Tax Analysts®Tax Analysts®

My Subscriptions:




Washington, D.C.

Friday, June 16, 2006



President, Tax Analysts


Member, President's Council of Economic Advisers


Special Assistant to President Clinton

for Economic Policy

Visiting Scholar, NYU's Wagner Graduate School of Public Service


Senior Fellow, Urban Institute
















* * * * *


(9:08 a.m.)

MR. BERGIN: I'm Chris Bergin, the president of Tax Analysts, which is the non-profit publisher of Tax Notes magazine, Tax Notes Today, State Tax Notes, Tax Notes International, and many other print and online tax publications. Welcome to the latest in our series of roundtable discussions on key issues in tax policy.

The topic for today is healthcare and the tax code. This is the second year that we have conducted a series of roundtable discussions. If you are new to our discussions, let me first say welcome. Let me also take just a little time to explain our process.

I will open things up with some brief remarks to introduce our topic. I will then introduce our distinguished panel of three speakers who are seated around the table here. Each of them will address aspects of our topic for about 10 minutes. After that, we will open up the discussion, and you are all encouraged to participate, whether you're sitting around our not-so-round table, or anywhere else in the room, just wave and I will find you.

We're recording this event and we'll post the transcript to our Web site, as we do for all our discussions. Also for media purposes, we're on the record. So when you're recognized, please tell us who you are. Also, please speak in to a microphone. For those of you in the audience, we have handheld mikes that we will quickly get to you. I will moderate the discussion; it will end at 10:30.

As I said, the topic for discussion today is tax-free healthcare. And we ask the following question: "Build it up or blow it up?" You know, in our last roundtable discussion last month, we asked the question: "Is the tax code the answer to encouraging saving in this country?" We had a great discussion, but we did not come to a consensus on an answer.

For several days after the discussion I thought about what people had said around the table, and eventually I came to my own consensus. And that is, the Tax Code is such a mess that it can't do anything to address the savings crisis. When I think about today's topic, and for the very same reason, I come to our discussion with a bit of a preconceived notion. I don't think the current tax system is the answer to addressing the healthcare crisis either.

Just look at the tax system. I still do my own tax returns by hand. The process is stupid and it's maddening, and since I'm not Rockefeller or Carnegie, my returns should not be that complicated. Phase-ins, phase-outs, hidden limitations, the AMT, qualified dividends, it goes on and on. The individual tax system tries to do too much in addition to its main purpose, which is to collect revenue. And by doing too much, it does nothing that well, including collect revenue.

The so-called tax gap, which is the difference between what's owed and what's collected, is huge and it's growing. And that's in a system where the government directs employers to collect a lot of our taxes for it. Then there is the estate tax. One year it's going to be here, the next year it's going to be gone; the year after that it's coming back with a vengeance. I think we've gone a little too far when we have a tax system that makes timing your death an important part of tax planning.

On the corporate side, we have a 20th century system that's not working that well in a 21st century economy. As our very own Marty Sullivan reported in the pages of Tax Notes recently, the Europeans are lowering their corporate tax rates and broadening their bases, making their economies more competitive. We're not. It pains me to think that it could be possible that we would be behind the Europeans in anything but soccer.

Our tax code is quite simply broken and it's broken badly. I know we have a healthcare crisis in this country, but here's my simple conclusion: I just don't think you fix one broken system by using another broken system. In fact, when it comes to every social problem we face in this nation, maybe — just maybe — our first answer should be don't go looking in the Tax Code; your answer is not here.

So with regard to the question we will discuss today, maybe the answer is neither to build it up or to blow it up. Maybe the answer is: Physician, first do no harm, leave the tax code alone. The healthcare structure that's now supported and promoted by the tax code took years to build.

Instead of building it up or blowing it up, maybe we should start dismantling it brick by brick. And maybe we should look for the cure to our healthcare crisis somewhere else.

We have three excellent panelists here today. And I suspect they don't share my views, certainly not in its entirety. Perhaps they will show me the error of my ways, and I'm eager to hear.

They are in the order that they will speak: Katherine Baicker, a member of the President's Council of Economic Advisers; Jason Furman, special assistant to President Clinton for economic policy, and now a visiting scholar at NYU's Wagner Graduate School of Public Service; and Len Burman, a senior fellow at the Urban Institute and co- director of the Urban-Brookings Tax Policy Center.

Thank you all for being here. It's a pleasure to have such an excellent panel, and I'm going to get out of the way now. We'll start with Katherine.

MS. BAICKER: Thank you so much for having me and I'm looking forward to the discussion as well. I know we've all been debating this issue for a long time, and we may say some of the same things, but I'm looking forward to hearing the new input from everyone in the audience.

I don't think I need to convince anyone that there are serious problems with our healthcare system today. You know about the serious problems with the tax code, some of which were just outlined.

On the healthcare side of things, our spending has been growing rapidly for decades, and promises to keep doing so without stop. We're currently spending a good chunk of GDP on healthcare, and that's going to rise to about 20 percent in 2015. That's a lot of money to be spending on healthcare. You would hope that we're getting our money's worth, but I don't think I need to convince you that we're not.

You see that we spend more than twice as much of our GDP on healthcare as a lot of other developed nations, and you'd be hard- pressed to find outcomes where we're doing anywhere near twice as better. In fact, on a lot of outcomes, we're doing worse. So we're spending a lot more money, our outcomes don't seem to warrant that kind of expenditure.

Even within the U.S., you see that the areas of the country where we spend a lot more money on healthcare are not areas where people are doing better or where they're happier with their care. So even the dollars that we're spending aren't going to the highest value users in the U.S., either between healthcare and other goods, or even within the healthcare system.

Now, spending a lot of money on healthcare wouldn't be a bad thing in the absence of those inefficiencies. Healthcare is invaluable to lots of people. I would like to live forever, and I'm sure many of you would as well, but that's not what we're getting out of the healthcare system.

So with that huge chunk of GDP going to healthcare, it's even more important that we get our money's worth out of it. So why aren't we getting our money's worth out of the healthcare system?

Well, one of the main culprits is the tax code. A lot of the — and I'm going to focus today on the private side, because that's where I think the tax code is both the cause of a lot of problems and a potential source of solutions.

There is a host of other problems on the public side as well. I don't even want to start talking about ways that we could improve the Medicare system or the Medicaid system. So let's put that aside for now except for one benchmark comparison. A lot of people don't realize, although those people probably aren't in this room, that we spend more money through the subsidization of privately provided insurance than we do on Medicaid. The federal government does.

So the federal subsidies of private health insurance are around $200 billion a year through payroll taxes and income taxes that aren't collected, that would be collected if private health insurers were not subsidized through the tax code.

So this is a lot of money, and that's a problem in and of itself, but the bigger issue is that because we subsidize insurance that you get through your employer but we do not subsidize out-of- pocket healthcare spending or most other insurance, you have a strong incentive to consume all of your healthcare in the form of employer-provided health insurance.

If you go to the doctor and it's covered by your employer plan, you pay with pre-tax dollars. If you go to the doctor and it's not covered by your employer plan, for the most part, you pay with post- tax dollars. That means you have every incentive to get an employer- provided plan that covers everything.

So if we provided any other kind of insurance that way — if we provided automobile insurance that way, where you got a discount through an employer-provided plan, your auto insurance would cover — you know, new paint and detailing the interior and gas and everything you could think of that's a predictable expense, because it would be much cheaper to get it that way. Well, that causes a lot of problems.

It means that you have little incentive to monitor the price of the services that you're consuming once you're in that first dollar employer-provided insurance plan. It also means that people who don't have access to employer-provided insurance are paying more for their healthcare than everyone else. As it turns out, the higher your income is, the more likely you are to have access to employer- provided health insurance.

If you look at people with income more than 400 percent of the poverty line, only 11 percent of them don't have access to employer- provided health insurance. If you look at people under 200 percent of the poverty line, that number more than triples. So there are really some distributional implications of that as well.

So it doesn't make a lot of sense to subsidize this one form of health consumption and not other forms. There are a couple of different ways you could level that playing field. You could level it by removing the tax exclusion all together so that all healthcare is purchased with after-tax dollars.

Now, the advantage of that is that you're no longer pushing people in to first dollar coverage plans. The disadvantage is, first of all, we want people to have health insurance. If as a society we make a decision that we're going to cover people if they go to the emergency room, there is an incentive to have people in insurance plans where they get more effective care. There is a public good aspect of people having good healthcare, and you know, it's a married good, we want people to have healthcare in the same way we want them to have food and housing, et cetera.

The other problem with eliminating the exclusion altogether is that through historical accident, people all get their — you know, a large majority of people with private health insurance who aren't on Medicare and Medicaid get it through their employer. That's our main mechanism of risk pooling, it's how our whole insurance system is built, and dismantling that, blowing it up in one fell swoop would have some serious consequences for risk pooling coverage, et cetera.

