ROBERT E. MOFFIT: Today, on the eve of the anniversary of the Medicare program, we are in the midst of a great national debate over the future shape of Medicare. This debate will have a profound influence on the quality of life of everybody in this room. So, the question is: What will Medicare bring for the future, not only for senior citizens, but also for the rest of us as taxpayers and recipients of health care services?

With us today are two outstanding experts, intimately familiar with the Medicare program. Dr. Thomas Saving is a distinguished professor of economics at Texas A&M University and a senior fellow of the National Center for Policy Analysis. He is a public trustee of the Social Security and Medicare trust funds, and was also a member of the President's Commission on Social Security. He is the author of Medicare Reform: Issues and Answers and the co-author of The Economics of Medicare Reform.

Our other speaker is a great colleague. Jeff Lemieux is the senior economist with the Progressive Policy Institute and also executive director of a brand new think tank in Washington called Centrist.org. He is the author of a number of centrist proposals on health insurance coverage, Medicare reform, balanced budget issues, and entitlement spending. In the area of Medicare reform, Jeff was the staff economist for the National Bipartisan Commission on the Future of Medicare headed by Senator John Breaux of Louisiana and Congressman Bill Thomas of California.

From 1992 to 1998, Jeff also served as a principal analyst at the Congressional Budget Office (CBO), where he helped to estimate the cost of national health care reforms and the impact of Medicare reforms, including the Balanced Budget Act of 1997.

Before his service at CBO, Jeff served as an actuary at the Health Care Financing Administration. Previously, he served on the staff of the DRI/McGraw Hill economic forecasting firm.

Robert E. Moffit, Ph.D., is Director of the Center for Health Policy Studies at The Heritage Foundation.

THOMAS R. SAVING: Jeff and I are going to address different aspects of this issue. I am going to stress what the future of Medicare would look like as it is currently structured, and not how we might restructure it, although I have spent a lot of time working on that issue. I am going to give you an idea of the shortfalls implied by the current set of transfers to elderly citizens from workers.

One of the things that we always like to do, as you know if you read the trustees' report, is to present the shortfalls as a percent of GDP. As a trustee I have campaigned to present the shortfalls as a percent of federal income tax revenues because I do not believe that many people understand what percent of GDP really means. Certainly the 535 people on the Hill understand federal income tax revenues. So I have translated these numbers into a percent of federal income tax revenues, assuming that federal income tax revenues remain at a 50-year average, which is 11 percent of GDP.

You hear people questioning whether the recent tax cuts mean we have given away the store. Rather, if we look at the past 50 years, in spite of huge changes in the tax code, the share of federal income taxes in GDP has remained almost constant. It is really pretty much unaffected by changes in the tax code because reductions in marginal tax rates have been offset by the incentive to produce more. That is important.

The Future Tax Burden
One of the important things about financing these programs is getting people to work longer. By working longer, we would increase revenue while reducing the burden placed upon the system.

But that aside, expressing projected shortfalls in Social Security and Medicare as a percentage of federal income tax revenues gives us an idea of the potential impact of the funding shortfalls in these programs. By 2026, the year that we say the Hospital Insurance (HI) Trust Fund is going to become exhausted, we are going to be transferring almost 30 percent of all federal income tax revenues to Social Security and Medicare. (See Chart 1.)

When I was testifying today before the Senate Special Committee on Aging, I said that Social Security is the easy problem and we should go ahead and solve it. Medicare is the real serious problem. By 2042 (I chose 2026 and 2042 because those are the dates of so-called trust fund exhaustion), the shortfalls that we as trustees estimate imply that we are going to be transferring almost half of all federal income tax revenues to these two programs. That is an incredible change from the current situation where these two together are providing net revenue to the Treasury. For the case of Medicare alone, in 2026, one-fifth of all federal income tax revenues are going to go to Medicare, whereas currently it is about 7 percent. (See Chart 2.)

