December 15, 2003 | Lecture on Health Care
ROBERT E. MOFFIT: Twenty minutes ago the President of the United States signed the Medicare bill into law. The debate about what the bill says is over. It's now what the law says, and that conversation will continue.
Today we're privileged to have with us the director of the Congressional Budget Office, Douglas Holtz-Eakin. Dr. Holtz-Eakin was appointed on February 5, 2003. He served previously as chief economist for the President's Council of Economic Advisers. He also was a senior staff economist there. He represents the Congressional Budget Office on the Federal Accounting Standards Advisory Board.
Doug Holtz-Eakin is also Trustee Professor of Economics at the Maxwell School at Syracuse University, where he served as chairman of the department of Economics and the associate director of the Center for Policy Research. He is the author of numerous articles in journals and professional studies.
Our second speaker, who will comment on what the director has to say, is Jeff Lemieux. Jeff Lemieux is the executive director of Centrists.Org. It's a nonpartisan think tank dedicated to health care policy, economic policy, and budget discipline, among other things. Jeff is also the senior economist of the Progressive Policy Institute, which is a center-left think tank here in Washington that is associated with the Democratic Leadership Council.
Before joining PPI, Jeff was the staff economist for the National Bipartisan Commission on the Future of Medicare, which was chaired by Senator John Breaux of Louisiana and Congressman Bill Thomas of California. Prior to his service with the Bipartisan Commission, Jeff was also a principal analyst at the Congressional Budget Office. At CBO Jeff was in charge of estimating the cost of national health care reform plans and the impact of Medicare changes. Between 1990 and 1992, he served as a member of the Office of the Actuary of what was then the Health Care Financing Administration, the agency, now called CMS, that runs Medicare.
Robert E. Moffit is Director of the Center for Health Policy Studies at The Heritage Foundation.
DR. HOLTZ-EAKIN: Thank you, Bob. The Medicare prescription drug bill is now the law of the land, but I would like to begin my remarks by at least metaphorically turning back the clock to a time prior to today's signing and sketching the setting into which this bill and the discussion about the bill was launched, as well as the setting in which it will be implemented.
That's a world in which seniors, or Medicare-aged individuals, will spend about $1.8 trillion on prescription drugs over the 10-year budget window between 2004 and 2013, and over the more narrow period when the bill is actually implemented, 2006 on, about $1.5 trillion. This amount constitutes about 50 percent of current Medicare outlays. It's a big number and it will grow quite rapidly. Indeed, it has grown on average about 9 percent per year.
Those seniors receive their coverage in a variety of ways. CBO's estimates, which were prepared for testimony before the Ways and Means Committee earlier this year, indicated that about 25 percent of all seniors did not have prescription drug coverage of any kind. That percentage represents about 10 million people. However, about 75 percent of all seniors did have some sort of coverage for their prescription drug costs. About 30 percent of that coverage came from employer-sponsored plans, about 16 percent from Medicaid, about 12 percent from Medigap, and the residual from a variety of federal or state and local plans for which seniors were eligible.
Looking across that population, about 90 percent of all seniors filled some sort of prescription for drugs in the year 2000, according to data from the Medicare Current Beneficiaries Survey. So about 90 percent of seniors had some prescription drug costs. These seniors spent an average of $1,500 on prescription drugs. Given the prescription drug coverages that were available, one of the interesting things that the data highlights is the absence of any particular relationship between the scale of out-of-pocket costs and the income of seniors. Looking at the flip side of that, there was no particular systematic relationship between income levels and the degree to which individuals had coverage for their prescription drug costs.
I'm going to spend a fair amount of time today talking about the costs of the prescription drug bill and what is now likely the future of the law. But before doing so, I think it's fair to say that Congress enacts programs for their benefits, and that it passes budgets to hold the line on costs. As the director of the CBO, my staff and I spend an enormous amount of time talking about the costs. But in many cases it's useful to reflect on the benefits.