So if we were starting over, would we start with this system? No. But now that we're here, blowing it up seems a little reckless.

The other way you could go to totally level the playing field is to subsidize all healthcare consumption, let everybody purchase healthcare through insurance or not through insurance with pre-tax dollars. So Cogan, Hubbard, and Kessler have made this policy proposal in a book you might have seen. That would involve people being able to consume, you know — doctors, visits, whatever kind of healthcare they want to purchase with out-of-pocket dollars, and write that off, not pay any taxes on that.

Now, the advantage of that is also you have a level playing field, so people are no longer pushed in to these first dollar coverage plans. The disadvantage is that now you're subsidizing healthcare uniformly relative to all other goods, you might think people would consume too much healthcare, it's a very expensive proposal, it has some pluses and minuses.

There is a third avenue that I'd like to lay out, which is to build on health savings accounts as a way for people to consume healthcare out-of-pocket expenses with pre-tax dollars in a way that doesn't have quite the same adverse incentives to overconsume healthcare that uniformly subsidizing healthcare consumption would.

So health savings accounts, I will sketch out very briefly, hopefully. People have some sense of how they work. You can put money in this account and use the account dollars to buy any out-of-pocket healthcare that you encounter, but in order to use the account or in order to contribute to the account, you have to have a high deductible health policy, where the deductible is about $2,100 for a family. I'm sorry — yeah, $2,100 for a family at least. It could be higher, and the out-of-pocket maximum is at most $10,500 for a family, but it could be lower.

So in order to have the tax preferred account, you need to have a high deductible health policy. So this removes the disincentive to have catastrophic insurance. So you can buy a basic catastrophic health insurance policy and then consume your out-of-pocket expenses with pre-tax dollars.

So you no longer have an incentive to get a really generous first dollar employer plan.

It also doesn't have the adverse incentives to overconsume healthcare that just making all out-of-pocket expenditures tax free would, because as long as you're going to have healthcare spending over your lifetime that exceeds the amount that you can put into your account and that amount is capped, then you still have an incentive on the margin to think about — would I rather use my last dollar on healthcare or on some other good, and that's still a dollar-for - dollar tradeoff once you've hit your maximum contribution, and these accounts can be bequeathed to your descendants. So as long as you put some value on that, the incentives line up a little bit better.

Now, there are lot of questions about this approach. Would it actually reign in healthcare spending? Do we think that people are really price-sensitive enough to change their behavior enough to bring healthcare spending under control? Wouldn't people all be over their deductible or all over their out-of-pocket max, do we really think this is an effective tool? Are people even capable of making those decisions when they're sick? Are they really going to say which doctor is most cost-effective? And then there are questions about the distributional implications, what does it do for low-income people, what does it do for people who don't have access to employer-provided insurance, et cetera.

Maybe we'll get to a lot more of those in the question and answer period. I know I don't have time to through them all, although I have pithy facts for each and every one, so if you want to ask those questions, I promise to sprinkle them in.

The one question that I will address is can this actually bring healthcare spending under control. I think that's the first order question, and I think the answer is yes, this can be a really valuable tool in that direction. Obviously in isolation, there is a lot of other stuff that needs to happen, too. But let me give you a couple of reasons to think that this is a promising approach.

First of all, a lot of spending is done by a few number of people with high health expenditures. So you may have in mind a little fact like 70 percent of the dollars are spent by the 10 percent of people with the highest spending, and that's true. But that doesn't mean that this kind of set-up would not bring healthcare spending under control, because about 50 percent of our healthcare dollars are spent by people who would still be in the price-sensitive range of the typical high deductible health policy.

So there's a deductible, then there is cost sharing up to an out-of-pocket max. Fifty percent of our dollars are spent by people below that out-of-pocket max. So there are still price incentives in place for a lot of our spending. I'll take 50 percent of our spending being price-sensitive. That sounds like an improvement.

Also, only about 20 percent of dollars are spent by people in emergency situations, where we're talking about people going into emergency rooms are with life-threatening conditions. So a lot of our dollars are spent in an environment where people who were thinking about getting high-value care, rather than just getting the most care, would have an opportunity to do that evaluation.

There is a fair amount of evidence from the medical literature that in fact, people do use available information about price and quality to change their health consumption patterns, even in emergency situations. So if you look at the hospitals that people go to when they have heart attacks, the better the information is about quality, the more people go to the high-quality, high-value hospitals. So people would use that information.

And the third point on that front is that people have in mind, oh, you throw somebody into one of these plans, they're out on their own. How are they going to negotiate with providers, how are they going to be able to negotiate a system that an insurer couldn't? This is a tool you add on to the other tools people have available.

So people with high deductible health policies still get their insurer's negotiated rates with providers. They're not negotiating on their own, they still get access to their insurer's disease management tools, they can consume preventive care not subject to the deductible. So there are lots of tools available to help people consume high value care.

So I think I will stop there and leave to the discussion period any other issues you might have about the distributional implications of this. And I hope that I've given you a sense that with the right information available, including price and quality information that people need to make good decisions, they will make better decisions about allocating their resources both within healthcare and between healthcare and other stuff that they might want to consume, so that in the long run, with our dollars allocated more efficiently, we don't have to worry so much about how much we spend on healthcare because we'll know we're getting our money's worth.

MR. BERGIN: Thank you.


MR. FURMAN: Thank you, and I should say that for every one of Kate's pithy facts, I have a somewhat less pithy response to it, but you'll be spared those as well, at least until the question time.

I think that Kate's diagnoses of what's wrong with the health system is — she's hit on at least one of the important aspects of the problem, which is the tax code, and I think one of the great things about the president's proposal this year is that a lot more people are aware that first of all we're spending $200 billion a year, which is more than we're spending on Medicaid, subsidizing the employer-sponsored system. We get a certain amount for that.

We have 180 million people covered through it. We have a certain amount of pooling and other insurance aspects that we'd like from it, but it suffers from a lot of defects. One thing is an unfairness. A high-income person might get a $4,000 subsidy for their health insurance; a low-income person might get a subsidy of more like $1,000 or maybe even $0 for their insurance.

Not only is that unfair, but that $4,000 is a much larger subsidy than you need to have for that high-income person to get insurance. You're just spending a lot more on them than you need to. And that $1,000 or $0 may not be enough for the low-income person to get insurance. It has some of the efficiency costs that Kate described in terms of the effects that it has on people's incentives to — the form that their insurance takes, to get insurance that basically isn't what economists think insurance should be, and pre- pays a lot of their cost, and as a result makes them less cost conscious and spend inefficiently high amounts on healthcare.

So I agree with all of that. I'm not completely convinced as to what the magnitude of those problems are and how huge they are, and whether people don't have a very strong cultural disposition towards wanting their healthcare paid for by someone else or paid for in advance, and how much they'd completely change were the tax incentives different, but no doubt sort of which direction those incentives go.

But then when you listen to something like the prescription coming out of the administration, it just doesn't follow at all from the diagnoses. The core principle behind it, and behind the Cogan, Hubbard, and Kessler proposal that we heard as well, is basically a "two wrongs make a right" type of theory, which is we have $200 billion of tax incentives. These tax incentives have all these problems, and you can sort of eloquently list every one of them, and how do we solve that?

Not by taking away any of those $200 billion, but by adding another wrong on top of it, and hoping that that second wrong cancels out the first wrong and somehow turns into a right.

And the circumstances in which that could happen — I mean, there are times when two wrongs do make a right. I don't think this is one of them. The second wrong is, in the case of the administration's latest proposal, an extra $15 billion a year, and that's on top of the $200 billion. That's at a time when we have a large fiscal gap. The $15 billion is very regressive.

If you look at it, half of the cost of it is raising the amount you can contribute to health savings account from about $5,400 a year for a family to $10,500 a year. These are the maximum amounts; not everyone could contribute that. You're talking about something that only the highest income families would really benefit substantially from — being able to save $10,000 a year tax-free on top of all the other ways that we can save.

It's very regressive, it's very costly, and it's not going to address a lot of the problems that we care about in our health system, and will make some of them worse. You often hear these types of plans going under the rubric of consumer-driven healthcare. And there's something a little bit incoherent about that, because the basic approach says we trust consumers to decide which doctor to go to, which hospital to go to, which medical treatments to buy, but we don't trust them to choose their insurance plan.

When it comes to their insurance plan, the government has decided this is the right deductible to have; this is the right out- of-pocket limit to have. If you take one of these government-approved plans, you're going to get special tax breaks. If you do. If you choose to have another type of insurance plan because you like getting insurance a different way, you like a different deductible, you just have different preferences than the people that drew up these regulations have, you don't get any of those special tax breaks.

There's something fundamentally incoherent about it in terms of a theory of rational sovereign consumers. They're really smart about healthcare; they're really stupid about insurance. That theory also then contributes to a limitation. We hear all these differences. You know, you get a tax benefit for insurance in the employer market but not in the individual market. Or you get a tax benefit for insurance but not out-of-pocket.

