Congress and the president need to consider only what action might be appropriate, and, since 2007 is an election year, the chances of corrective legislation in 2007 are quite slim.
One measure of the cost of the total Medicare program is the projected Medicare expenses as a percentage of the GDP. The 2005 trustees report estimates that the total Medicare expenditures will be 3.5 percent of GDP in 2006 when the prescription drugs are added to the program, increasing to almost 14 percent of GDP in 2078. The panel was provided with projections that showed the total health expenditures, including Medicare, would be 40 percent of GDP in 2079.
Parts B and D are financed after retirement by the government and the retiree. The 2005 trustees report shows that premiums are expected to grow by 5 percent a year compared with less than 4 percent a year for Social Security benefits. As a result, premiums will continually increase as a percentage of Social Security benefits. By 2070, the trustees expect the total Parts B and D premium paid by beneficiaries will be 30 percent of the average Social Security benefit. On top of this, the out-of-pocket expenditures for benefits covered by Parts B and D will be 50 percent of the average Social Security benefit. The effect on any individual can be much different from the average, but the fact remains that with no change in Medicare, a very large share of Social Security benefits will have to be paid for premiums and the beneficiaries’ share of health care expenses.
Note: The 2005 trustees report and the 2004 technical report can be found at www.cms.hhs.gov/publications/trusteesreport.
EDWIN C. HUSTEAD is senior vice president and director of the Hay Group in Arlington, Va.
Contingencies (ISSN 1048-9851) is published by the American Academy of Actuaries, 1100 17th St. NW, 7th floor, Washington, DC 20036. The basic annual subscription rate is included in Academy dues. The nonmember rate is $24. Periodicals postage paid at Washington, DC, and at additional mailing offices. BPA circulation audited.
This article may not be reproduced in whole or in part without written permission of the publisher. Opinions expressed in signed articles are those of the author and do not necessarily reflect official policy of the American Academy of Actuaries.
Up To Code: