August 12, 2004

August 12, 2004 | Lecture on Health Care

A Vision For Health System Change

ROBERT E. MOFFIT: One of today's biggest domestic policy questions is, "What should our health care system look like? More specifically, how do you bring about the vision of a better health care system?" Our first presenter will be Dr. Daniel "Stormy" Johnson, who is a practicing radiologist from Metairie, Louisiana.

Dr. Johnson is also a visiting fellow in health policy at The Heritage Foundation. He's been with us in that capacity since 1998. He is a former President of the World Medical Association and previously served as President of the American Medical Association. A Vietnam veteran, he received his doctorate in medicine from the University of Texas in Galveston. He is a Clinical Professor of Radiology at Tulane University and was co-founder and President of the American Society of Head and Neck Radiology. He received the Gold Medal Award from the Radiological Society of North American in 1997.

Our first respondent is The Heritage Foundation's own Vice President for Domestic and Economic Policy Studies, Stuart Butler. Stuart has been involved in the debate on health care policy since the 1970s. In the 1980s, Stuart led the debate over health care reform, arguing that we should have a system in America based on consumer choice and market competition. His manifesto, A National Health System for America, written with Ed Haislmaier, was published in 1989 and focused on the tax treatment of health insurance and how to change this into a national tax credit program to bring about a new, consumer-based health care system. Stuart has also written for such publications as The Journal of the American Medical Association, Health Affairs, The New York Times, and The Washington Post.

Our next presenter is Stan Dorn. Stan has been working on health policy at the state and national levels for almost 20 years, focusing primarily on low-income consumers, people on Medicaid, S-CHIP, and the uninsured since January of 2002. Stan has also been the Senior Policy Analyst at the Economic and Social Research Institute, Director of the Health Consumer Alliance, and Director of Health Policy for the Children's Defense Fund. Stan is a graduate of Harvard University and also the University of California at Berkley.

Stan's presentation will be followed by that of Dr. John Goodman, founder and president of the National Center for Policy Analysis. The Wall Street Journal called Dr. Goodman "the father of medical savings accounts." He is the author of seven books, including Patient Power--Solving America's Health Care Crisis, and received the Duncan Black Award in 1988 for Best Scholarly Article on Public Choice Economics. He received his Ph.D. in Economics from Columbia University and has taught at Columbia, Stanford University, Dartmouth, Southern Methodist University, and the University of Dallas.

Finally, Ken Thorpe is the Robert Woodruff Professor and Chair of the Department of Health Policy and Management at the Rollins School of Public Health at Emory University in Atlanta, Georgia. Ken has taught at Tulane University, the University of North Carolina, Harvard University School of Public Health, and Columbia University. Ken was Deputy Assistant Secretary for Health Policy at the Department of Health and Human Services during the Clinton Administration. In that capacity, he coordinated all of the financial estimates and program impacts of President Clinton's health care reform proposals for the White House. Ken has authored or co-authored over 80 articles and book chapters on health policies. He received his doctorate from Rand Graduate School, his M.A. from Duke University, and his bachelor's degree from the University of Michigan.

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

DR. DANIEL JOHNSON: Are we going to have changes in the health care system? Is it going to continue? Absolutely! But what will it be like?

Two Options

In the final analysis, there are two major options that we have: a true market system without central political control, price controls, or limited choice, or a Canadian-style single-payer system. I think the latter is a definite possibility. We have a situation now where everybody is unhappy and so we are going to go in one of those two directions.

If everyone is unhappy with the current system, that creates an opportunity for positive change. I think we need to take advantage of this. Will Rogers said it best, "If you find yourself in a hole, stop digging." We need to think about that. In terms of where we are, it is hard to find anything good about this system. There have been some important political advances in the last six months--and maybe in the year before that--but we basically have a Medicare program that is bankrupt and we have a Medicaid system that is bankrupting states everywhere. No matter how you calculate the number of uninsured, it is increasing at a rate that is unacceptable. We have physicians who are disgusted with all of this, and particularly with their inability to adjust their fees to compensate for rising costs. They are simply saying, "I'm not going to see any more new patients, particularly in these particular categories."

We have people who can't get access to care, so they go to the last place they should go--the emergency room. Throw on top of that the skyrocketing costs of medical professional liability insurance premiums and you have a very volatile mix. Nobody cares about the doctors dropping out of practice until someone needs to go to the doctor.

Market-based Solutions

Is there any hope for this? Professor Regina Herzlinger of the Harvard University Business School says it is the business community that can solve this and I am inclined to agree with her. But, before we try to fix this, we ought to figure out what the problem is.

Our health care system is disjointed and disconnected. The majority of us in this room understand what the problems are and what we ought to do about them, but we do not tie that together in a message that is understandable by the average person. I don't think we are reaching the constituency. We can discuss it among ourselves, but we can't discuss it with the people who really count.

Let me ask you some questions. Should everyone have health insurance? If so, why? If not, why not? Is this a question of "This is the right thing to do because of altruism"? Or is it the right thing to do because of common sense?

In Louisiana, there is a law that says that if you own an automobile, you have to have liability insurance. About 20 percent of the people in Louisiana can think of better ways to spend their money than buying liability insurance. Consequently, the rest of us who do carry liability insurance pay much more for our insurance than we would if everyone obeyed the law.

I suggest that the same thing occurs with health insurance. Those of us who have insurance and/or provide it for our employees pay more for it than we would otherwise pay if everyone had insurance.

Who ought to pay for the insurance? Who should pick individual insurance plans? Should it be government, the employers, or the people who actually use it? You say you also want to control costs. What do you think about price controls?

What about the notion of "First Dollar coverage"? How do you think that is related to price control? Most people in business don't like "First Dollar coverage," and yet, business went to something called "pre-paid care." Pre-paid care is not a term that critics of the HMO industry use--"pre-paid care" is an HMO industry term. Is "pre-paid care" any different from "First Dollar" coverage? Is it any wonder that it did not succeed in solving the cost problem? Who ought to fix the prices?