Let us now look at the way we run Medicare at the moment. Those of you who are in the Medicare business already know that Medicare is in deficit, because some 75 percent of Part B is already financed by a transfer from general revenues, and the HI surplus is not big enough to cover the Part B deficit. Now it is true that for the past two years the transfer has been bigger than 75 percent. It is 78 percent because we have underestimated pretty consistently the rate of growth of Part B expenditures. So we set the premium at the beginning of the year with an estimate of what those Part B expenditures are going to be and, as it turned out, they have been much bigger than that for two years in a row. That might suggest that there is something systematic going on that we would want to change.

But of course, premium adjustments do not make the elderly happy. They are likely to let their congressman know that they are unhappy. But further, by 2042, again, when the Social Security trust fund is supposedly exhausted, we will be transferring a third of all federal income tax revenues to Medicare.

Now, a transfer of that size is just not going to happen. And this transfer is without prescription drugs. So I have a graph about prescription drugs because that is the topic at issue right now. What this graph shows you is the per capita expenditures and the percent paid by third parties. As we have increased third-party payments for prescription drugs, per-capita consumption has increased dramatically. (See Chart 3.)

Thirty years ago, I would say let's take a look at the rate of growth of expenditures for physicians, hospitals, and prescription drugs. You could see hospitals grew the fastest. That is when something like 80 percent of hospital costs were paid by third parties. Physicians by that time had only maybe 40 percent and their costs grew much slower. Only about 5 or 6 percent of pharmaceuticals were paid by third parties, and they were actually declining in price.

That has all changed now because everything is pretty much third party-paid, and therefore, patients act as if it is all free.

Medicare currently pays for about 5 percent of prescription drugs. As you are aware, there is a very narrow range of prescription drugs that are reimbursed by Medicare Part A. Part B, of course, would be paid by Medigap and other things, to the extent that the patient has coverage.

Future Drug Spending
Based upon the recently passed Senate Medicare Bill, the CBO estimates Medicare would pay about 25 percent of prescription drug costs for seniors. However, when we tried to apply the technique used by CBO to the distribution of expenditures by seniors, we got a number more like 42 percent instead of 25 percent. However, I am going to use CBO's number because I have not yet resolved why our number is so much different than their number.

Assuming that the prescription drug bill that ultimately passes results in Medicare paying the CBO-estimated 25 percent of seniors' prescription bills, this would increase by 20 percent the transfer that we are going to have to make in 2026 from federal income tax revenues. If you were to do what the Democrats proposed last year, well over a third of all federal income tax revenues would have to be transferred just to Medicare.

By 2042, last year's Democratic proposal would result in well over half of all federal income tax revenues being used to cover the Medicare benefits. If you added Social Security to that, by 2026, you can see that you are already up to more than a third of all federal income tax revenues being transferred to Social Security and Medicare. So we are going to go from the two programs together making a positive contribution to Treasury revenues to taking 30 percent or a third of all federal income tax revenues.

Future Debt
Now, let us get to the issue of what we owe the current participants in these two programs. Something we have done for Social Security this year in the trustees' report, and I hope we will do for Medicare next year, is estimate the present value of the unfunded liability for what we refer to as the 100-year closed group, that is, all those currently 15 years and older. No matter what we do to try to reform the program, we have to deal with the debt that we owe the current participants. (See Chart 4.)

Let me begin by noting that if we passed the Senate prescription drug bill, we would add $2.6 trillion to the debt that we owe the current participants. Even without any prescription drug benefits, what we owe current participants in Medicare net of their Medicare tax and premium payments is $13 trillion. The prescription drug bill is going to add 20 percent to it--and remember, that $2.6 trillion is almost the size of the outstanding federal debt right now. So the prescription drug bill is terribly expensive in this perspective.

How about the new generations? How are those who start working tomorrow and after them going to pay off this debt? If the system really works in the long run, the 100-year closed debt is fully paid by the next generations. They come in and start paying taxes for a long time before they ever collect anything. So all those taxes that they pay are designed to pay off the debt owed to current participants in the program.

Well, I will give you an idea of how much future generations will contribute toward this debt. Instead of contributing anything to paying off the $13 trillion, the new generations are going to impose another $25 trillion in debt to the system. So they don't contribute anything to paying it off; they just add to the problem. The prescription drug part of that is another $4.9 trillion.

Between the two programs, that is Social Security and Medicare, we owe current participants about $24.4 trillion. Adding the prescription drug provision to that will make it $27 trillion. The future Medicare participants starting tomorrow will cost us $25.3 trillion if we don't have prescription drugs. If we do, that is an additional $4.9 trillion. So we are looking at $49.6 trillion of debt for the system as a whole as it is currently constituted, and $57 trillion if we add prescription drugs. (See Chart 5.)

Medicare is a system out of control. Of course, we all knew that. The way Medicare is set up and the growing proportion of per capita income it is using pose serious problems in financing it. I am not one of these people who argues that the percentage of GDP that Canada and Germany consume in health care is the nirvana of health care. I don't think that is right.

There is a reason why health care for the elderly especially is rising as a percent of their income, and we need to understand that reason. And then we somehow have to make incentives matter.

One of the things that the Centers for Medicare & Medicaid Services (CMS) has done is to try to control cheating in Medicare. There is a lot of health care fraud. So we spent some time on this and discovered, yes, there was systematic upcoding; meaning that physicians would do one thing for you, but put down a code for payment that is related to that but one step up, so they could get more money back.

So we have a bunch of policemen watching what is going on. If consumers cared what it costs, they would be doing the checking for upcoding. When your physician gave you a shot for influenza, and said he gave you a pneumonia shot that costs twice as much, you would complain about being double-charged.

But until that happens, we cannot control these costs. It is our job to find a way to make people care what it costs. If we cannot do that, we cannot solve this problem. We are going to have to solve this problem because we are not going to find this money. That means we are going to have to take away health care in some way, or ration it seriously--which is what other countries do, as you know--or we will have to find another way to reform it.

Jeff is going to speak about that challenge and how we might meet it. Thank you.

JEFF LEMIEUX: When the Medicare Commission tried to find solutions in 1998 and 1999, they were primarily motivated by graphs like those shown by Dr. Saving. The idea was that Medicare is going to take up a larger and larger percent of GDP, and that it is unlikely that we are going to be able to find revenues to pay for that. So we have to find some way to slow down the growth of this program spending, hopefully without deep and draconian benefit cuts. Control the costs and ideally create better value.

In the past five years, scholars in the health care area have also taken a new direction. They have started to realize that if we really want to save the U.S. health care system, if we really want to try and get better value in health care, we need to stop keeping the health system oriented toward what is called acute care.

In other words, doctors are patching us up when we suddenly get sick or are injured and then sending us back on our way. Our system does that very well. But we need to pay more attention to what happens to people with long-term or chronic illnesses, especially between hospitalizations, between episodes of sudden illness or injury.

There are some good reasons why we should think about chronic care as we move forward with the House and Senate Medicare bills.

The Medicare Bills
Now, here is a four-bullet description of what is going on with the House and Senate Medicare bills. In 2004, we have a discount card. This should help a lot of seniors. There is some low-income assistance attached to those discount cards. Again, very nice. Low-income seniors especially need extra help purchasing prescriptions when they don't have other coverage.

We also have an effort starting in 2004 to re-establish private plan options. I think this is a very good idea. In the long run, the hope is that competition between Medicare's fee-for-service system and private plans is the best hope for slowing the growth of spending without draconian benefit cuts, especially as Medicare grows more capable of making decisions internally and not relying on congressional laws and congressional micromanagement.

Many Democrats don't really like the idea of moving in 2010 toward a federal employees-style premium competition. But my calculations show that the risks of that to the fee-for-service program are actually very low, and that this could be a very good thing. My only regret on the 2010 premium formula is that the government-run program probably needs more flexibility if we are going to ask it to compete directly with private plans.

Then we have got the big wart on the face of this thing. The 2006 stand-alone drug benefit is something that no economist and few health policy analysts would have designed. It looks like just the sort of thing that is designed for political reasons, but which analysis and economic thinking has not played into very much at all. It's ugly.

How did we get to this point? Well, for two straight elections, the House Republicans needed some sort of decent-sounding drug benefit to pass so that they could go home and claim they were for drug benefits. They did not really care if it worked; they were trying to develop something that sounded as good as possible in the newspaper without really worrying about the consequences of how it would work if it was actually implemented.

Then the Senate tried some other approaches, but they never really got CBO and the other congressional support agencies excited about Federal Employees Health Benefits Program-style premiums and figuring out really what the risks and the rewards were. Nor did we ever have much congressional support for Medigap options adding drug benefits for people, defining contribution drug subsidies, or catastrophic low-income plans.

So when the President and the congressional leaders decided they wanted a Medicare bill, the only thing that was really teed up were these bills that were patterned after the House Republican bills: a stand-alone, high-premium drug benefit, where if you have drug spending that is zero, then you would be asked to pay a premium of about $420.

As your drug spending rises, your premium would stay the same, your benefits would gradually start to grow. If your drug spending exceeded about $1,000, in the House plan, you would hit the break-even point, actually between $750 and $1,000. What this would do, all other things being equal, is cause people who expect their drug spending to be very low to not bother. People with drug spending that is very high would say, "Yes, I need to do this."

The Senate bill is a little bit different. The break-even point is a little bit higher and the mechanics are a little different. The premium is about the same, but it is the same sort of plan. If you have very little drug spending, you might not want to buy this. If you have higher drug spending, you would.

All other things being equal, what Congress will have to do to maintain a good enrollment in this is to add money, to make sure that virtually everyone is enrolled so that the premium doesn't start to skyrocket, because people with low drug spending don't take it and people with high drug do.

I mean, the logical course of action for this sort of drug benefit is to increase the benefit package, especially to fill in the so-called doughnut hole in the middle of the benefit package where you don't get any benefits until your catastrophic benefit kicks in, and also to keep the premium low. If you do these two things, then you can get high enrollment and the whole system could work.

What Congress has done instead is to impose a late enrollment penalty. They have said we are not going to worry about adverse selection. We know our drug benefit is not that good and the premium is high and a lot of people are not going to want to buy it, but we are going to force them to buy it. We are going to put in a penalty that says, "If you don't take this benefit now, you will pay a higher premium if you have to get into it later."

Moreover, they have created incentives in here for employers to force enrollees in retiree health coverage plans to pick up this new benefit.

A Politically Designed Drug Benefit
This system was never designed to work; it was designed for political appearances. To make this sort of drug benefit work it is going to need more money or more government control. And to be honest, there are a lot of Republicans who really don't care if it is going to work. The idea is to neutralize yet another issue for the next election.

By the same token, there are a lot of Democrats who hope it fails. Their idea is if it doesn't work very well, then great: We will get more government control and more money in the system and we will end up with a trillion-dollar drug benefit just like we wanted all along.

Here is what I have computed. It is a different approach from Dr. Saving's presentation. I took the CBO estimate of the Senate bill, and it works out virtually the same if you take the CBO estimate of the House bill. I took a baseline of Medicare spending as a percent of GDP and then added in the drug benefit based on the CBO cost estimate, and extended that for an additional 17 or 18 years out to 2030.

The CBO estimates, if you look at them very carefully, show not only a big jump up in Medicare spending when the drug benefit is phased in, but they also show a permanent accelerated growth of Medicare spending after that point. It is not a huge acceleration--it is about a quarter of 1 percent a year--but it is enough to make a big difference over time.

In both the House and the Senate bills, the CBO projects a permanent acceleration of growth. If you extend this out to 2030, you will end up having spent an extra 1 percent of GDP on this drug benefit, compared with what the Medicare baseline would have been without it.

Now where is the conference committee likely to go with this? I don't think anybody knows. These are two bills that are going to be very difficult to reconcile. I think that it is very unlikely that the House will agree to the Senate bill and allow for a compromise based on the Senate bill that includes a lot of House Democrats voting for the compromise and a lot of House conservatives voting against it.

I think it is also pretty unlikely that the Senate will agree to the House bill. That would require a couple of conservative Democrats to go with a solid Republican majority, and they have declared they want a bipartisan approach instead.

So it seems to me the only thing they could do to get this bill out of conference is to go through item-by-item and try and split the difference. My hunch at this point is that it will be very difficult to do this, but if they do, the drug benefit will probably look a little bit more like the House's drug benefit, and that the Senate's fallback position will be sort of melded with the House's.

The House members will say that they have a fallback position: It doesn't really have direct government control, but it has 99.9 percent government financial backing, which is virtually the same thing. Then, finally, the premium formulas that would start in 2010 will probably be diluted down to a demonstration, which might be very interesting. But again that delays the hope of trying to create a competitive system in Medicare that could save some money over the long haul.

A Better Course of Action
Here is what I think should be done in a perfect world. In 2004, really try and get this drug discount card up and running and make it as universal as possible, so that seniors can get at least the promise of not being ripped off at the pharmacy. They are at least getting some discount, some group purchasing is helping them get a decent rate on drug purchases. Of course, beef up the low-income benefits.

In 2006, instead of the stand-alone high-premium drug benefit that has all these problems and requires all these contortions, it would be much more stable and more controllable in the long run to just add a catastrophic benefit to those discount cards. There is no reason to charge a premium.

What would have to happen is that members of Congress and the President would have to go back to people and say, "Look, we tried to create a full drug benefit. We went through all these contortions, and it probably will not work, so we are just going to have to back off and do a catastrophic benefit instead."

Given the budget situation over the next five years, we are looking at very deep deficits persisting. Therefore, many liberals who currently vilify the premium support FEHB system are going to come back to the idea that maybe it is not such a bad deal after all; that if we analyze it very carefully, if we work on it very hard, we can make sure that it will not threaten Medicare's fee-for-service program.

We can actually use a premium support-style system to help get the fee-for-service plan modernized in terms of its internal workings, and allow people at CMS to actually run that program more like a business for the betterment of seniors' health care and for the betterment of chronic care.

Then the last thing on my list. In spite of whatever we do, we still have to pray for the next generations. The federal budget has swung from surpluses into deficits, where revenues have fallen from 20-and-over percent of GDP down to an expected 16.0 percent next year according to the President's mid-session budget.

In the meantime, spending, based on pressures from defense and other areas, including Medicare, has now risen above 20 percent of GDP, and we have deficits of about 4 percent of GDP.

This isn't going to go away. Based on my best projections of a bounce-back in revenue as a percent of GDP, with the stock market returning to some of its vigor, and capital gains and bonuses coming back into the economy, we are still looking at an enormous gap between spending and revenues.

When you extend this out even further to 2030, we are looking at something that is economically untenable. As this starts to hit home, perhaps as early as 2005 during deliberation of the Budget Act of 2005, legislators on both sides of the aisle are going to have to come back together and figure out how to be a little less partisan in all these things and how to work on entitlement reform in such a way that it doesn't seem so threatening to liberals.

Here is what I expect in 2005. Essentially, the era of easy choices ends. We have been cutting taxes and raising spending now for several years without any attention to the consequences. We can say, well, we are trying to fight recession and do other things, and that is all fine. But we have also done long-run things that are going to cause budget deficits to persist throughout the decade. The mess hits the fan in 2005, after the next presidential election.

We have sunset provisions on tax cuts that are going to have to be addressed. We have the entitlement programs continuing to increase. We have appropriations problems. We have a whole host of things that are going to cause the Budget Act of 2005 to be a really important watershed.

At that point, on the issue of Medicare, the whole idea of premium support, a federal employee-style system is going to come back. Liberals are going to say this might be better than flat-out cuts, as we are looking forward 10 or 20 years. Hopefully, conservatives will be able to find ways to work with liberals and make them feel more at ease with competitive systems. Nobody thinks that revenues are going to be able to grow to 25 percent or higher of GDP to pay for the spending that we are to incur.

I hope that the Medicare conference comes to this improved drug benefit, compared to the ugly one that they have scheduled for 2006. Other elements of the drug bill and the Medicare bill are fairly commendable. It is a big question whether or not the legislators will be able to pull this out at the end of the year.

DR. MOFFIT: Thank you very much, Jeff and Tom. Ladies and gentlemen, it is time for your questions.

SPEAKER: Jeff, the Senate version closes the "doughnut hole," so to speak. Don't you think politically, they might shift to the Senate side because it closes that hole a bit better?

MR. LEMIEUX: Well, my rationale is less logical than that. The main flashpoint is going to be the premium support formulas in 2010. If it looks like the House is going to have to make a big give on that and turn it into a demonstration program, then it seems to me that the House gets its way on a lot of other things, including the structure of the drug benefit.

The structure of the House drug benefit, with bigger up-front benefits and a larger doughnut hole is probably more appropriate if you are trying to do a stand-alone high-premium drug benefit and entice lots of people to join up. So for both of those reasons, it is just sort of a balance between House gives and Senate gives. If you are going to do things this way, it would help avert adverse selection a little bit, get a little higher participation. That was the reason for my hunch, not that it was better than the other.

SPEAKER: This is a question for Jeff. I don't understand the logic behind the claim that if we move to this competition approach, you are going to slow the increase in costs in the Medicare program. Marilyn Moon, who has looked comparatively at increases in costs in the private insurance market comparing it with Medicare, finds a slower rate of growth in the Medicare program over 30 years.

Everything that I have seen indicates that you have higher costs in the Medicare+Choice plans than in traditional Medicare. The recent CBO analysis says that you are going to increase costs for the Medicare Advantage portion of the reforms.

There is a claim on the one hand that you are going to save money by going to this approach, but then there are a whole lot of people who say, in the same breath, that you need to increase funding to these plans to get them in. I don't see how you can believe both those things at the same time.

MR. LEMIEUX: It is very easy to explain, actually. For all the good work Marilyn has done, her piece on comparing private health insurance premiums and Medicare costs over 30 years totally missed the point. Neither spending growth in Medicare nor in the private sector is exogenous, as economists would say.

Essentially, when Medicare starts to save a bit of money in some way or another, then employers put pressure on private insurers to save some money, too. When private insurers are saving a lot of money in some way or another, then Congress faces pressure to try and save money in Medicare.

Marilyn Moon said that Medicare grew slower over 30 years. Then, Joe Antos from the American Enterprise Institute did a study that said, "Yes, but its benefits have gone up, so if you weigh it in that sense, the private plans have been more efficient." There have been other studies that try to control for these things. It is really a lot of great work by really smart people on kind of a dumb question.

Now, to get to the second part of your question, what goes on with competition is more than what cost estimators can sometimes deal with. What CBO said about this bill is, if you pay more to private health plans, it will cost more. That makes a lot of sense.

What CBO did not say in this bill, but what a lot of economists believe, is that over time, if you create a competitive system, it will cause changes in behavior, both at the health plan level, at the administration of the program level, and at the consumer level. These changes have the potential--not the certainty, but the potential--to slow the rate of cost growth.

Economists have looked at systems like the federal employees' plan. This is hardly perfect in its design, but it does seem to have created a bit of market awareness, both at the plan level, government level, and the consumer level. This has helped keep its cost growth reasonably under control.

It is something of a leap of faith to assume that competition in this way is going to lead to a slowing of Medicare costs. There is no objective evidence that you can really say is definitive, but it does make a lot of economic sense to assume that this is something that we should try before we launch into benefit cuts.

SPEAKER: In terms of trying to figure out what the compromise might be on premium support, I don't see that a demonstration project is going to do it. We have already been through three demonstration projects on competitive pricing, and the politics trumps it. Do you have any comment about that?

MR. LEMIEUX: Some of the competitive demonstrations that were proposed in the past had to do with private health plans competing against one another, but there would not be any change in the fee-for-service premium. So this would be a little different, in that the fee-for-service would be part of the mix. So it would seem more fair from the private health plans' point of view, so they would not oppose it so bitterly.

Of course geographically, it is going to be hard to find a region to do a demonstration, the purpose of which is to force more competition and force more efficiencies. "Why are we forcing efficiencies into the system in my district and not the other district?"

So how you structure that demonstration, I really don't know. But there is a lot in this bill that, when you think through how it would actually work in real life, is kind of fanciful.

SPEAKER: I am Joe Antos, AEI, one of those not-so-smart people working on a dumb question.

I agree with Jeff. There really is a good study that addresses the real question fairly and directly by Michael O'Grady at the Joint Economic Committee. He did the most accurate comparison you can make that is policy-relevant.

He compared actual Federal Employees Health Benefit Program (FEHBP) performance with Medicare, and he also was able to tease out the drug components. It was really a very close comparison. What he found over the course of the past 15 or 20 years, FEHBP in fact did better than Medicare. So, I would recommend that particular study.

Now let me try to draw both Tom and Jeff out on something that they both said, but Tom said it very provocatively. That is, what are the elements that you see that would actually get consumers interested in the price of the product? We are all interested in the quality of the product, but we don't have any handle on that. But what about the price of the product? What are we going to do to get consumers involved?

DR. SAVING: I think that unless we do that, we are not going to succeed in any way. Part of what we can do in Medicare is to have a much better way to estimate the expected costs.

I think we can give people vouchers for what their expected expenditures are. We can allow the private insurers to bid for those people. There is no fixed capitation. It would not be based on age or other characteristics but on expected expenditures. So you are all alike. I think that is one approach.

We cannot do the same thing that General Motors does with cars; everyone pays the same. Well, why should they? Why should the young worker who has a much lower expected cost, pay the same as somebody who is chronically ill? We have to find a way to make that work. But until we do that, we are not going to succeed.

Then the second thing we have to do is to get rid of first-dollar coverage. As long as we have first-dollar coverage, customers are not going to care what it costs. If we are looking at prescription drugs, and if they are not expected cost, if the chronic people are not given a bigger voucher than everyone else, then no one else should get into the program. We have to price it right.

I know in health care we call that risk adjusting, because we don't use the word "risk" properly. Expected value, not risk, is what we are talking about here. Risk is something else, and we know that in every other area of economics except in health care. In health care, we call expected value "risk," whereas everywhere else we call it "expected value," and risks are uncertainty around that expected value.

You can separate risk from expected value, and you can buy risk insurance and it can be separately provided; it doesn't have to be provided by the health care industry. And it can be separately done.

For Medicare, of course, CMS can be that provider of risk insurance. But in the market for general health care providers, casualty providers can provide health care risk insurance. That would really allow someone to self-insure for the component that they now buy from a health care provider.

Instead of what we now see, you would see doctors, pharmaceuticals, and hospitals advertising the price. I am dreaming of the day when I am driving to Houston and I see the big sign that says: Come to M.D. Anderson, $99 a night, everything included.

Why don't CVS and Eckerd and Walgreens advertise Viagra at $5.95? Nobody advertises that. Why? Because the customers do not care what it costs. We have to make them care. Then for all the commonly used prescriptions, like the grocery store ad, they will be putting the ad in the paper and we will have competition.

We have to have a way to make that work. Part of it can be medical savings accounts, where people get to pocket the money at the end of the year. It has to be their money. They have to get to keep whatever they do not spend.

MR. LEMIEUX: There is a distinction. I think that the idea of trying to make consumers pay at the point of service--in the hospital, the doctor's office, or the pharmacy--it would in fact cause people to think carefully. But I think that a lot of moderates and liberals in particular worry that poor people will say, "I can't afford it," and wealthier people will say, "I can." Unless the voucher is just right, as Dr. Saving said, we could have some real problems.

A lot of moderates and also the Heritage Foundation (I don't want to accuse you of being moderate) have talked about price sensitivity at the point of purchasing insurance; not so much the point of entering the hospital or the pharmacy, but price sensitivity at the point when you make an insurance selection.

That insurance that you buy might be more or less generous, it might have fewer or greater numbers of restrictions. But that is sort of what the whole federal employees-style competition is about.

Even if people choose the type of insurance that does not cause them to have much immediate price sensitivity, hopefully their insurer is watching out for the prices that they face and trying to get good deals on their behalf. In the case of people with chronic conditions, some insurers are actually starting to learn how to help them take better care of themselves and save money in that manner.