I would have hoped by this time in the debate over this legislation to have some clean message for you about the benefits. Instead, I've come to a simple conclusion after listening to people talk about this legislation. This bill is the Rorschach test for the future of health care policy in the United States. Listen to any group talk about this law and you hear widely divergent opinions about the benefits that will be embodied in it as we go forward.
Some examples stand out. On demonstration projects for PPOs, is this the end of Medicare as we know it? Or this is a desirable aspect of competition introduced into a government program? Or perhaps it's just an ineffective fig leaf adopted and put into the bill.
On the income testing of the Part B premium, is this the end of a universal benefit for Medicare? Or is this too little too late? Or is it just perhaps the beginning, the leading edge of some effective cost controls in the Medicare program?
On the coverage of low-income dual-eligibles between Medicare and Medicaid, one view I've heard is that this is more effective targeting for low-income beneficiaries. At the other end, I've heard that this is simply an undesirable federal takeover of a state program and an expansion of federal obligations.
For the private delivery of the prescription drug benefit, a topic that CBO studied at length, is this simply the sensible application of market principles to the delivery of a government program? Or is this a risky venture into the unknown with unknown consequences for beneficiaries?
Another area is the use of the true out-of-pocket notion in the delivery of the prescription drug benefit. One view is that this represents a sensible targeting of the benefit to those individuals with a genuine need, or with real out-of-pocket expenses. At the other end, it's been interpreted as just a clear incentive to drop existing coverage by employers and other groups and, as a result, is an undesirable feature of the plan.
Finally, with regard to the doughnut hole, the famed section of the bill where there is no copay by the government, is this a good thing because it allowed the benefit to be front loaded and will attract more seniors into the program, thus mitigating any selection problems in the delivery of a prescription drug benefit? Or is this simply a bad thing and a way to fit a $1 trillion benefit into a $400 billion budget? Or perhaps it's just an insurance product that has never been seen running loose in nature and, as a result, has no merits at all.
My main message today, however, is to simply sketch a little bit the likely costs, and particularly the federal budget costs, of the legislation as enacted. I hope everyone in this room has memorized a single number: $395 billion, the CBO's estimate of the 10-year cost of this bill. This cost represents not really a 10-year cost, but because the benefit doesn't begin until 2006 and ramps up from about $26 billion then to about $75 billion by the end, it's closer to an 8-year budget cost. However, I think the way that cost is configured is of some interest.
In many of the iterations of the bill we saw roughly the same kind of structure and the same spending target, or about $400 billion on prescription drugs. For providers, roughly break even while providing more generous reimbursements in some cases but paying for them by recapturing costs, in particular the reimbursements for outpatient drugs in oncology and some other areas of the current Medicare program.
With regard to the delivery of the Part A and Part B benefits, spend some money to increase the outlays for private plans, now Medicare Advantage, the PPOs and HMOs that are out there on the landscape, and balance that with some higher recapture of costs through a greater level of the Part B deductible, indexing that deductible for inflation and introducing for the first time an income testing of the Part B premium. That will affect only those with more than $80,000 of income and is really a small fraction of the population, about 3 percent of beneficiaries. Roughly speaking, that structure is now built into the bill, and the question is how it will evolve going forward.
The first thing to note is that the bill arrives at roughly the same time that the baby boom generation is beginning to enter into its retirement. The CBO projected last summer that the growth in the number of Medicare beneficiaries will rise from about 1 1/2 percent a year between 2005 and 2008. Between 2009 and 2013 that growth doubles to 3 percent. In the years thereafter we see the steady retirement of the baby boom generation. So from the point of view of demographics, one would expect that the out year costs of the bill would rise as the number of beneficiaries increases as well. Indeed, if one takes into account the rise in prescription drug costs at historic rates and the increased number of beneficiaries, it's easy to expect this bill to cost over $1 trillion in the second 10 years and perhaps approach $2 trillion in 10-year costs between 2014 and 2023. In rough orders of magnitude, that would suggest the bill is at about $190 billion in 2023.
There are some wild cards in any long-term forecast and the CBO finds it useful in guiding Congress to provide long-term outlooks as a rough indicator of the way the compass is pointing from a fiscal point of view. Those wild cards usually take two forms. One set of wild cards are technical wild cards, economic wild cards, the sheer difficulty of looking into the future and trying to get a handle on the overall cost of a bill or the path of the economy.
In this case the most relevant wild card is the relative pace in the growth of prescription drug costs. As I mentioned earlier, drug prices and spending in particular, the combination of prices and utilization, have risen faster than health care costs as a whole. For those familiar with the area, health care costs as a whole are rising much faster than the economy. So we have not just the usual story in which health care costs rise quickly, but one where the particular kind of health care costs, prescription drug spending, are rising even faster than that.
In our long-term projections we take the view that the growth rate of prescription drug costs is about 3 percentage points faster than health care spending as a whole, but then over the next 25 years it ramps down to about 1 percentage point faster than health care as a whole. That's a forecast constructed on the grounds that trends that cannot continue, won't. But we have no particular information about the degree to which that rate will come down or the timing of the decrease. Therefore, it's sensible to assess the future with some notion of risk, and an upper bound to that risk would be to assume that the drug costs continue to rise at 3 percentage points faster than health care spending as a whole and thus consume a greater fraction of our economic resources.
On the flip side, there also are policy risks. Although the bill has now been the law for something near an hour, it may not in fact survive in this form. Congress and the Administration may revisit it at some point in the future. If they do so, it will affect the cost of the bill going forward and the command of economic resources in the economy
The most obvious and the simplest way to get a handle on the potential cost differences is to imagine filling in the doughnut hole, or the gap where there is literally an end to the government's share of the costs prior to hitting the catastrophic limit. Closing the doughnut hole has been widely discussed as a possible future change in the bill. So in thinking about the future, one has to think about the degree to which one fills in the doughnut hole or adds to the coverage in some way.
I asked the CBO staff to crank up a spreadsheet and give me a rough estimate of what closing the doughnut hole might cost the federal government on the Medicare prescription drug benefit. Now, there's an important question you have to ask yourself when you do that. Would it be the case if you filled in the doughnut hole that you kept the current split in cost, 75 percent government subsidy, 25 percent in premiums to beneficiaries? If so, filling in the doughnut hole comes with an increase in premiums and we would not see the estimated $35 premium that you've seen quoted regarding the bill as currently configured. Rather, you'd see a much higher premium if that were the way it was done.
An alternative of course is that one could fill in the doughnut hole or otherwise enrich the benefit and not adhere to that 25-75 split. If so, the premium would remain at an estimated $35 and then rise thereafter, and the government's tab would go up correspondingly.
In the numbers that I'm going to talk about, we assumed that we maintain the current structure of the bill, 75 percent subsidy, 25 percent picked up by premiums. As a result, to the extent that either we're wrong on technical reasons and drug care costs are higher than anticipated, or policy events overtake us and the bill turns out to be richer than as it was currently passed, in both cases, the premium is expected to rise and the government's share of the bill would be correspondingly smaller.
The Range of
What does that look like from the point of view of the possible outlays in the future? One scenario would be that we keep the bill as it's configured and we take a look at the drug costs. If we've guessed it right, drug costs will ramp down from 3 to 1 percentage point faster annual growth. If so, this bill ends up at something like $190 billion in 2023 and is on track to be something like 1 percent of GDP out at 2050.
At the other extreme it could be the case that we simply have constant growth in prescription drug prices, something rising at 3 percentage points a year faster than health care as a whole, and the prescription drug benefit becomes richer. In this particular example we fill in the doughnut hole in the benefit. If so, in 2023 this bill would cost in the neighborhood of $360 billion and it will be on track to be about 4 1/2 percent of GDP by 2050. In between those two estimates lies an enormous range where the actual cost of this bill could reside due to an economic future or a policy future that is different than what is embodied in the law.
Those are the extremes. You can imagine configurations in between. One could imagine we get the technicals right but not the policy changes. If so, it's something like a $320 billion benefit in 2023 and we're on track for something like 2 percent of GDP in 2050. Or you could do the reverse combination and say the policy is not going to change but the economic or drug price future is not what we anticipate, in which case it's not quite as expensive in 2023--on the order of $200 billion--and it's only about 2 1/2 percent of GDP in 2050.