The Cost Problem

What is clearly driving the mess we have is cost--and costs are continuously going up. A recent study says the cost is going over 15 percent of GDP. Cost is significant. Just two months ago, Drew Altmann said in an article in the Washington Post that now we have this feeling that there is nothing we can do about this and that people are tired of the "market-driven" strategies. Mr. Altman apparently thinks that we have a "market-driven" system, and that we have been pursuing "market-driven" strategies. I've already spoken to that. We don't.

We could go through a list of what is causing the cost problem and have discussions on any one of those things, but I want to suggest to you the reason we have a cost problem is that the person consuming the services is insulated from the cost of those services by third party payment. Anything that doesn't address that is doomed to failure.

The person who is buying the services is not the person who is using the services. The patients themselves are insulated from the cost. The employers react to this in whatever way they can. With respect to physicians, we have had deflation in prices. In my own practice, I have not raised prices--either the actual prices or the discount prices--for 10 years. Yet my employees have this radical idea that every year that they should get a raise. Sooner or later the lines cross and you shut the door. So we have deflation and consumer price insulation--a volatile mix to say the least.

Pain and Power

There are two issues that I want to discuss with you. One is pain and the other is power. What I would like to do first is present the Bob Moffit "Pain Soliloquy" here. We are all suffering pain in the current health care system. How do we respond to this problem of pain?

Liberals say they feel your pain.

Moderates want to share the pain.

Conservatives deny the pain. "What pain?"

What is your attitude toward pain? I'm a physician and I can tell you that I don't like pain. Why don't we just get rid of it? Why don't we try something that gets rid of the pain? I think that is possible. It is not possible in all venues, but it is in health system reform. As a matter of fact, we can do that but for one thing. And that is power.

Power has presented us with a classic confrontation between the control freaks and those who would allow people to make decisions for themselves. Who are the players in all this? They comprise the same list, but with the patients at the bottom. Patients are currently incidental to this process and we need to reverse that in my view: Put the patients at the top.

Is there a better vision for reform? Can markets work in health care? Can you have a market if the person who is using the market is insulated from the cost of the goods and services in the market? Can you have a market where you have no choice? Which market would work better? A market with limited choices or a market with expanded choices?I would like to propose the following: that we have three bullets for our vision for health system reform. It's not anything new or different.

  • Expand the choices.
  • Allow individual selection and, where appropriate, ownership. This applies to Medicare, Medicaid, and the private sector.
  • Defined contribution. Have the government in the public sector (and the employer in the private sector) put up the same amount of money no matter what choice the person makes.

Those are the three points of vision. Focus your attention on how to accomplish that vision by doing the things that are necessary.

Paternalism and Pooling

Let me just speak to the three bullets quickly:

The notion of personal choice is very important both in expanding the choices and different kinds of delivery mechanisms.

There are a variety of financing ways and any type of system that expands the choice should allow for innovation. It should not stifle innovation and should not prejudice one of these against the other. The notion of individual selection and ownership eliminates the portability problem. If people can choose and own their own insurance they can take their insurance with them from job to job. They should have their periodic right to change if they don't like the choice that they made last time. The notion of defined contribution means that the one providing the subsidy provides it the same no matter what choice the person makes.

Another barrier is paternalism--the notion that you are unable to make decisions for yourself, while on the other hand, I am far more capable than you and willing to do so.

The next notion is pooling. The Federal Employees Health Benefits Program (FEHBP) works like a voluntary choice cooperative (as opposed to a voluntary purchasing cooperative): It sets the rules and then gets out of the way. Let the plans be qualified as to whether they are solvent or whether they have truth in advertising, but let people pick whatever they want without trying to drive them in one direction or another. Therefore, I offer the term "Voluntary Choice Cooperatives." The FEHBP basically does that. It has outperformed the private sector over time, which I think is important to point out. That was with no consumer driven option in the choice of plans. That shortcoming has been corrected by Kay James, Director of the Office of Personnel Management (OPM), and consumer driven options will be increased this year and possibly next year.

Single-payer System

As Alain Enthoven has pointed out, a single-payer system cannot stand up to a properly constructed market system. What do doctors think? Because of the hassles involved in practice now, a significant and growing number of physicians would go to a single-payer system. That is something you need to understand because it is a part of the dynamic for potential change. The inefficiency is unbelievable and the system is growing more and more complex: That seems to be the American way. We need to address that and government has a role there. Set the rules and get out of the way. Make the system simpler.

Tax Credits

I want to speak about tax credits. The Heritage Foundation's senior health policy analyst, Nina Owcharenko, has shown us in a clearly written paper how to use tax credits to accomplish universal coverage and how to use it to accomplish a defined contribution. Yet, for ordinary folks, tax credits are very arcane. We have to make some changes in the tax laws and use the tax laws more creatively. The tax credit situation is a detail and a means to an end in my view. It confuses the vision when you run with that as the lead solution.

What is the outlook for success? I think we need to focus on diversity and figure out a way to celebrate our diversity and to understand what a potential benefit it is to our vision. So celebrate diversity, leverage diversity, and use it to make possible a simple vision of expanded choice, individual selection, and ownership where appropriate. Require the government in the public sector, the employer in the private sector, or any subsidy giver to provide the same subsidy no matter what choice the person makes. Center on those three items as a simple vision that we could all agree upon--based on what we all believe. Then spend time working on how to enact that.

Remember that the whole notion of consumer-driven health care really boils down to something that I have advocated for a long time, which is to put the patient in the driver's seat. Instead of thinking that people are too stupid to make decisions for themselves, put that person in the driver's seat. Because it is a complex world, the patient needs to be able to look to someone for help--so I would add, "with the doctor riding shotgun."

STUART M. BUTLER: I want to concentrate on the steps and the strategy we might take in trying to take these elements of the vision that Stormy laid out and to move in that direction in broad policy ways.

Operating Principles

I think of four operating principles--from a public policy point of view--that would help us move in that direction.

First, assistance should be focused on those who need it most. We should be focused on making sure that these people have enough insurance to protect the rest of us from them.

Second, the medical care, insurance, and coverage choices that people have should not be dependant upon their place of employment. The notion that you are playing roulette every time you change jobs and hoping that your employer will provide you with the choices that fit your particular needs is a thing of the past. We should be looking at a system in which the choices are not dependent upon your place of employment.

Third, we should think of the place of employment as the place in which you do the bookkeeping--where subsidies are available to you from the government or from the employer. Yet that workplace is really a convenient location for making decisions that you want to make, not the place that determines what your coverage is.

Fourth, I think it is important to look at the wonderful system of federalism that we have in this country, which allows and encourages states to fine tune the kinds of structures that would work best for organizing insurance--to the extent that that should be done. I think if you look at it that way, you can then start broadly applying these kinds of principles to the task at hand. How do we move from here to there?

I think it is important to recognize that within this poor system, we can--in terms of the working population--break this down into three parts. If you work for very large employers you certainly need a reformed system, but you generally have coverage and you have more choices than most people have. If you happen to be in a government program, big reforms are needed (by people in those programs who are eligible) to at least have something they can depend on. If, however, you work for the small-business sector, the notion of getting insurance--particularly through the place of work--is a metaphysical concept. It does not exist. It is a massive, dysfunctional system for people in that particular sector. In my view, if we are going to move from here to there, it is that sector that is the best place to begin focusing on "un-insurance."

Employer-based Coverage

If you look at the small-business sector and compare it with other parts of the current system, only 55 percent of firms with 10 or fewer employees offer coverage--compared to about 99 percent of very large employers. If you do not have coverage through your place of work, you get little or no assistance at all--unless you enroll in government programs. People who run small firms know this. They know the headaches of trying to organize and obtain coverage for employees and the problems associated with that. It is a pretty dismal system. That is why I think it is important to start with that small-business sector.

How could we do that? I would suggest three strategic elements.

First, money. Stormy mentioned defined contributions: giving the same assistance to everybody. I think that when it comes to government assistance, it is very important to move toward giving equal assistance--or at least, to give more assistance to those who most need to obtain basic care.

Today we have a system in which the cost of tax expenditures associated with our assistance is about $188 billion by the latest measurement. Each year we provide various tax breaks to people with which to obtain insurance. Most of that goes to people who have higher incomes and have very elaborate health systems. The people who do best in the current system are rich hypochondriacs. The people who do the worst are the people who are poorly paid and actually need assistance.

It is very important to move in the direction of changing this. I think the way forward should start in a practical way: by looking at the various proposals that have been mentioned--tax credits and refundable tax credits (which are really vouchers run through the tax system).

We need to begin to focus that kind of help on people--particularly in the small business sector--who really need help. When people receive that assistance, they should not be restricted on using that help to pay for insurance through their place of employment. They should be allowed to obtain it from anywhere they wish. In other words, we need to break away from the current system of employer-determined coverage toward a system in which people are actually making the choices that they want.

Second, I think it is very important to encourage a much wider degree of diversity and choices on intermediates and on the organizations through which one can get help. For most working people today, it is pretty much only the employer. Yet I am in favor of looking at steps to encourage the states to take the lead in this and form all kinds of different arrangements and all kinds of different reinsurance systems, pools, and intermediates. Remove the barriers--including the tax barriers--to those kinds of alternative organizations.

Community-based Organizations

I think it is very important that we encourage community-based organizations to be intermediates as well. When you look at the population of the uninsured, particularly among minority populations, it is largely African-Americans and Hispanics. It is very important that we enable people to deal with doctors and hospitals--not by just by looking at the "Yellow Pages" and trying to strike their best deal, but also by being able to go through intermediaries that they trust and that can act on their behalf. For example, the FEHBP has unions that carry out this function. I am also in favor of looking beyond unions--to churches, to farm bureaus, to minority organizations, and so on. I think we need to take steps to remove the regulatory and other barriers. I believe that a system of tax credits would increase the demand for services through these kinds of organizations and would cause them to flourish and expand.

Third, we need to rethink the place of employment in terms of getting insurance. Under the current system, the employer determines whether you have insurance and what you shall have. I would like to see the workplace become a shopping mall for the coverage that you want and the kind of service that you want. In order to encourage employers to facilitate these changes, we should allow payroll deductions, a change in withholdings (to reflect tax changes and subsidies), and maybe even the ability to put money through some contribution to your choice. The consumer makes the choice about the type of coverage and which organization he or she wants to go through. He or she would simply inform the employer, who would do the paperwork. That is how I see the future--as an evolution of the employer-based system.

In closing, I think it is important to think not only about where we want to go and exactly how we want to get there, but also to think about how people view the health care system.

One of the dirty secrets of health care policy is that while everybody complains about the system, Americans are generally disinclined to make big changes. They get very nervous. They don't want to see big disruptions. Therefore, I think it is very important to move forward with that in mind. If you move forward, it is very important that the first steps look much like the current system. That is why continuing to think of the workplace as the place where you sign up is a small, but important, step. Still, to a lot of people this looks like more choices, more individual power, and more diversity. That is why, politically, it is a very important step to take.

STAN DORN: I am going to focus on a key policy mechanism supported by many advocates of market solutions--that is, refundable/advanceable tax credits to subsidize the purchase of health insurance by the uninsured. I am going to ask the nitty-gritty, empirical question, "How can we make them work?" This question breaks down into two further questions, which I will discuss in turn. One is, "As a matter of policy, how can we make sure the credits effectively cover the uninsured?" Second, "As a matter of politics, how can we make sure that bipartisan support is available?"

Tax Credits

First, how can we make sure that tax credits are effective? The short answer is that we have no idea. In the early 1990s, the so-called Bentsen/Child Health Coverage tax credits went into effect for literally one year before they were repealed. That was not a happy experience. Very few people took advantage of the credits, horrendous marketing fraud was reported, and there were lots of administrative inefficiencies.

Now we are engaged in our country's second experiment with health insurance tax credits, for the first time by furnishing credits directly to insurers during the year, when premiums are due. The Trade Act of 2002 established 65 percent health coverage tax credits (HCTCs) that pay for private health insurance purchased by some workers who have been laid off and certain early retirees. Advance payment directly to insurers began in August of 2003. We do not know whether this program will succeed or whether it can overcome the obstacles it faces. Early on there seems to be low participation, but we do not know if this problem is going to go away with time. We do not know what the causes are. There are a lot of ideas out there but it is still very early in the program's operation.

In the next month or so, we are looking forward to finishing and releasing a report, funded by the Commonwealth Fund and the Nathan Cummings Foundation, that will take a look at some of this early implementation. However, the short answer is that between the Bentsen credits and the Trade Act credits, we have not yet managed to make tax credits succeed in covering the uninsured. That is a problem.

Here is another problem. There are disagreements about what it means to succeed in covering the uninsured. There are disagreements about what kinds of health insurance policy the uninsured should receive. For example, do we want to encourage high-deductible plans or more comprehensive coverage? Similarly, intelligent people of goodwill disagree about the importance of limiting tax credits to the uninsured or the desirability of providing the same subsidies to similarly situated people regardless of whether they purchased insurance in the past.

I am going to ignore these fascinating questions, and instead make one obvious point. Regardless of how you feel about these important issues, health insurance tax credits need to be taken up by eligible people if they are to accomplish the goal of providing health insurance to the currently uninsured. That question--how to design health coverage tax credits so that they are used--will be the focus of much of my time this morning.

Thanks to the Institute of Medicine's (IOM) comprehensive synthesis of the peer-reviewed literature, we know that health insurance coverage makes a dramatic difference to the prompt detection and effective treatment of such chronic illnesses as cancer, heart disease, and diabetes. According to the IOM, 18,000 Americans die prematurely every year because they lack health insurance. Why does this happen? Many have no affordable choice but to gamble with their health, delaying going to the doctor and avoiding filling all their prescriptions, hoping that they somehow gain insurance coverage before health problems degenerate into emergencies that require hospitalization and endanger their well-being.

The challenge facing advocates of market-based solutions is to show how non-coercive policies with significant consumer choice can gain adoption and then achieve significant progress in reducing the number of uninsured. I am convinced that this can be done, but it will require significant commitment of resources, both intellectual and financial.

Achieving Take-up

In order to achieve take-up, I think one important issue to look at is enrollment mechanisms. We are all grateful to Lynn Etheredge of George Washington University for bringing us information about what has happened with 401(k) plans, in which enrollment context makes all the difference. Roughly 30 percent of eligible people enroll in individual retirement accounts (IRAs) on their own. However, over 80 percent of employees eligible for IRAs enroll if the employer offers automatic enrollment. In other words, the exact same tax benefit yields very different take-up levels, based simply on ease of enrollment. The bottom line is that the harder you make it to take up health insurance, the fewer people take up health insurance.

Yet that is not the only thing that matters. The size of the credit is also critically important. In 2003, average employer-sponsored health insurance for an individual worker cost a little less than $4,000 per policy. The individual worker, on average, had to spend about $800 a year out of his or her own pocket to purchase that coverage. With a tax credit like the Trade Act HCTC covering 65 percent of premiums, in order to purchase average employer coverage, the individual worker would have to cover the 35 percent remaining share at a cost of approximately $1,400.

Even if the worker selects a policy worth only 80 percent of average employer coverage, the worker's 35 percent share would amount to $1,120. With a less generous $1,000 tax credit, the worker would need to pay $2,200 a year, or nearly three times what that worker would pay for a more valuable employer-sponsored policy.

The question is: Is that feasible? Put differently, will many uninsured Americans pay those amounts left uncovered by tax credits? It depends on who you are looking at. Some of the uninsured have high enough incomes that they could afford to pay thousands of dollars a year for worker-only coverage, even after receiving help from a tax credit. However, about two-thirds of the uninsured have incomes under 200 percent of the federal poverty level and have limited ability to come up with even $1,000 to pay for health insurance.

It is not easy to determine what the percentage of poverty level is below which you cannot come up with that much money. Any measure of poverty is an over-simplification, and the ones that we have today are no exception. The same percentage of poverty can purchase very different amounts of goods and services in New York City than in Iowa City, for example.

The variations obscured by a single, national poverty level go far beyond geography. For example, if you have to pay for child care, you have less money available for health insurance than if your child is old enough to take care of himself or herself or if you are childless. If you are receiving public benefits like food stamps or Section 8 housing certificates, you have more money available to purchase health insurance than if you have the same level of earnings but do not receive any public benefits.

Mercer Study

There was a wonderful study done by researchers at the University of Washington and at Mercer Consulting, using funding provided by the Health Resources and Services Administration. They looked at 576 different types of households in eight Washington counties--high-cost areas like Seattle and low-cost, rural areas. They looked at families with kids, families without kids, individuals receiving and not receiving public benefits, and individuals who were both sick and healthy. They asked how much money would it cost to meet just the basics--food, shelter, and clothing--and still come up with $120 per year required as the minimum premium contributions to join Washington's health coverage program for low-income workers, the Basic Health Program.

It turns out that very few households with income below 150 percent of the federal poverty level can pay $120 a year for health insurance and still meet their other basic needs for food, shelter, and the like. Every household type, in every location, needed to have income above the poverty line in order to pay $10 a month for health insurance--and still pay for clothes, housing, and so forth. Yet in most contexts, the vast majority of profiled households had to have incomes over 150 percent of poverty--and in many cases over 200 percent of poverty--in order to have the discretionary income required to pay even $10 a month for health insurance.

That is consistent with a lot of research that suggests that low-income households are very price-sensitive when it comes to the decision to take up health insurance; maybe not higher-income or middle-income households, but certainly low-income households. Studies by researchers at Lewin and Urban suggest that if even 3 percent of household income must be dedicated to insurance premiums, then less than 40 percent of eligible individuals take up insurance. If you have a four-person family at 200 percent of federal poverty level, the 35 percent cost discussed previously would take up 3 percent of family income to buy coverage just for the head of household--which means less than a 40 percent take-up rate. Additionally, the $1,000 tax credit would result in consuming 5 percent of family income, which means you would have even lower take-up rates.

To use plain English: if you want tax credits to be taken up so they have an impact on the low-income uninsured, the credits need to placed in an administrative context in which they are easy to get, and they must be sufficiently large that low-income households are not required to pay very much.