So, rather than having consumers purchasing health services directly and seeing the prices on the billboards, to have an agent working on their behalf, which they choose and they face a financial penalty if they choose an inefficient agent--that is a good thing.

SPEAKER: Al Milliken, Washington Independent Writers. What kind of feedback are you hearing from consumers, particularly seniors? Do you see either party getting a clear political advantage in the short term, specifically in 2004?

MR. LEMIEUX: In the late 1980s, seniors who had good retiree health coverage rebelled over the Medicare Catastrophic Coverage Act. They didn't want to have to pay the extra surcharges that were involved with Medicare catastrophic and prescription drugs back then, because it would not give them anything. It was an extra fee without any benefit.

There is a similar dynamic in these bills. If you have good retiree coverage (perhaps you have steered your career toward a large firm or a government employer precisely so that you would have good retiree health benefits), you get very little out of this.

CBO said it flat-out: The employers have a disincentive to wrap around the government coverage and they have a strong incentive to force you to purchase the government coverage. CBO thinks that 32 to 37 percent of people with good retiree health coverage will lose it under these bills if they are enacted.

There is a constituency there for rebellion. I am not sure if it has gelled yet, or if it will, or if it would take time after enactment for people to figure out what is going on here. That is one of the reasons that I have described this benefit as ugly. It was intended to look good in the newspapers, but it has some unnoticed consequences that could hurt its workability in the future.

The way to get out of this problem, of course, is to add to the cost. The solution to all the problems with this type of drug benefit is to spend more money. That will be difficult in future years.

DR. SAVING: The firms themselves have a real issue, because retirement funding is now in a very difficult position. As you know, all these retirement programs are underfunded. These health care programs, with their rising costs, are also underfunded. I suspect the firms, politically, are going to be very much in favor of this kind of entitlement program because it is going to be a significant subsidy.

As everyone agrees, they are going to bail out of the health care business. It would be a good thing for them to do in the sense of portable health care. Get them out of the business, but we need tax code changes to accompany that.

DR. MOFFIT: I think that the rebellion is underway. It is rather significant that you have The Wall Street Journal and The Washington Post editorializing, within a period of a week of each other, to stop it. The National Association of Retired Federal Employees, the United Autoworkers Union, the AFL-CIO, and the Heritage Foundation are all saying this is not a good drug proposal for seniors and we ought to go back to the drawing board.

This is not the political slam-dunk that some of the political geniuses in this town thought it was a few weeks ago. The more that people actually see the details, and how it will affect them, you are going to see a backlash not dissimilar to the kind of backlash that took place in 1988, with the Medicare catastrophic.

SPEAKER: On that note, can this patient be saved? Is there anything that can be done to these two bills to make them a net-plus rather than a net-minus?

MR. LEMIEUX: Well, I have already said it, so I will just quickly repeat it and then let Dr. Saving have his say. There are lots of good things in these bills, and it is wonderful that there has been sort of a bipartisan approach in the Senate. It is nice to make some progress any time on health care issues that are usually stymied by ideological gridlock.

But this 2006 drug benefit is a real tough thing. It would just be so much easier if they would scale back the promise; and say, "Listen, we cannot afford to give you this drug benefit that we have been promising, but what we can afford to give you is free catastrophic coverage.

We can do that in such a way that it is controllable. It doesn't hurt retiree coverage; it is just less than what people are expecting. So, the politicians have to go back and scale it back. I think if they did that, they could come up with a system that makes a lot of sense.

DR. SAVING: I agree with some of that. The problem is that these bills started out as prescription drug bills instead of Medicare reform bills. What we need is Medicare reform that combines Part A and Part B, while including prescription drugs. And we need to go back to the drawing board.

This legislation is a very bad idea. The catastrophic part doesn't affect enough seniors. If you go back to just catastrophic, you only pick up some 15 percent of seniors.You want to pick up the voters and you don't get very many.

But the idea is to reform Medicare. We ought to get after that. Throw this thing out and go back to Medicare reform that would include prescription drugs.