Those kinds of numbers suggest that two bottom lines are relevant from the point of view of thinking about the fiscal future of the United States and for the bill in isolation. First, this is a bill with not very well-known outyear costs and not very well-known outyear benefits. Indeed, as I mentioned at the outset, there's a tremendous disagreement about the degree to which this will be a future of one type or another from a policy perspective.
To my eye at least, the growth of information about this law and individual evaluations of costs and the benefits means that it is highly unlikely that we'll end up exactly on the course that we have set today
Second, the potential for large outyear costs and the tendency of fiscal policy to emphasize more spending means that once that money is spent you'll have to address how best to finance the spending. These financing debates will press particularly hard as we discuss how to accommodate the large rise in entitlement programs more generally--Social Security, Medicare, and Medicaid--into the fiscal policy of the United States. The funding pressures of these other programs adds to the particular need to take a good look at the track of these programs as we move past the year 2014 and out into the future to 2030, 2040, and 2050.
Douglas Holtz-Eakin is the director of the Congressional Budget Office.
I've been calculating what happens to total Medicare spending when this new law is implemented, assuming the doughnut hole is not filled in. Under my calculations, based on the CBO baseline, I get to about 1 percent of GDP as the long-term cost of this bill once the baby boomers are fully retired or pretty much fully retired by 2030. That's a number that I think makes a lot of sense even though it is a highly uncertain estimate.
What does this mean? There are four big federal government entitlements: Social Security, which is expected to increase by about 2 percent of GDP as the baby boomers retire; Medicare, which after this drug benefit is expected to increase by over 3 percent of GDP as the baby boomers retire, closer to 4, actually; Medicaid, which because it pays for nursing home care for beneficiaries who are poor or who have divested themselves of assets to qualify for Medicaid coverage, is expected to increase especially rapidly as the baby boomers themselves actually get old; and, fourth, interest on the public debt.
The current improvement in interest on the public debt over the last 5 years or so has been very impressive, but now we're back in deficits and the national debt is growing. The deficit is expected to be over 4 percent of GDP next year. If we continue to add deficits at that magnitude for the next decade and then the baby boomers retire, the interest cost of paying investors for loaning money to the federal government would eventually become the largest entitlement on this trajectory, about 8 percent of GDP by 2030, larger even than Social Security or Medicare. The entire cost of these four entitlements would exceed the current federal spending budget, which is about 21 percent of GDP, and would exceed current federal revenues, about 17 percent of GDP, by a large amount.
Why does it make sense to project such a calamitous budgetary situation that far out into the future? Two reasons. The first reason is that even though, like Doug said, things that can't go on forever don't, it makes sense to project spending on its current trajectory so that policymakers will have an idea what they're up against. It's true that we're unlikely to allow interest, Social Security, Medicare, and Medicaid to be 23 percent of GDP in the year 2030. Something will give. On the other hand, if that's the trajectory, it's important for policymakers to know so that they can begin to think about reforms and other action.
The second reason why thinking about long-term costs is important is the case of Medicare: I don't see transformation in this bill. I was looking at my newspaper clips yesterday, and the L.A. Times had a story on the Web last night that described the Medicare bill as a "landmark overhaul" of Medicare. I had to scratch my head at that because I really saw quite the opposite in this bill. In a sense, this was the sort of bill that Medicare reformers in 1998 wanted to make sure never happened again. When the Balanced Budget Act in 1997 passed with hundreds and hundreds of lines of provisions and very little time for CBO and other policy analysts to really figure out how it would all work, a policy mess resulted. There were plenty of mistakes in both the drafting of the bill and the estimates of the bill (some of them that I made and my other colleagues at CBO made others). Staff wrote things into the law that had unintended consequences. For example, the law was supposed to improve private plan participation in Medicare, but it ended up cutting it by not quite half.