Lessons from Medicare and Medicaid

I suggest that in order to ensure that tax credits actually enroll uninsured individuals of modest means, you need to think about enhancing the level of subsidies for the lowest-income consumers who qualify for the credits. Yet how do you identify the low-income people who need extra help? Relying on year-end tax forms may not get the job done, because family income can fluctuate dramatically during the year. The Medicare bill provides an interesting model, because Medicare is not in the business of assessing current income, just as the IRS is not in the business of determining how much income a family is earning right now. The Medicare bill says that if you want low-income subsidies, you have to go to your local Medicaid agency or the Social Security agency, which is already in the business of means-testing; demonstrate your income; demonstrate your level of assets; and then if these agencies say you are eligible, you qualify for the low-income subsidy. We could do the same thing with tax credits.

In fact, while building on the Medicare bill, we could take it one step farther. The Medicare bill commanded these Medicaid and Social Security agencies to develop simplified application procedures. But the state agencies that run the State Children's Health Insurance Program (known as "S-CHIP") already have simplified application procedures; that's one fewer set of wheels you need to reinvent.

So here is how it would work. If you have a basic credit of 65 percent and an individual wants a higher level of credit because they have low income, he or she would have to take the responsibility to go to the S-CHIP agency, demonstrate his or her low-income and then get an enhanced level of subsidy.

I'll discuss one other policy issue beyond take-up and that is marketing fraud. We have had moderate problems with Medicare HMOs marketing fraud to seniors. There have been severe problems with the Bentsen/Child Health Tax Credits and with Medicaid HMOs, and if we are not careful, horrendous marketing fraud could be a problem with health insurance tax credits.

I'll give you one example from California. Literally for years, the newspapers were filled with stories of employees or contractors with Medicaid-participating insurance companies going to Women, Infants, and Children (WIC) programs giving food to poor pregnant women and children and saying, "Sign here or else you'll get deported"; "Sign here or else you lose your Medicaid"; "Want some booze? Sign here, and we'll give you a free bottle." Finally, Governor Pete Wilson decided to implement what many Medicaid programs across the country now do, and that is to contract with a private-sector, neutral enrollment broker. In that case, it was Maximus. Insurers could continue to market on TV, billboards, and so forth, but if the individual wanted to sign up for a particular Medicaid HMO in California, he or she had to be signed up by Maximus. That solution worked, in a way that criminal and administrative sanctions did not. It pretty much stopped the marketing fraud in California. You could build in similar mechanisms to tax credit plans.

There was a tradeoff involved, which would apply to tax credits as well. Obviously, the kind of policy that Pete Wilson put in place would prevent insurers from employing the full range of marketing techniques in reaching out to eligible individuals. Under that approach, insurers can provide information, but they can't actually go to Joe Sixpack, sit down with the forms, and sign him up along with his kids. That means there is the potential for less take-up. There is unquestionably a tradeoff there. But you don't want to advocate for tax credits and then, a year or two later, see headlines in all the newspapers about tremendous marketing fraud. That is not in the long-term interest of people who favor market solutions for health coverage and, more important, it is not in the best interests of the uninsured individuals whom tax credits are supposed to help.

Politics

Let's be clear: You can try to pass health reform without bipartisan support. Obviously, policy has been enacted in the U.S. with support from only one party, but I think we all agree that you are more likely to get coverage expansion and market-based reforms enacted with bipartisan support.

I'm going to talk about two strategies for getting bipartisan support, if that is the route you want to pursue. First, tax credits could be coupled with public program expansions to serve the very poorest uninsured. Wharton School economist Mark Pauly has proposed that persons under 125 percent of the federal poverty level should get public health insurance free of charge. This means zero premium costs for Medicaid programs, S-CHIP programs, other public programs, or public employee programs. I think that Mark has recognized that if you do not have money it is not realistic to ask that you need to spend money to get health insurance. Conservatives who would resist any move toward the single-payer system can nevertheless draw a principled distinction and say that, to cover the very poorest Americans, it makes sense to employ a program like Medicaid, which imposes very few health-related costs on those who lack the ability to pay.

Let's get a little more specific about the kind of public program expansion that could be supported, in good conscience, across the political spectrum and coupled with tax credits. We know that 36 percent of the uninsured have incomes below the federal poverty level. What few people realize is that 52 percent of these poor, uninsured Americans are not children. They are not parents. They are childless adults. Why is that the case? Unbelievably, federal Medicaid law forbids state Medicaid programs from providing coverage to childless adults unless they are pregnant, severely and permanently disabled, or elderly. Empty-nesters are included in this prohibition as well, since they are not currently caring for a dependent child who lives at home.

Kaiser Commission Study

There are a few states that have obtained waivers or used their own dollars to provide coverage for childless adults--14 states plus the District of Columbia, as of the first of this year. In the next couple of months we are hoping that the Kaiser Commission on Medicaid and the Uninsured will publish a study we have done looking at eight different states that have enacted coverage for childless adults.

What we found is very interesting. First, there is tremendous take-up by the near elderly, these very empty-nesters who couldn't have Medicaid coverage. About a third of enrollees in childless adult programs were in their 50s and 60s. It was just remarkable. There are political as well as sound policy reasons for helping those older, uninsured adults who are at particularly great risk of serious health harm if they do not have health coverage.

Second, we found that there was tremendous political support for these programs. In several states, both Democratic and Republican governors facing budget crises proposed eliminating coverage for childless adults. In all of those states, Democratic and Republican legislators said, "You're out of your mind, Governor. If you have two equally needy and equally hard-working families, how on earth can you justify subsidizing one and not the other only because the kid living in one home is age 17 and the kid living in the other is age 22? That makes absolutely no sense at all. We want to support people who work. We want to provide assistance to people who can't pay for coverage themselves."

Without exception, these states rejected singling out childless adults as proposed. Instead, policymakers spread cuts more evenly across the beneficiary population. That is a slight over-simplification, but the bottom line is that there may be more political support for doing away with this inequity than many realize.

How could you fix Medicaid so this is not a problem any more? At a minimum, you could give state Medicaid programs the option to cover poor, childless adults just like they cover children and parents, by changing the state Medicaid plan. Don't require state officials to come to Washington--in former Wisconsin governor Tommy Thompson's words--"on bended knee to kiss the federal ring and request a waiver." Let them just do it by changing their state plans and automatically qualifing for federal matching funds. There are lots of different ways you could structure it and different levels of federal funding that you could provide, but the bottom line is that you could give states the option to cover this group. That is one strategy to help some of the very poorest uninsured.