By limiting your solution to high deductible plans, you're limiting your solution to about the 15 percent of the population, privately insured population, projected to take up those high deductible plans. For 85 percent of the population, all of the problems in the tax code, everything Kate outlined, everything I outlined, will still exist under this type of plan, because they won't be in high deductible plans.

So they'll suffer from, to the degree I think those are problems with our health tax system; it's just not going to solve them for 85 percent of the population.

Now, in terms of what its actual impact on the health system would be, probably the two most important variables are cost, and in particular, cost-effectiveness, and second of all, uninsured, what does it do to the number of uninsured. In terms of cost- effectiveness, it's really important to pay attention to the latest proposal, which is not sort of are you for HSAs or against HSAs; we already have HSAs.

Should we take those HSAs that we have and expand them? Would that increase health spending or reduce health spending? I think there is a very good chance that taking our existing HSAs and expanding them would actually increase health spending, and here is why.

We already have — Treasury projects that without any new laws at all, we're going to have 14 million HSAs in 2010. These expansions of HSAs would take those 14 million people who would have been in HSAs anyway, and most of them will have more of an incentive to spend more money out of their HSA than they would have had in the absence of this proposal.

So you have 14 million people who are now potentially spending more on their healthcare. There's a number of technical issues that Kate and I could go into, it turns out the incentive — the reduction in the after-tax price of healthcare is anywhere between zero and 30 percent for the typical person, depending on the assumptions you use, but they are going to spend a substantial amount more. Then this proposal is going to attract seven million new people according to Treasury estimates. Seven million new households to HSAs. And those households might spend a bit less. The amount less that they spend is limited by the fact that most of our health spending is driven by the highest-spending people. So they're not going to reduce their health spending much.

It might be driven, reduced — mitigated a little bit by psychological things that HSAs — you build up your money in the account, you get a debit card; it seems free. You want to spend it more quickly, so you don't build up too much.

But whatever. Even if those seven million people reduce their spending, you have to have them reduce that quite a lot to make up for the over 14 million that already would have been in HSAs who are going to be spending more. So it's far from obvious to me that it will reduce health spending. There's a good chance, from my perspective, it would increase it.

The second question is what does it do to the number of uninsured. And the title of the this session is — you know, the employer-sponsored system, build it up or blow it up.

The employer-sponsored system is certainly not great. Anyone who sort of worships at the altar of it is probably a little bit confused, but if you start to undermine it without creating anything as an alternative, you can create a lot of problems because of the historical accident you talked about, Chris.

And you look at John Gruber of MIT's estimates, his are that you take away the tax advantage for the employer-sponsored system, and this proposal actually gives you more of a tax advantage for the individual system for a number of reasons, and nine million employers drop coverage, or nine million employees choose to leave their employer-sponsored coverage.

A bunch of new people get insurance because of the incentives in the plan, and on net, you have a 600,000-person increase in the number of people without insurance. So this is $15 billion a year, and 600,000 more uninsured. Now, [off mike] and I think that number is too low. They think there would be more employer dropping; Treasury thinks there'd be a lot less employer dropping, so they think three million new people would become insured as a result of this proposal.

Any of these numbers though are just a lot less new coverage than we should get for $15 billion a year, and then when you unpack that coverage, a lot of the new people getting coverage are somewhat healthier, and a lot of the people losing coverage are people who are less healthy and less able to purchase affordable insurance or insurance at all, in more of an individual market-type of framework.

So it's a proposal that would increase health spending, might increase the number of uninsured. Even if it reduced it, you're spending a lot of money for a small reduction.

So what would I do instead? Well, first of all, rather than two wrongs make a right, let's just reduce or get rid of the first wrong in the first place.

I'd love to see the exclusion cap just scrapped, and the money used for a more efficient and equitable way to cover people. I recognize this isn't the most politically salient thing in the world, but I'd love to get it to the status of the gas tax, where you have Tom Friedman twice a week writing a new column about how it's a no- brainer, and he wants to see some politician get up and say — it's sort of nowhere near that status. I think all of us in the sort of "not running for office" community of health analysts have an obligation to raise it to that status.

Second, I'd like to see more cost consciousness on the part of consumers. I don't think it's going to solve all of our problems. I'm humble about what we know in terms of cost consciousness. It's not obvious to me that a $2,100 deductible is the right way to make people more cost conscious. Certainly, it seems like we want a smaller deductible for low-income people and a higher deductible for high-income people. You might be able to implement something like that, income-related coinsurance in the context.

Medicare, I'd like to see more things of all — I mean, managed care is in effect saying you have 100 percent coinsurance for things that managed care chooses not to cover. So in a sense it's almost more restrictive than a high deductible policy, and then low coinsurance for things that the plan thinks you should have cover — zero for preventive, which is also an option in HSAs, which Kate will remind us. I'll remind you in advance before she does.

The third thing is, we want plans that have more pooling, and the employer-sponsored system is one way to do that pooling, and it could be strengthened by strengthening tax incentives, for instance, for small businesses, which is one of the weakest links in the employer-sponsored system.

The much bigger no-brainer to solve our pooling problems is not through the Tax Code, not through bribing people, but through doing the other thing that government can do, which is force them to do things. It's mandating insurance at the individual level or the employer level, another thing that would be nice to get to the same intellectual status as the gas tax.

And finally, if we were in the world of [off mike] measures and you think most of these other ideas are pipedreams, let's remember that the single most cost-effective way to expand health coverage is to expand Medicaid. And if you let everyone up to, for instance, 100 percent of poverty to 225 percent of poverty, something like that, into Medicaid, you're going to get bang for your buck for each person you expand health insurance to than any of these different tax credit, tax deduction types of approaches.

MR. BERGIN: Thanks Jason.


MR. BURMAN: Well, it's pretty intimidating to follow Kate and Jason. Those are two people I have enormous respect for. They're really smart, they're really thoughtful, and they really cover the entire waterfront. So I was trying to think of what's left for me to say. Basically, things that are either not very smart or that are just wrong.

So here I go.


MR. BURMAN: You know, Chris started out by saying that he thought — basically it was just to get rid of the tax subsidy, and anybody that looks at this looks at what the tax subsidy has done and the problems in the healthcare market would have to conclude that we're not spending the money very well.

But you know, this isn't quite the no-brainer. You know, getting taxes out of the market isn't quite the no-brainer than it is for a lot of other public policy problems.

You know, when I was at Treasury, we would often fight bad ideas. Most of them actually came from Jason. And we'd start out by saying there wasn't a role for government, even though people are —

MR. FURMAN: Really bad ideas —

MR. BURMAN: You know, even though people are suffering, even though we feel their pain, you know, it's just not a role for government. Well, in this case, there clearly is a role for government. You know, even the hardest-core free market advocate would have to conclude that — well, if they're sensible — they would have to agree that all of the conditions for efficiency in a market don't exist in the healthcare market.

Basically, there is the problem of moral hazard that Kate was talking about, exists with insurance. People want insurance, people need insurance; you need insurance just to achieve economic efficiency, but the kind of insurance that wouldn't give people the incentive to over-consume is very hard to come by.

It's a problem of adverse selection, if people can choose whether to buy insurance or not to get insurance, the people who are going to want insurance are the ones who are especially sick, the insurers are going to want to sell insurance to the ones who are especially healthy, and that just creates a big problem. Basically, you end up with a system where the people most likely to get insurance are the ones who need it least, and the people who need it most don't have — they have an incentive to buy it when they need it but not before.

There are problems of information. People don't know what they need to know to make good decisions. That's even true for the doctors, and it's certainly true for the patients. There's a thing in economics called the principal-agent problem, that the people on one side say the insurers have got different incentives from the insurers; doctors have different incentives than their patients do.

Doctors on some level are selling something, and so they actually have an incentive to sell people treatments, whether or not they're cost-effective. If you provide emergency room care, there's a free rider problem. People don't have an incentive to get insurance because they know that they can get covered when they go to the hospital, and there's a whole long list.

So clearly, you know, there is a role for government. The real problem is figuring out what's the best way to meet that role. And I agree with both Kate and Jason when they say that even though the tax subsidy isn't necessarily the most sensible way to provide coverage, there's an enormous number of people who get their health insurance at work. And if you just threw out the subsidy, there'd be a big risk that you'd have fewer people with insurance. So you have to come up with some better alternative if that's what you're going to propose.

Unlike a lot of other things that economists talk about, this really is a — really important problem. There are 46 million people that don't have health insurance, probably more now. The number rises constantly. Small employers have huge problems in providing health insurance. Their costs are higher than they are for large employers, and the kind of risk pooling mechanism, the thing that makes the employer base a relatively good way to provide insurance, which is that you have a large number of people who are brought together for reasons that don't have anything to do with their health status, that doesn't really apply for small employers. You might have four or five employees, if one of them gets sick, you have a really hard time getting insurance.