As I look forward at this bill, it seems like history is repeating itself. It's not really what I would call a transformation or a true overhaul of Medicare. I think it's very important for the media to reflect on what's going on here and decide whether this is a transformation or just business as usual.
As I look down the possible avenues of transformation, chronic care looms very large. Chronic care is what the clinicians tell us is the future of improving health care quality and possibly saving money in the health care system. There are a few small provisions in this bill on this part of Medicare's future. But I really don't see a chronic care transformation here. With a stand-alone drug benefit, any time you parcel out benefits into subcategories that don't necessarily talk to one another, that works directly against improved chronic care.
What does this bill do for competition? This bill has some things that improve competition. It would hopefully resuscitate the HMO sector in Medicare and also provide room for a PPO sector in Medicare, more room than is currently there. It would create a bidding system, which none of us fully understands just yet.
But the competitive aspects of this bill, (the demonstrations that go into effect in 2010), are not really enough to justify calling the bill a pro-competition landmark or an overhaul of Medicare. Maybe a few precursor steps, but not transformation.
So my first conclusion is that this is not reform and that it is business as usual. What's especially distressing is that the last time we did a bill like this in 1997, Congress was somewhat apologetic. Congress said we should have a Medicare commission to figure out how Medicare should really be governed, not by this catchall of hundreds of congressional provisions which we don't really know are in the bill until after it's been enacted.
This time there has been no apology. I think with the Balanced Budget Act and with this bill there is almost a comfort level that's been found by Congress in creating this sort of congressional micromanagement of Medicare, and that worries me a little bit.
The second conclusion is that the lessons of health reform and of the BBA exercises in both 1995 and 1997 haven't been lost on our legislators. These episodes apparently teach that it's best to leave the analysis until after the votes are held. In the early 1990s, when we were talking about health reform, there were perhaps 10 CBO reports on various aspects of health reform before any legislation even moved to committee. In addition, there were outside analyses from esteemed estimators, such as the Lewin Group, and from the Administration. There was a lot to look at when people were thinking about whether to do those health reforms.
After the early 1990s, however, analysis increasingly took a back seat to political goals. The next big health legislation was the Balanced Budget Act of 1995, which went forward with far less analysis. Even so, people had an idea that it was a shift toward a defined contribution system in Medicare and it didn't pass. In 1997 there was even less analysis, as CBO was really scrambling just to get the basic estimates done in time.
I think the conclusion is now that we really don't want to see too much analysis from CBO and others because it could hurt the legislative prospects of such a complicated piece of legislation, which relegates CBO to just explaining what's in the bills. The cost estimates that came out in July were very helpful explaining what was in the House and Senate bills.
CBO's other roles include trying to create an economic perspective. If you look at a Congressional Research Service summary of a bill you will find excellent descriptions of a bill's contents, but no analysis of how the bill would affect, for example, the economy or an industry. It's up to economists like those at CBO to tell us something about the larger context of legislation. In regard to the recent prescription drug legislation, CBO should have spent more time answering questions like, What does this mean in the long run? What does this mean in the larger sense of health care in this country?
More generally, how well would it work? The whole subject of the law's likely effectiveness has been something that CBO hasn't had time to work on, but hopefully CBO analysts will do so over the holidays before they get wrapped up in other things. They need to focus on whether or not this bill will work as drafted, if there are additional technical changes that need to be made in the law, if there are particular regulatory approaches that should be taken to make the thing work, and so on. The larger analytical community needs to analyze the political feasibility, which is something that CBO cannot really do, but the rest of the analytic community can decide whether or not these things are politically feasible.
So those are my two basic conclusions: First, it's not transformation; therefore, it will cost a lot. Second, we're on a legislative path of leaving the analysis until after the votes are had, which has its ups and downs. It may be easier to get legislation that way, but we certainly don't know what we've got until after it's done and now it's all up to us to go back, read the bill extremely carefully, and work with CBO and the rest of the congressional agencies to figure out what we've got here and what we should look forward to in 2004 and beyond.
Jeff Lemieux is the executive director of Centrists.Org.