A second strategy is to target credits carefully, and I think Stuart's proposed targeting makes tremendous good sense: Focus on low-wage workers in small firms, but have the credits be used in group settings with key market features, perhaps using as a model such as the Federal Employees Health Benefits Program, affectionately known as "FEHBP."

Getting Bipartisan Support

To get bipartisan support, you have to think about why some people hate the non-group market. Some people hate it because it seems to impose high prices and low benefits based on factors entirely outside an individual's control, such as age, gender, and prior health history. It just seems wrong to some people that, because factors outside your control mean that you really need health insurance, you are going to get pre-existing conditions excluded or that health coverage is going to cost four times as much. The other concern is for the dynamic scoring effect, that is, over time the better risks are funneled into one part of the market, and the other part of the market keeps all the worst risks, which increases prices, driving out the best remaining risks, which drives up prices further, etc.--a death spiral, in short.

Now, I think that with FEHBP, you can get a lot of the key features of market-based reform without running into those problems with non-group coverage. In particular, market pressure--rather than detailed government regulation--is the primary force in driving private plan decisions in FEHBP. Is FEHBP open to innovation? As Stormy noted, this last year we saw consumer-driven health coverage put on the menu of FEHBP-covered options, and health savings accounts are likely to appear there soon. The contrast between Medicare and FEHBP is dramatic in terms of the range of changes that private insurers have made in FEHBP versus the difficulty Medicare has had in changing regulations and statutes. Everyone knows how difficult that is. With FEHBP, you have much more openness to health plan innovation, significant consumer choice, and significant (but not unlimited) variety in health plan offerings from which consumers can choose. Also, with quite a range of different sorts of plans, there are incentives for consumers to buy less-costly coverage. If you pick a more expensive plan, you have to pay more. In other words, there is an incentive to select less-expensive coverage, but consumers are free to choose more expensive plans if they think it is worth the extra cost. Sounds like a market, doesn't it?

There are a lot of liberals who like FEHBP. It provides comprehensive benefits--or at least it gives consumers an option for comprehensive benefits. They may not have to take it, but they have access to it. The coverage you are offered and the price you pay is not affected by age, gender, or prior medical history. Issue is guaranteed. There are ways one could make this thing work, using a FEHBP-type model to give tax credit beneficiaries and others access to a real health insurance market without feeding into liberals' nightmares about the non-group market.

Second, you could charge new enrollees based on premiums for federal workers. There are mechanisms you could put into place to prevent adverse selection and to prevent a death spiral. The Heritage Foundation's Bob Moffit has recommended back-end risk adjustment invisible to the consumer, which has also been recommended in other contexts by analysts from the Urban Institute. Therefore, I think there is potential bipartisan support for this important mechanism. The bottom line is that for people who want to see the country move toward a market-based system of health care, one could have bipartisan support for a policy that coupled giving states the flexibility to cover the poorest, uninsured Americans and establishing a system of refundable tax credits to cover low-wage workers at small firms, as discussed by Stuart, giving such workers (and perhaps others) access to a broad range of choices in a real health insurance market. I don't see any reason why we couldn't get it done.

JOHN GOODMAN: I also want to talk about Stormy Johnson's idea that government should make a certain amount of money available and let people choose where the money goes. I have a completely different approach to this.

Two Policies

We have basically two sets of policies. We have tax policies, and we have spending policies, and they are totally disconnected. On the tax side, we have these very generous subsidies that Stuart Butler talked about, which, for a middle-income family, stack up something like this: The ability of the employer to pay premiums instead of wages, not taxed, means those premium dollars escape, say, a 25 percent income tax, a 15 percent FICA tax, and maybe a 5 percent state and local income tax. So government is really paying for about half that cost of the insurance.

The problem is that most of the people who are uninsured do not have the opportunity to get that kind of subsidy, so if they buy insurance, they have to do so with after-tax dollars. For a middle-income family, buying insurance with after-tax dollars means that, effectively, you have doubled the cost of the insurance. You have to earn twice as much money to be able to pay the taxes and buy the insurance with what is left over.

So we have very generous tax subsidies for some people to get insurance and virtually no subsidy for other people. Yet what if they choose not to buy the insurance? What happens to them? Then we are over into the "Public Safety Net" part of our health system, which no one has really talked about this morning. What do we do for people who are not insured and who cannot pay their health care bills? It turns out we are fairly generous to a lot of them.

The Example of Texas

The state of Texas has done perhaps the most comprehensive study trying to determine how much Texas spends on free health care for people who cannot pay their bills. It turns out it is about $1,000 per uninsured person per year. $1,000 per person is $4,000 for a family of four, and in many Texas cities you can buy a private insurance policy for a family for $4,000.

Admittedly, this safety net money is often wasted. It is spent on lots of different programs, and they are overlapping and duplicative, but nonetheless, we are spending quite a lot of money for free health care for people who are not insured.

I do not know of any other state that has been as thorough as we have been in Texas at trying to add up all the dollars and cents, but there is an Urban Institute study that makes national estimates, and those numbers are pretty consistent with our numbers in Texas. A recent article in Health Affairs estimates that the uninsured really only pay about one-third of their health care costs, and the other two-thirds are paid from these other safety net sources. That, if you add it up, looks like it is consistent with the $1,000 number.

What does all this mean at a practical level? I will concede that it is very different in different parts of the country and in different communities, but here is how it works in Dallas, Texas. If you are uninsured in Dallas and you do not have much money and you have a health care need, you are likely to show up at the emergency room of Parkland Hospital. Parkland is also where you are likely to go if you are a Medicaid patient, and it just so happens that the uninsured and the Medicaid patients enter the same emergency room door. They see the same doctors. They get the same treatment. If they are admitted to hospital rooms, they are admitted to the same rooms on the same floors and, again, get the same type of care.

Next door to Parkland Hospital is Children's Hospital, which is where children would go. If you are an uninsured child or a child on S-CHIP or a child on Medicaid, you are likely to be at Children's Hospital--again, going through the same emergency room door, seeing the same doctors, and getting the same care.