And as has been noted, we're spending a boatload of money on tax subsidies. Over $200 billion of income in payroll taxes. It does do some good things. You know, most working-age people get insurance at work, and employer base is a good way to pool people. Even if we didn't have a tax subsidy, it's likely that large employers would offer their employees a way to get health insurance, because it really is a particularly efficient way to get insurance.

And by encouraging people to buy their insurance, the tax subsidy by encouraging people to get insurance, as Kate mentioned, deals with this adverse selection problem to some extent. It gives people incentive to buy even before they know they have health needs. But clearly, we're also not getting our money's worth.

The open-ended subsidy does encourage people to get too much insurance, overly generous insurance that exacerbates this moral hazard problem, and the subsidy is upside down. I actually did a study several years ago looking at the distribution of the tax subsidies from health insurance. This was in 1998, so you probably would double or triple these numbers now. But for people with incomes under $10,000, they got an average subsidy of $177, for a number of reasons.

One is they did no income taxes. So an income tax exclusion is worthless to them. Their employers were much less likely to offer insurance than higher-income people. When they did offer insurance, they paid less towards the insurance than they did for higher-income people. For people earning $200,000 and over, the average subsidy was almost $2,000, and those people will get insurance under any circumstance. So the subsidy in that way is just kind of irrational.

As Gene Steuerle, who is sitting at the table has mentioned times, throwing more money at the problem really, really isn't the answer. And basically all of the major proposals out there would do that. And there's a good reason, and Jason sort of hinted at that, because these proposals are made by people who run for office and they don't to take back what's already out there.

But the health savings account proposal of $15 billion a year, the Hubbard proposal, basically just throws the barn doors open and subsidizes everything.

The problem is that healthcare is at the core of very, very, serious budget problems. Not just through the tax exclusion but also through Medicare and Medicaid, and sometime relatively soon we have to figure out a way to reign in these costs. And adding another $15 billion a year doesn't seem like a good place to start.

As Chris mentioned, I think, actually that — the version of what he said that I'll completely agree with is the solutions won't be found in the tax system alone. I think Kate mentioned this as well.

One of the good things about the administration's proposal is they take seriously the problem of information, that doctors and patients need a lot more information to be able to make good choices. They worry about the problem of this risk selection on the part of employers.

I think there clearly is a role for government to just get a lot more information about what works and what doesn't work, and to disseminate that information. You have to deal with this problem of adverse selection in any kind of a free market system, and particularly in a system outside of employers, there's such a strong incentive for insurers just to cover the healthy people.

And I think there really — that's a serious problem, and the solution to that is not going to be found entirely in the tax system. But it's also true that if you were going to throw more money at the problem, it's very likely that the best place to throw the money, as Jason mentioned, might be not in the tax system, but through expansions and programs that provide coverage to low-income people, the ones who are least likely to be covered now. Expansion of Medicaid, expansion of a program that covers the state childrens' health insurance program and so on.

The idea of high deductibles is not a bad idea. I mean, high deductibles could help. Improving cost consciousness I believe would reduce spending by a little bit, probably not by a lot, but it could also make things a lot worse, especially for those who are chronically ill, who would hit the high deductible every single year. And for people with low incomes, for whom even, you know, it sounds like $2,000 doesn't sound like a lot of money to people in this room, but if you're only earning $13,000 or $14,000 a year, it could be huge, and hitting that deductible even in a single year could be catastrophic.

And as Jason mentioned, it's not clear why we should pick one method to control costs. Better option, and I think actually all of us, at least offline, would agree that a better option would be just to cap the exclusion and encourage people to find cheap health insurance that works for them, which might be high deductibles or might be more aggressive managed care or something like that.

Subsidies for non-group health insurance, actually they have a certain logic to them. It seems kind of unfair that you provide subsidies to people who get insurance at work, but you don't provide any subsidies at all to people who have to buy the insurance on their own. And obviously, since people who don't get insurance at work are the ones who're most likely to be uninsured, it makes sense on some level to provide subsidies for non-group insurance.

The problem though is that if you provide — and this is Jason's point — if you provide generous subsidies for non-group health insurance, then particularly for people who are healthy and young who can get a good deal in a non-group market, they're going to say, well, I don't want to get insurance at work anymore. Raise my wages. You want to provide insurance for the other people, that's fine, but I'm not going to pay for it with lower wages.

When the healthy people don't want to be in the employer group, the employer group doesn't become a good pooling mechanism anymore, and that's a serious problem, especially for small employers, small groups. Think about for small employers. Small employers are struggling to provide insurance. One strong incentive they have now to provide insurance is that it's the only way they can get tax- subsidized insurance for themselves and their families.

Well, if they can buy insurance in the non-group market and get the same subsidies that they would get at work, they have a much smaller incentive to put up with all the hassles of providing insurance to the employees, and that's a really serious problem.

So in terms of reallocating the subsidies, just subsidizing non- group insurance I think is really fraught with difficulties, and the reason I think that John Gruber estimates that the new proposals would increase the number of people without insurance is specifically because it's tilting the playing field in favor of non-group insurance.

I'm running out of time, so — the obvious solution for all the people here who don't have to run for office is to cap the subsidy, turn it from a tax exclusion into a refundable credit, so you actually provide benefits to low-income people, or better, provide it in the form of a voucher.

Problem with tax credits is that you get the money — particularly if you're buying the insurance with your own money, you get the money after the end of the year. The Treasury has this Rube Goldberg mechanism for making the refundable tax credit basically look like a voucher, but it's not clear that it would actually work.

And the more sensible thing would be to give people a piece of paper and say you can buy insurance at work or outside of work, and this will help you pay for it.

The other thing I think is really important is that any subsidy you provide for low-income people should be in either the employer- sponsored insurance or in the non-group market, and you shouldn't provide an additional advantage for non-group insurance.

One thing that I had proposed in a paper a number of years ago, with a Gruber who wasn't John — it was Amelia Gruber, who was a research assistant of mine — was that if you were going to provide subsidies for low-income people and it was going to be available in the non-group market, you should make insurers do something for the money.

Right now, basically the subsidy is available to anybody who buys insurance and qualifies. Well, what we proposed was that if the non-group market wanted to qualify for subsidized insurance, they would have to solve these problems of selection, they would have to offer insurance that was really completely portable, that if you started buying insurance when you were healthy, at a low rate, and over time you became less healthy, that there would be a way to guarantee that you'd always get the cheapest insurance from your insurer or from any other insurer you bought it from who qualified for the tax credit.

The basic idea is use the credit as a mechanism to induce market change. I don't know whether this would work. I mean, the insurance market would have to figure out a way to make this kind of fully portable insurance work, and you can imagine the reasons why it wouldn't. But if it didn't work, it wouldn't cost the government anything, because nothing would qualify.

There's probably a version — I think Jason actually was hinting at — there's probably a version of health savings accounts and high deductible health plans that would actually solve a lot of the problems that concern Jason and — actually Kate and I. And that is a subsidy that was more targeted towards low-income people.

One of the good things I think of the recent set of proposals is that I think there is a growing understanding that the poor need help. There's just a question as to how much help they need. And one possibility would be to come up with an HSA proposal. One possibility, as Jason mentioned, is you could just have the deductible vary with income. So if you're earning $15,000, the deductible might be $100, and if you're earning $100,000, the deductible might be $10,000. Feldstein and Gruber had a proposal like that.

Another possibility would be to have a refundable tax credit that would basically equal the entire deductible for low-income people, and maybe almost nothing for higher-income people — and also to subsidize the cost of insurance for low-income people. That would give low-income people still a strong incentive not to spend money out of the HSA. If they could leave it in there, then it would give them retirement money that they could use later on. And it wouldn't put them at big risk if they turned out to be really ill.

The other thing is you have to do something about the non-group market. You have to have a place for people to buy insurance where they're guaranteed of getting it at relative — at fair rates, guaranteed portable insurance.

One possibility again would be to tie the credit to insurers who are offering fully portable insurance, and also put some responsibility on the insured. You could say that the full credit would only be available if you'd maintain continuous coverage, either through an employer or any individual in non-group market, and that would deal with the insurers' concerns that only the healthy people would want to buy in.

This is a fascinating issue, and we could probably talk about this all day long, but I appreciate Tax Analysts setting up this forum and I look forward to hearing what everybody else has to say.

MR. BERGIN: Thank you, Len, great panel discussion so far.

I'm going to open this up now. Let me just remind everybody, if you have a question, I'll find you. If several hands go up at the same time, I'll try to go one, two, and three and let you know when I'm coming back to you. And please state your name, even if we've known each other for years — I always look at Sheldon when I say that. Just to help us out with the transcript. The transcripts generate a lot of interest, so we like to get them out quickly.

Sheldon, I'll help you out.

I'm Sheldon Cohen, and is this thing on? I guess it is.

MR. BERGIN: Yeah, it is.