Now, you might wonder what difference it makes whether you are signed up for one of these programs or not signed up. As far as the patients are concerned, it is clear they do not seem to think there is much difference. To the hospital, however, it makes a great deal of difference. They have paid staff that literally go through the emergency room patient by patient, family by family, trying to get people to sign up for one of the programs they are eligible for. Interestingly enough, more than half the time, they fail to get a signature on the dotted line. After patients are admitted to their hospital rooms, they literally go room by room, still trying to get them to sign up.

The patients are not in a big hurry to do this. As it turns out, in Texas--this is also true of many other states--you have three months, if you like, to sign up and still get the bill paid for. So there is no reason to sign up at the time you receive care. You can always sign up later.

What I am describing to you is a system that is very much like the health care system that exists in Toronto or the health care system that exists in London--except that in London we claim that everybody is insured, and in Toronto we say everybody is insured. In Dallas, we say there are a lot of people who are uninsured.

In all three cities, we have a lot of people going to the emergency room for their health care. In all three cities, we have a lot of people who do not pay for their care. In all three cities, we have a lot of rationing, particularly rationing by waiting. Yet I would guess that at Parkland Hospital, people probably get more care and better care than they in Toronto or in London. At Parkland Hospital, if you walk in with a migraine headache, you might get an magnetic resonance imaging (MRI) scan. I bet they have never given an MRI scan to a patient with headaches in Toronto.

What the Studies Indicate

We have had a number of studies about what happens to people who are uninsured--I would guess more than 100--and there are problems with these studies. First, none of them are very good because they do not really distinguish between who is really uninsured and who is de facto uninsured (people who just haven't bothered to sign on the dotted line).

Second, they are not very helpful because they are really studies of unmet needs, and the only thing you can do with studies of unmet needs is use them for campaign rhetoric. They are great for election speeches, but they do not tell you what to do about the problem.

Third, what we really need to know is how people are getting their needs met. If you want to expand the process of getting needs met, you need to know what people are doing right now, and we know very little about that. I have just mentioned two or three studies that have been done, but that is about it as far as I know. We have hundreds of studies on the unmet needs, but very few studies on how needs are getting met.

Implementation

How can we implement Stormy's idea? You make sure that, whatever you are going to provide to people in terms of free care, you are at least as generous to them if they decide to go over and be part of the private insurance system. If $1,000 is the right number, then we need a $1,000 tax credit, and we need to allow them to apply that same $1,000 to the purchase of private insurance. That way, we are not encouraging people to be uninsured. We need just as much encouragement for private insurance as we have for the public safety net.

We do not need more money. This is a point that seems to be missed in the debate about all of this. We have enough money in our health care system. What we need is for the money to follow the individual.

The flip side of a tax credit is a tax penalty. If I offer you $1,000 to obtain private insurance and you turn me down and decide to remain uninsured, then you do not get the credit, and that means you pay $1,000 more in taxes. That is your financial penalty, if you like, for being uninsured. In fact, right now, people who are uninsured pay more in taxes precisely because they are uninsured than the rest of you who have insurance.

The problem with this system is that those extra taxes, this financial penalty for being uninsured, goes to Washington, but the free care is delivered locally. So we have no connection between what we are doing on the tax side and on the expenditure side. If you want the system to work, it has got to be seamless; the money has to follow the individual. That means that as people go from being uninsured to insured, the $1,000 has to follow them. If they go in any other direction, the $1,000 has to go in the other direction.

Social Safety Net vs. Private Insurance

Imagine for a moment, that all of the people in Dallas who now rely on this social safety net go over and get private insurance. That would mean that we have got to take the $1,000 per person that we are spending over in the public sector and switch it over to the private insurance sector. Yet we can afford to do that because we do not need the public sector any more if no one is in it. In other words, the public sector can expand and contract as the number of people who use it expand and contract.

Conversely, imagine that everyone in Dallas who now has private insurance decides to become uninsured. They give up all those tax subsidies and we use that money to fund the safety net. We do not need to spend more money. We do not need to do much more than just make sure that money follows individuals. There is enough money in the system now, but we need to integrate what we are doing on the tax side and on the expenditure side.

What I am describing is a system in which we have got to fund a safety net. We have got to make sure that if more people do become uninsured, their $1,000 is there--not individually earmarked, but we have got to make sure the money is there. Then, it seems to me, we have satisfied our social obligation.

That would be a system in which we have a form of universal coverage. Government has made its commitment--$1,000 per person, $3,000 per family, or more if you like. Yet the point is, money follows individuals. The public commitment is there. Once we have done that, it seems to me largely irrelevant what the take-up rate is on any of this.

I see nothing wrong with a lot of people wanting to rely on Parkland Hospital and that kind of system. Maybe rationing-by-waiting is the way they prefer to get their care. Maybe they do not want to put any of their hard-earned dollars into more coverage or into a system where they have to wait less. Those are not decisions that have to be made by others. They are not decisions that have to be made by politicians. They are decisions that can be made by individuals.

So, again, the idea is to integrate our spending programs with our tax programs. If we do not do that, we will never solve the problem that Stormy is talking about.

KENNETH E. THORPE: I liked a lot of what I heard today. I like having broad pools. I like the discussion about divorcing the provision of insurance through the employer and focusing on larger population-based pools. I like the FEHBP-type concept of giving a multiple choice. We, unfortunately, called them alliances, but the elements that we are talking about here--with some important nuance differences--are really very similar.

I want to take a step back and look at some stylized facts that I think will place the discussion in some context. I am hoping to also broaden the discussion a bit. We have been focusing almost entirely on insurance and ways of getting insurance to people on the demand side, which is clearly very important, but I also want to focus on some issues on the supply side which I think are of critical import as well that we have not spent as much time on.

The Uninsured

Stan already talked about this, but I think it is important to take a step back and really look at who we are talking about in terms of the uninsured. About two-thirds of them live in families under 200 percent of poverty, so you are talking about a single individual earning under $20,000 per year facing an average premium today of well in excess of $4,000, and families facing premiums of $9,000 or $10,000 a year. Here we are talking about a family of four earning, perhaps, something on the order of the mid-$30,000 range. The typical person works for a company that does not offer insurance; two-thirds of them are in that situation.

I think one of the take-home messages here is that we do have enough money in the system. We are talking about a very substantial--and I think we need to be up-front about it--redistribution of who pays because, frankly, in terms of getting these folks enrolled, it will take (no matter if it is done through tax credits, direct subsidies, or some other mechanism) a lot of public dollars to do it.

We can talk about how to finance it, in terms of not having an existing tax break and so on, but the amount of money we spend on health care is not going to change much from the $1.4 trillion. Yet we are going to have a very substantial reallocation--a redistribution of who pays that dollar--and those are the politics of it. That is when the fun starts about reallocating burdens, and then we get into disagreements about how much to move to the federal ledger, how to finance it, and what other federal priorities we are going to sacrifice in order to accommodate that. That, to me, is 90 percent of the debate right there.

Could we do this? Of course. I think it has been discussed up until the point when the states ran out of money. They were very rapidly expanding coverage to childless adults through CHIP. They're using private health plans; they're using Medicaid. The nice thing about the CHIP program is that we let the states make the decision of moving toward a private market, a Medicaid market, or whatever. Yet they were very actively moving in this direction.

We could have helped them out by making it easier at the federal level, in terms of the waiver process. Yet certainly the states, in many cases, have gone up to 200 percent or 300 percent of poverty level covering parents, covering childless adults through this program. That is where a lot of the opportunities are here.

The issue that we face now is that the states are out of money and are pulling back. I think we need to rethink the fiscal relationship between the states and the feds in this program if we are going to realistically use it as a vehicle for expansion.

Another thing that we need to recognize is the cost side of this, because the trends are only moving more and more in this direction. If you look at where we spend the money, it is dominated by people who spend over $2,000 a year. Eighty-two percent of total private health insurance spending is traced to people who spend $2,000 a year or more. There are very few people in that category, but they dominate the spending.

The flip side is that we have got a lot of people--perhaps 60 percent to 70 percent of the privately insured--who spend 18 percent of spending. They are inexpensive; they are generally healthy. Yet, over time, that is not where the spending problem is.

The Problem of Complex Epidemiology

In fact, if you look at the growth in spending over the last 15 or 20 years, it is dominated by this crowd that spends more than $2,000 per year. It is increasingly dominated by people that have two, three, four, or five multiple chronic illnesses. The growth in spending is dominated, not by a rise in the cost-per-treated-case of an illness, but by a rise in treated prevalence of chronic disease.

You can go down the list: pulmonary disease; asthma (which has gone up by 100 percent over the last 25 years); and mental disorders--in part because physicians are more likely to diagnose it and recognize it. Yet we also have new medical technologies on the pharmaceutical side to treat it, so we have expanded our reach. Additionally, hypertension incidence and prevalence are up.

Go down the line and look at what is driving this system in terms of health care costs: It is these folks that have multiple chronic illnesses. It is these folks, for the most part, on whom any discussion about trade-offs on the margin with consumer-driven products is probably not going to have much of an impact.

This is not to say they could not have a big impact on the other consumers. Yet on the margin they are thinking about--and for people who spend $1,000 or $1,500 or $2,000 a year--those types of products are probably quite important in forcing those types of trade-offs.

I worry that--in terms of really getting our hands around the cost problem--we are missing the boat by not focusing on where the money is and what is driving the growth in the system. There is a set of complex causes that deal with the rise in the underlying epidemiologic prevalence of disease. I did not even mention diabetes and heart disease--as well as more and more complicated patients.

That leads me to the next piece of this puzzle. Insurance is a big part of this discussion, and it has to be. Yet if we do not focus on the supply side, the delivery system side, of this at the same time, we are going to miss a big opportunity.

No matter how we put money into the system, throwing money into the existing system in terms of how we deliver services would waste a lot of money. It is not going to provide the type of care to the patients that are really driving health care in the system. It is throwing money into a delivery system that, in many parts of this country, is dysfunctional and broke. We need to think about good models of how we are going to fix this and fund them.

System Fragmentation

The delivery system is highly fragmented because it responds to the way we pay for services on a piecemeal basis. It arose out of a system that has the physician or provider waiting for the patient to show up when he or she is sick and then getting paid for actually doing something to the patient.

So the fragmentation in the system is purely a function of how we pay for services. It is a terribly poor provider of care to chronically ill patients. Much of that care is effectively delivered outside of the physician's office and outside of the hospital. A lot of it is focused on care at home. A lot of it has to be focused on lifestyle issues with respect to smoking and weight and diet.

We have some good demonstrations looking at treating patients with congestive heart failure and really focusing on making sure that CHF patients are following their diets and keeping their water levels at a point at which they do not get fluid buildup in the lungs and end up in the emergency room. You can do that, but nobody gets paid to go and take care of that patient at home or set up the system to monitor it or provide the incentives to prevent it.

If you look at other parts of the delivery system, not only is the care part of this poorly designed and highly fragmented, but one of the reasons why the growth in spending is also rising is that it is incredibly inefficient in terms of the productivity of health care.

If you look at the 18 major service industries, health care is number 14 in terms of productivity growth over the last two decades. Why? It is largely because it has a 1950s version of an information technology platform.

If you look at any other service industry in this country, they have an information platform and a technology base that far exceeds the capacity that we have here. It is more important in health care because we have a highly fragmented multi-payer system, and not having that type of an electronic platform to process claims and do the transactions adds to the cost of the system. So another part of this puzzle is very low productivity traced to an outdated information technology structure.

Systems Building

I think we can build these systems. It is not a real mystery, clinically, how to take care of a patient with congestive heart failure or how to provide the best care for the diabetic patient. The trick is that you now have patients that have multiple chronic diseases seeing multiple physicians and taking multiple meds.

The issue is building a delivery system that is integrated, seamless, and has an appropriate clinical and information technology platform to provide care to the patients that are driving spending in the system. If you look at the demographics, obviously, we are talking about (in the next 15 to 20 years) more and more folks falling into the chronic side of this equation, and I see more and more folks driving that 82 percent figure even higher.

Who finances the care and what kind of insurance we give people is an important piece of the puzzle, but simply providing health insurance coverage without fixing this side of the system is not going to be nearly as effective as where we need to go in terms of providing quality health care, and it is certainly not going to address the issue of what forces are driving health care spending in this country.