When I first came to the Revenue Service in the dim, dark, distant days some 55 years ago, there were still people around from World War II, and we were in the middle of the Korean War. All this comes out of avoidance of inflation problems in World War II. That is, we decided that we would encourage people to take non-cash income instead of cash income, in order to avoid inflation problems.

I mean, that's the initial decision that gets us into this sticky stuff. And once you get into the tax system, it's like a narcotic. It hides the cost, it hides the administration, nobody gives a damn thereafter, and it just festers.

Having had some responsibility for administering the system, we pretend it's a cost-free administration; it is not. It is very debilitating, and it's a very costly administration, because it distracts from other things that the Service is supposed to be doing.

And once we've gotten it — I mean, it's hard to break that narcotic habit. We're going to have to find some way to wean ourselves off of it.

Vouchers were once a mechanism that was used quite often. I mean, during World War II, you got a voucher from the defense establishment that you needed to build a war plant, then you got ability to get materials that you couldn't otherwise get. The Service didn't issue a credit to build a plant.

It's just easier now — I mean, the whole problem with our tax system, I am convinced, is the appropriations process. It is easier not to collect the dollar than it is to collect the dollar and spend it.

My name is Brooke Albright. I'm actually a medical student and I'm a member of AMA. And one of our biggest focuses right now is covering the uninsured, so that's why I attended this conference. My question is actually a two-part question. First — maybe Jason sort of touched on this, and it is, have you guys explored, or what other areas have you explored as far as pooled plans?

I know you had the small businesses and the employer-based programs, but other areas that you could pool? Have you considered other areas?

MR. BERGIN: Any chance that like Jean Lambert or Chris Jennings or someone that knows what they are talking about wants to answer that?

SPEAKER: Or we can just talk and then shake our heads.

MS. LAMBERT: I'm Jean —

SPEAKER: We are enjoying this a lot.

I'm Jean Lambert. I used to work with Jason, which is why I think he pointed to me, and Chris Jennings and a few other people around the table. You know, there are a number of very interesting ideas out there, with the concept being how do you, potentially through pooling rather than direct regulation, figure out ways to spread risk and ensure that both rates are affordable and people have access without kinds of other sorts of problems?

I mean, we have seen people structure regional purchasing pools, we have seen state-based purchasing pools. I think a lot of the details matter. Who has access, is it through — you know, are individuals allowed in or are groups or employers allowed in? But there is a bunch of options out there.

The Federal Employees Health Plan is typically the model that people look at. I'm not saying people should be allowed to buy the same insurance that people in the federal employee system have, but the model itself is kind of an example of a pool that most people point to.

MS. ALBRIGHT: I actually have, on behalf of the medical students, an idea that I would like to share with all of you. One of the largest uninsured populations is the healthy population, the 18- to 30-year-olds that are in school, that are going to a university that is tuition-based. And several of the students, I know even in medical school, are uninsured. And so one of the solutions that we have sort of tossed around is the idea of mandating like another pooled population in tuition-based universities.

And that is that part of your — I mean, they are able to force us to pay a recreational fee, why can't they force us to pay a health insurance fee? And then offer that tax incentive to universities for providing us that health insurance.

I mean, that's something that — we don't have incomes as a student, so a tax subsidy is not of interest to us. First of all, we don't even understand it. So you know, to provide a tax subsidy to the healthy populations that are younger, that have less understanding of the tax issue and also less incentive because we don't have income — it doesn't cover us in any way, so it doesn't give us any incentive.

So one way that you can get us to get insurance is to mandate it through universities and it's covered by tuition, it's covered — we are going to pay it back eventually anyway, so it's not taking away from the pool of money from the government, it's something we are going to pay back.

And we don't mind paying a health insurance fee because we do get sick and it is a big burden on our parents when we get sick and they have a big bill from us.

So I think that's another maybe pooled area to explore is tuition-based universities. And you can even go as far as the private schools and like junior high and high schools that are private tuition-based to mandate that as well.

MR. BERGIN: It's an interesting idea.

MR. KLIPPEL: I'm Charlie Klippel from Aetna, and we are the largest writer of college-based insurance in America. And I would tell you that there are a number of states and a number of systems that have essentially mandated coverage requirements at the college level, and in some instances at the graduate school level. And I actually agree with the point that it is a — in that instance, the mandates tend to work fairly well.

That is, you really do get a lot of people covered and the pooling is good when you do that and the cost is not high.

It is a selected population, so it is younger, healthier people. So I think that's a very good idea and I think it has proven in many states to be an effective idea. I'm not sure you can extrapolate it to solve all the problems of the uninsured or — and I think, you know, it would be interesting to see whether the Massachusetts model, which purports to mandate coverage for all individuals, will in fact have the same kind of effect, because you are now cutting across a different population base.

MS. STRONG: Hi; Katie Strong with the U.S. Chamber of Commerce. And to build on what you were just saying about the Massachusetts proposal, one aspect of that does include a mandate-light plan for that young healthy population that is uninsured. That will be a much more affordable product to them, and it will not be required to cover all the extensive listed mandates that Massachusetts does. That's another model we should certainly look at.

MR. OOMS: I'm Van Ooms from the Committee for Economic Development, which is an organization whose trustees are essentially business executives and university administrators. Just to make a note that I think, unlike some other problems where we talk about them for 20 or 30 or 40 years and don't seem to make much progress, the urgency here may be a little bit stronger than we think, and it may be accelerating, because it looks like the competitive pressures out there are developing in such a way that the incentives and even the possibilities and feasibility for business to continue in the healthcare business are becoming more and more limited.

We have not done a survey, and so the evidence here is rather anecdotal, but we are doing a project on healthcare, trying to look at what might follow the employer-based model if indeed the employer- based model simply disappears or — obviously that doesn't happen overnight, but gradually erodes and the — you know, the recent experience with Delphi and what's happening in other companies, not only in healthcare but on pensions as well, suggests that you have sort of enormous pressures on the cost side here which are going to a lend a real urgency to this, I think, sooner than perhaps many of us expect.

MR. BERGIN: I think I'm hearing a couple of things here. It goes back to what you and I were talking about before. The ideas here — I haven't heard a tax idea yet. A lot of us who are in the tax area respect the people who are in the medical area. I went to law school to study tax and thought I wouldn't be involved in any life-and-death decisions; tax is, you know, simpler that way. And Len and I were talking before about in this area, there are life-and-death decisions where it meets the tax code. So maybe the tax code is not the best answer for a lot of it.


MR. BRUNORI: David Brunori, Tax Analysts. I have a question. I keep hearing this doom and gloom from our speakers and some other folks about how we are heading for this impending crisis, but isn't it true that for those 180 million people who are perhaps confused about how good the employer-based insurance is, aren't most of them perfectly happy right now?

I mean, is this a real political problem or is this a — I mean, is it a problem that experts and economists and really smart people in Washington are identifying?

But I have to tell you my own personal experience. I have insurance, my kids get sick, we take them to the doctor, I get sick, I go to the doctor; never think about it. I mean, I like the doctors we have. I have very good insurance from Tax Analysts, but I don't think I'm atypical.

And I don't see politically for the — you know, there's certainly a problem with the 40 million people who are uninsured. That's a serious problem, and there is a problem with the small businesses out there in terms of being able to afford insurance for their employees. But for the vast majority of Americans, I don't think this is a problem. I mean, I really don't.

MS. BAICKER: I'd love to jump in on that.

MR. BERGIN: Please, Kate.

MS. BAICKER: I think a lot of people, including this last questioner, have highlighted that there is a disconnect between people's perceptions about the expense of the healthcare system and the real expense of the healthcare system, that most people who have private insurance think "I like my doctor, I like my insurance plan," but then they think, "Gee, I wish my wages were higher."

And in fact, about 25 percent of the potential increase in wages over the last 10 years has been eaten away by the increase in health insurance costs. So would you like your wages to have grown 25 percent faster? Most people would say yes.

MR. BRUNORI: But that's not a conscious thought most people have.

MS. BAICKER: Right. Exactly. So I'm highlighting that people — some of the consequences of the rapid growth in health expenditures may be invisible to people, because it takes a couple of steps for increases in the cost of your benefits to pass through to increases, or a lack of increases, in your wages.

Also, people may be nervous about switching jobs if they or their family members are sick.

If you ask people in surveys, "Are you worried about your health coverage?" A lot of people say yes. They may be happy with their current coverage now, but they are fearful of losing it, either because they change jobs or lose their jobs or because they want to retire early but they can't. Then think about the tax burden that's imposed not only by this tax expenditure on the subsidization of employer-provided health insurance, but also on Medicare and Medicaid.

The unfunded liability of the Medicare system is about five times as big as the unfunded liability of the Social Security system. And both of those problems are hard to persuade people of the urgency because they seem off in the future.

But if you ask people "Are your taxes too high," they would surely say yes. So the disconnect between the bills that you see — okay, some people in surveys report that they would like to have more money in their pockets.

And if you ask people "What are your health insurance premiums," they tell you the dollar amount that they pay. They don't tell you the dollar amount that they pay plus the amount that their employer pays on their behalf. So part of the reason that we are having trouble getting to better solutions is that the problem is obscured by all these layers of tax provisions, and you know, multi-step transmissions of healthcare costs to people's pockets.

And people have this idea that we should somehow expand coverage and people should have all the healthcare that they currently consume now and people who don't have access to it should get more, and that that doesn't cost any more money.

There is just nothing for free. And so we need a mechanism that allocates our resources more efficiently, given that those resources are scarce. And I think we would all agree that getting healthy people insured at day zero and then having everybody insured continuously until death, which would be in the indefinite future, would be great. And so the question is how do we get young, healthy people into the system?

And I would hate for the solution to that to be focused on what we do with current chronically-ill people who are uninsured. They really have a separate problem. That's a distributional problem that we need to address through a separate mechanism; whether it's extra money for them to buy private insurance, whether it's getting them into state risk pools, whether it's getting them into public programs. In an ideal insurance set-up, people would get insurance before they get sick.

Given that we are in a world where some people are sick and have predictably high expenses every year, you don't build an insurance system around catering to them, because that's not insurance, that's a transfer. So let's build the insurance system that will get everybody covered where people are aware of the costs of their coverage and making rational trade-offs in a world of scarce resources.

MR. BRUNORI: I agree and I understand the issues that we are talking about here. All I'm saying is that no one is going to vote on this issue. In November, no one is going to go in and say, well, how does my — this candidate come down on healthcare, because it's not on people's radar screen.

MS. BAICKER: I would disagree with that.

MR. OOMS: I mean, I think wage — you know, a very large part of the population has seen essentially stagnant wages for a long period of time. Compensation has been going up, and that's why, you know, when you get into a political argument about this, somebody says, oh, wages are stagnating, and somebody else says, no, no, compensation is rising. Well, it's the ever larger wedge of the healthcare costs of course that is the difference between those.

But as the employers begin to shift — well, not begin because they have been doing it for some time — but we are well along the process of employers trying to shift, having to shift a lot more of the costs on to the workers from the healthcare. Pretty soon, somebody is going to figure out that the fact that my wages never go up has something to do with the fact that the healthcare system is broken.

So I don't think it is a political freebie. I think you've got — I mean, I think you've got in the sort of structure of this a huge political problem brewing, and I guess my surprise is that that connection hasn't been made more directly already. But it will be; I don't think there is any question.

MR. BERGIN: Let me let Len jump in here and then I'll get back to the hands that were up.

MR. BURMAN: One thing that's kind of frustrating is — and I think this has driven our politics for like the last 20 years, is this idea that policies can only be enacted if everybody is better off. And I think actually part of the reason we are having these big problems —

Except people in the future.

MR. BURMAN: Well, yeah, except people in the future. Yeah, I mean our current budget problems are this idea that people want to pay lower taxes, people want more services from government, we'll give them the whole thing. And there's like this illogic. Basically what this — I mean, I think actually the problems in small businesses especially will produce some political impetus to solve this.

I think employers increasingly just don't want to deal with this. They don't want to have to explain to their employees why their wages aren't going up. They don't want to have to — for a small employer, they don't want to have to explain why they can't afford to provide health insurance anymore even — especially when one of their workers is really sick.

In some ways, I think actually the HSA proposal would help get us to reform, because as the current proposals are structured, it would hasten the demise of the employer-based health insurance market, and I think actually reducing the number of people with insurance over time and eventually produce a crisis point at which people would actually understand that doing something radical actually would be necessary.

It would be nice if before we got to that point, you actually had some leadership, you had a president, members of Congress who would explain that the current system is irrational. It's really not that hard to explain. That's basically all we've been talking about this morning.

And you can't replace an irrational system with a rational one without rearranging things a little bit. But people have been willing to accept that in the past. I read a speech that Ronald Reagan gave in 1982, where he said we went too far in cutting taxes; we are actually threatening our economic prosperity, and we actually have to raise taxes now or else we are going to undo all the good things that we've done. It's almost inconceivable that any politician would say that.

Now, even when the Democrats complain that the Republicans are cutting taxes too much, they basically come up with their own proposals that would increase the deficit by just as much, but just in a different way.

I think my comment is going to seem very small after those comments. My name is John Irons. The comment/question that I have kind of goes to — I'm a little bit disturbed by the talk about costs, because I think that the core problem with healthcare is the increase in costs, not the costs themselves.

So here is a quick thought experiment. Suppose you could decrease costs 25 percent across the board. There would be a decrease in costs. And with healthcare costs going up by 6 to 7 percent a year, we are talking about four to five years, maybe six years' worth of cost savings, and then you are right back in the same position.

And so in my mind, the core issue is how do you decrease the growth rate of costs, not necessarily increase or have some one-time cost savings. And in my mind, the tax code, however you want to do it, might save costs on a one-time basis, but it's very hard to see how that can really decrease the growth rate of costs. I hope I'm wrong, and maybe someone could tell me how I'm wrong with this. But if I'm right, that says to me that the key here is do what Len's approach is, is that the tax incentives ought to be structured in a way that changes the market rather than changing one-time costs. And that's the right approach. But I'm interested in hearing what people thought about that.


MR. STEUERLE: Gene Steuerle. I have a dream once in a while that I'm sitting before the Ways and Means Committee and all of a sudden there is an interruption and someone from the National Cancer Institute comes running in and says, "eureka, we found a cure, though very expensive, for cancer."

And all of the people in the audience are pretty happy. They are thinking about the good things that are going to happen to us in life. And then I look behind the podium and all of the members of Congress are wiping their brows and sweating and commiserating among themselves.

And I'm wondering why, why are they so upset? Oh, my gosh, Medicare costs are going to go up. If you're going to live longer, Social Security costs are going to go up, the cost of our employer- provided healthcare is going to go up. You know, we have all these horrible things.

And the only reason I make this point is — the healthcare debate — and I know a number of people in this room do this, if you sort of go from the healthcare debate to the welfare debates, the education debate, stuff like that, the healthcare debate is in this world of its own. It's so plush. I mean, even at the Urban Institute, the people who get the most money are our health group. I mean, it is plush everywhere.


MR. STEUERLE: It really is. The money that floats around in the system — and it's because it's an open-ended system, open-ended on the exclusion, open-ended on Medicare, open-ended almost everywhere — and as Kate points out, and all that open-endedness is hidden.

So people don't realize that we are now — in fact, when I go to health people and I say, what do you think the average amount we are spending on healthcare is per household? And I'd say 80 percent of healthcare people never do the calculation. All you've got to do is take the HHS estimate and divide it by the number of households in the population; you come to a number of a $16,000 per household.

And people think, you know, well, nobody can afford $3,000, nobody can afford $4,000. Wait a second, on average, we are paying for — somebody is paying for this $16,000 somewhere. And as you mentioned, John, it's growing at an extremely rapid rate, and that growth is causing several problems.

One is it's actually decreasing the number of people who are insured. So when we spend more in a lot of areas — I mean, you can even talk about the Medicare drug benefit doing this indirectly — but you spend more and you open up the system and you make higher prices possible, we actually increase the number of privately insured people.

And maybe we compensate by trying to increase Medicaid and offset that a little bit. But even the studies on Medicaid say if you increase Medicaid, you decrease private insurance because employers have less of incentive to cover the people.

So this notion that we can buy our way out of the problem and keep these costs hidden is just to me — is to me silly. And one of the main reasons people talk about things like a cap on the employer exclusion is to try to not just get the incentives right, because you can debate with behavioral economics how much the incentives make, it's to get the information systems right so people say oh, my gosh, this insurance is costing me $8,000.

And then when Van says, you know, by the way, when your insurance goes from $8,000 to $9,000, you've got to pay the excess. We'll give you the $1,000 in more wages, but you will have to pay for it — or we'll keep descending, we'll bargain over whether your wages are going to go down. It makes the system much more explicit. I don't know whether HSAs actually get at that solution the right way. I have a lot of problems with them because the HSA advocates have decided they want to propose capital income exclusions along with everything else, and that sort of confuses the whole debate.

But it does at one level try to improve the information system, and that to me is a crucial part of the problem no matter where you come out on the political spectrum, is the information system has to be made better.

And let me just add one final note there on how tough that is politically. I put out a proposal about 15 years ago which was one way to do this, which you disagree with, Chris, because it uses the tax code, is if you don't have health insurance at the end of the year, you lose something like some other tax benefit. Maybe it's your personal exemption or something, so this is going to affect the middle-income people; it wouldn't necessarily affect the low-income people.

And at least one member of Congress I know who took up this proposal then later on lost in an election, so I don't know it's a very —


MR. STEUERLE: He is actually back in Congress now: Jim Cooper of Tennessee. But that's actually now become part of, whatever — I should think of the rest of the plan, of the Massachusetts health plan, which is to say — it's a very small thing. The question is if you really believe people have to buy insurance because we have this back-up policy, we are going to provide everybody a back-up Medicaid policy. So we are insuring indirectly — everybody, you and me, if we don't have insurance.

Then you got to go towards the notion of an individual mandate. If people are going to participate in the system, what can they pay? And for a long time, people have proposed individual mandates; it didn't work out how to do it.

So Heritage says, well, we'll deny people their driver's license or something. The problem is when you are at $16,000 average cost or $8,000 per policy because there is a lot that's not in policies, it's hard to put an individual mandate at that level on people; it's too high.

But you can incrementally say, okay, if you don't buy insurance, we are going to start taking things away from you. So once again, you recognize the costs. I say that actually is a little bit in the Massachusetts plan, whatever else you think about the rest of it. I think the main thing we got to think about, how do we improve these information systems.

But going back to what David said, and on to this point I think Sheldon made, is the tax code has provided a great — has hidden it very well, hidden your plushness very well so far. And that's the biggest part of the problem. You want to hide spending, put it in the tax code.

Howard — I'm trying to get to everybody here.

MR. GLECKMAN: Howard Gleckman of Business Week. Kate, I have a question for you. To the degree we are talking about the uninsured, as I understand it, and correct me if I'm wrong, the uninsured are basically low-income people who make a little too much for Medicaid but low-income working people, small business employees, and then as someone mentioned, the indestructible 20-somethings who just don't want to buy insurance at all.

As I understand it as well, the early evidence on HSAs is that the companies offering them are overwhelmingly large companies, not small businesses, that the people who are taking them up are overwhelmingly the healthy and wealthy. So that if that early evidence is true, how is expanding HSAs going to retarget the proposal, the program?

MS. BAICKER: That's a good question. First of all, the uninsured are a very heterogeneous bunch. As you have mentioned, some of the groups who are uninsured, there are also — you know, 55-year-old chronically ill people, there are kids who are a little too old and a little too high over the income distribution for Medicaid.

There is a really wide bunch of people who are the uninsured. And that's why a lot of the programs that target specific groups of those don't make as big a dent in the number of the uninsured as we would like.

But you do highlight a couple of groups who would actually be likely to expand the coverage under HSAs. There are a lot of facts floating around about who the preliminary enrollees are in HSAs, which companies are offering them, which individuals are purchasing them in the non-group market. The evidence is a little thin so far because these plans were only first offered in 2004, although enrollment had tripled over a less than a year period from a million to three million people covered by HSA-based policies.

So what do we know about who is enrolled? First of all, a third of the small businesses offering HSAs previously did not offer any insurance. So in the last five years, we have seen an erosion of employer-provided insurance from 69 percent of firms offering down to 60 percent offering. That's a pretty big erosion. One of the reasons that they claim is that the insurance is expensive and there is no basic plan available. So a lot of small businesses in particular find the HSAs appealing because they are basic plans that offer catastrophic coverage without a lot of extra expense for workers who might actually only want minimal plans.

About a third of the people purchasing HSAs in the non-group market were previously uninsured. I think about 40 percent of the people purchasing plans in the non-group market have incomes under $50,000. So a lot of people have the idea that only the healthy and wealthy benefit from them, and I don't think that that's actually true.

The reason the plans are appealing to low-income people is that the premiums are much lower. So people who focus just on the deductibles are missing half the picture.

The average HSA plan has a deductible that's about $3,400 more than the average, traditional PPO-type plan, but it has a premium that's about $3,000 less. So it doesn't completely offset the deductible. If you hit your out-of-pocket max year after year after year, you are not going to fare better under one of these plans. But if you have an idiosyncratic year of high spending, which is what — the majority of the under-65 population that has a year of high spending did not have a year of high spending the year before.

So there is a lot of — in the under-65 population, there is a lot of variability in your healthcare spending. In that case, you actually end up with more money in your account and less money — you know, less higher wages, et cetera, if you are under the HSA plan than if you are under the traditional PPO-type plan. Now, that's not the solution for the chronically ill, and they need extra help, as we have talked about before. But the enrollment patterns that we have seen in the HSAs suggest that they are fairly widely appealing, and that taking the premium savings into account goes a long way towards offsetting the higher deductible that people would see.

MR. BERGIN: Everybody bear with me, I'm trying to get to everybody. Let me get — let Jason jump in real quick.

MR. FURMAN: Sure. I promised the non-pithy facts, but I'll try to be a little pithier about them. In terms of the incomes of people in HSAs, there is one survey which shows 40 percent have incomes below $50,000. That's of the individual market, which is already disproportionately low-income people. Another survey shows 33 percent have incomes below $50,000. Another survey shows 29 percent. All three of those numbers seem a little bit below the percentage of people in the individual market who have insurance and make less than $50,000.

So all three of those numbers seem consistent with HSAs being disproportionately attractive to higher-income people. But I think the even more important question is not sort of do you have slightly more high-income people or slightly less; it seems like we have slightly more, but it's not positive. It's what sort of incentives are we giving.

As I said in the beginning, the $4,000 we give right now to a high-income person is more than we need to give them to get them health insurance. HSAs follow that same model by using tax deductions. Then they follow that model on steroids by giving you this $10,000 savings account that gets by far the biggest tax preferences of any savings account or tax system.

And by the way, you don't even need to use it for healthcare, as Kate pointed out. You can bequeath it to your heirs and you get this huge, sort of, super — beyond what a consumption tax advocate would want for a savings account.

So if you look at who is actually benefiting from HSAs, the take-up seems to be disproportionately high income but not hugely and it's very uncertain right now. The benefits are very tilted towards high-income people. And if you allowed people to contribute $10,000 a year instead of $5,000 a year, you tilt those benefits much, much more in that direction. That's in terms of the income.

Then the one other issue in terms of the uninsured, you know, a third of the people — well, the HSAs used to be uninsured — we have no idea what that means. I mean, a recent Cato report came out that sort of spent most of its time debunking what it argued were the myths that people used against HSAs. This was one of the ones that listed from the people who liked the HSAs. There is a tremendous amount of turnover in the individual market.

Most people getting insurance in the individual market — not most people — a lot of them, were uninsured before, we have no idea what a reference group is for this, what a comparison, whether that's high or whether that's low. And even in a model like Gruber's which predicts a net increase in the number of the uninsured, it also predicts a lot of people that used to have — the uninsured will get insurance and a lot of people that used to have insurance will become uninsured.

And in his model the net is negative. So you still might observe something like this fact. So I don't think it tells you anything about whether this is helping your uninsured problem.

MR. BERGIN: We'll go over here.

MR. CANNON: As the guy who wrote that Cato report —


MR. CANNON: Michael Cannon with the Cato Institute. And I have a question for Kate, and it's about something else I brought up in the report, which is I was hoping you could give the administration's thoughts on high deductible plans with HSAs and why the administration doesn't seem to be favor allowing more different types of insurance to be coupled with HSAs. The administration has proposed a tax credit but only for high-deductible plans, has proposed to allow money from the HSA to purchase to purchase health insurance, but only for high-deductible plans.

And I guess I asked because I agree with Jason on this point that it does seem inconsistent to say that consumers can shop around for routine care but they can't shop around for insurance.

So I think there is a very good argument to be made for expanding the flexibility of the form that these high-deductible policies could take. So one proposal might be instead of saying, you know, a deductible where you pay 100 percent up to $2,100, what about having 50 percent up to $4,200? You know, there is an argument to be made that that would cover a broader spectrum of the spending distribution and might have bigger effects in reducing moral hazard over consumption of healthcare.

So in the long run, I think moving towards actuarially equivalent but more flexible forms of policies would be a promising approach. That said, there is some reason to focus on high-deductible policies, and that goes back to your question about level changes versus trend changes.

Why are we spending so much more on healthcare today? What's driving the rate of growth of healthcare, not just the level? It's not that we are going to the doctor more, it's not that we are going to the hospital more, it's not even that we are staying more days when we are in the hospital.

It's that we are spending a lot more money on technologically intensive interventions when we are there, and technological acceleration drives our healthcare spending. Now, is technology bad? No, technology is good, it's lifesaving. And on average, we are all better off with today's medicine than we would be with 1960s medicine. Nobody is arguing that on average that isn't good.

The problem is that on the margin, the technology is not being applied in the highest value cases, and because of the insurance structures that we have now, there is little incentive to develop cost-saving technology.

So right now, if there were a new treatment available that did the same job as an old treatment but cost half as much, there is a diminished incentive to develop that new technology, to invest in that. And heaven forbid it's slightly less effective and costs only a quarter as much, you are not going to develop that technology, or there is reduced incentive to develop that technology. And that's a large part of the driver of cost growth over the long run.

So in an insurance system where people are more price-sensitive and you have removed the disincentive to develop cost-effective technological change — not just health-enhancing technological change, there is a spillover effect there. The traditional health insurance premiums should grow more slowly, too, as the expansion of price sensitivity and consumer evaluation of the costs and benefits of health technology brings more cost-effective technology online.

So that's one of the reasons you might want to think about focusing on policies that will make people more price-sensitive, because in effect, everybody benefits from that, and those spillover effects aren't adequately captured by the individual's decision to enroll in a high-deductible health policy versus a traditional health policy.

Thanks, Kate.

MR. CLAXTON: Gary Claxton with the Kaiser Family Foundation. Just a quick point. I'm not sure we talked about the non-group market. I don't think HSAs make non-group insurance cheaper. I think arguably they make it more expensive because you pay — I mean, there was already high cost-sharing for non-group insurance, and now you can pay for the cost-sharing with pretax dollars. So maybe there is some increased price sensitivity because of the savings account being able to save. But that actually benefits healthier people.

So I think it's really unclear that we could say that HSAs are affecting the uninsureds or non-group insurance by making it cheaper. It may be something more sexy to sell. Maybe people want to buy the insurance to get the account, but I don't think it's making insurance cheaper.

MS. BAICKER: And there's just been a little bit of sloppiness on our part, I think, in distinguishing between current HSA policy and proposed expansions. So some of the proposed expansions of HSAs would make the premium for the policy deductible in the non-group market. And I know I have been a little sloppy in mixing those up.

MR. CLAXTON: I'm wondering whether that's sensible to the question, too, but —

MR. BERGIN: I'm trying to get as many people as —

MS. BAICKER: The answer is yes.



MR. BERGIN: Get as many people as I can in here until we end. So try to keep your comments real quick.

Over here, please.

MR. MILLER: Tom Miller, Joint Economic Committee. I would like to see a little bit more of a marginal analysis perspective on where this discussion started, which is what the effects of tax policy are on health spending and its efficiency. We seem to have kind of wandered off in some other directions.

My question would be whether in fact we are beating a tired horse, that if you look at kind of what's going on in the marketplace, we did a lot of damage with the tax code. But on the margins, we are seeing employers not expanding their coverage to be more compreh
ensive. We don't see the marginal employee saying, hey, this is great tax benefit, I've got to kind of take that insurance.

In fact, we are seeing a slight retrenchment in the scope of coverage, because at a certain point even with a subsidy, it gets too expensive to buy even when it's subsidized. The analogy would be to Medicaid, where say except for states like New York, at the margins even if you're getting $0.50 to the dollar, you still don't want to spend more money because it just gets too expensive.

So my question is as a marginal analysis, is changing the tax code going to make a significant difference in the arc of spending we have already built in the installed base? Most of the propulsion for higher spending is the existing level of the tax subsidy. Even if you capped it right now, the effects of third-party payment are strong enough to keep driving that higher. So is this tinkering at the margins, given that in effect, we have already done enough harm with over-subsidizing through the Tax Code and we need to find different tools beyond that?


MS. CREECH: Hi, I'm Kathy Creech. I'm a tax lawyer, I'll admit it, here in Washington, and I represent employers and providers in the health system. And really, I just wanted to make an observation which I don't think answers your question but is along the same lines, just based on my anecdotal experience with employers.

And that is that they are very cost-sensitive, that in fact, the system may have — we may describe it as plush, we may describe it as something that hides the subsidy maybe from somebody who is sort of walking down 14th Street in terms of how much they think their healthcare costs.

But I'm sure any employers that we work with or any companies — you ask the CFO how much they are spending on healthcare and they can tell you to the dollar. And so the sort of the idea that the employer system has somehow hidden the cost, I guess from my own observation, I don't see that as being really on the mark. And what I do see anecdotally is employers — you know, they want a healthy workforce. They want to protect their employees from catastrophic events so that they have a healthy workforce, and that they are doing that in the best way they can in the current system.

And some of that is moving to high-deductible health plans. You know, whether that's the panacea, the answer to all these issues, you know, I don't know; I'm just a tax lawyer. But I do see employers using all the tools that they can in the current system to try to make their employees understand how much this costs and to try to make their employees more price-sensitive, because the employer is price-sensitive.

And that's through HSAs, it may be through flexible spending accounts, it may be through HRAs, or all the alphabet soup that we have in the Code, and certainly transitioning to something that may be a little more simple would be beneficial from a policy perspective.

But I guess to sum up, my main point would be that I don't think the subsidy is really hidden from an employer's perspective.

MR. BERGIN: From the employer's perspective.

MS. CREECH: Yes. And that they are sharing —

MR. BERGIN: Yeah, I would agree.

MS. CREECH: That with employees and trying to make employees understand that.

MR. BERGIN: But it's hard.

MS. CREECH: It is hard.

MR. BERGIN: I can say that from personal experience, because as a tax editor — and I'm proud to admit that, I didn't pay much attention to this — as the guy who has to make the decisions and deal with the CFO, I am very sensitive to it now. I think there is a breakdown between the employer and the employee, and it's real hard, at least I've found, to sort of convey the costs to the employees while at the same time you are trying to protect them. I'm sorry, go ahead.

MS. STRONG: Thanks. Again, Katie Strong. I think that on the employer side, certainly you can go and ask any CEO, they are very frustrated because they can predict their costs from year to year on everything they do: Their supply, their labor costs. The one thing they can't predict from year to year is what their cost for healthcare is going to be, and that gets increasingly frustrating.

Also in our current system, employers give very little choice. You may have two or three options to pick from, but you don't give a lot of choice back to your employees of what plan they get into or chose to get into. Because it's very administratively burdensome to offer a healthcare plan, I think we are going to see in the next few years — I don't know how long it will be — a shift more from a defined benefit to more of a defined contribution. Unfortunately, we can't really get there until we really do some reforms to the market itself.

The way our laws are structured around insurance right now is there is an inability on behalf of the insurers to provide innovative products because they are so regimented by the state laws that they have to operate within. And I think we need to take a look at the current structure and how we regulate and over-regulate and over- mandate the insurance industry before we can really start addressing cost.

MR. JENNINGS: I've really been trying to say what I was going to say other than I want to welcome the tax community to the healthcare debate. It's not so easy and so we are — at least we certainly welcome smart people to the game.

SPEAKER: You should introduce yourself.

I'm Chris Jennings. I'm a has-been, used to be in the Clinton administration, that continued to try to figure out a way to do affordable accountable healthcare in this system. And I don't think there is a magic bullet anywhere. But I'm going to try to make a quick suggestion about constructive ways you all can play in the upcoming debate, which are that — I think the issue of the tax cap concepts make some sense as a potential issue.

I think primarily because no one can figure how you can get new revenue into the system and we are going to have to figure out how to do that going forward in thoughtful, creative ways — I think it should be linked to meaningful, clinically-based benefit, not 100 percent co-payment for all services, but I think you can talk about that. The only thing I would beg of you if you do decide to play in this area is that you talk about re-dedicating all of those dollars to healthcare and not for other silly uses. Otherwise, it will blow up. It will blow up —

MR. FURMAN: Only your silly use.

MR. JENNINGS: No. It's our money, you steal it all the time, this is our money, so —

MR. FURMAN: Chris is still bitter about, like, Medicare and Medicaid cuts in '97.

Yeah, you recall though we were the reason why we had a surplus in 2001 through savings. But anyway, we'll come back to that later. Now that we've somehow lost that surplus, but it didn't get reinvested into healthcare that much, which I think is — I lament that quite significantly. But I would urge you to think creatively about that. And I think that when people talk about high- deductible plans or what's happening in the fee-for-service or whatever, I really think you can maybe combine interests here. I think we are going to have to have higher cost-sharing in many ways. I don't believe you need 100 percent cost-sharing.

I believe in some places, you are going to need to have lower cost-sharing to induce the type of incentives you want in prevention and other type of service. I just think we have models to go before we have reached the deal that we want to feel comfortable with, how we subsidize benefits. I would suggest that we move towards more of an evidence-based, clinical-based benefit, and then start linking how we subsidize healthcare going forward, and rationally do so.

And the last thing I'm just going to ask the tax community to think about is, if you do think about caps going forward and you do link it to benefits, I do think you have a real problem for older workers.

If you have just a straight cap and then that cap — the benefit may be a meaningful benefit, but if you link it to a dollar amount rather than to a benefit, then you will find that that dollar benefit will not be adequate for older populations, particularly if you go into an individual market rather than group markets.

So I'll conclude with that, and again, I welcome you to the debate.

It's a good way to end this. This was great. Thank you, everybody, we had a wonderful panel. I appreciate you guys coming; I appreciate the audience and your questions.

One advertisement/interruption here. We will launch our new Web site in about a week, so please check it out. It's Lots of good stuff on it.

Thanks everybody, and to our speakers.
(Whereupon, at approximately 10:38 a.m., the PROCEEDINGS were adjourned.)

* * * * *