Paul Ryan's Roadmap Proposal

So, if you have been thinking that the Democrats overreached and tried to 'take on too much with health reform' you should stop reading now and write off Paul Ryan's (R-WI) Roadmap for America's Future Act of 2010 [hereafter Roadmap]. This is a overarching vision for taxes, Social Security, private health insurance, Medicare and Medicaid, all rolled into one. This would be a big, big bang.

From my perspective as a 42 year old with 3 children, who thinks we must cover everyone while being fiscally sane, I think our country needs a fairly fundamental reconsideration of what people can expect to be provided by government in the way of retirement and health, and what they will be responsible for own their own. So, I am open to big bangs, but that is typically thought of as an ailment or virtue that only affects liberals. Rep. Ryan paints his vision, and shows that desire for big bang is bi-partisan, at least for this Republican.

On parts of Rep. Ryan's vision, I think it is about right (Social Security). On parts of it, this is not how I would do it and I am not convinced it will work(Medicare), and on others (private health insurance) I am not only in between, but see alot of similarity between the Senate bill and what he proposes, even when he professes to not see it. [Interesting interview by Ezra Klein with Rep. Ryan on all this from yesterday].

The big picture effect of the Roadmap becoming law as compared to the what CBO calls the Alternative Fiscal Scenario (which is a mix of current law and certain changes they assume would have to be made) would result in the following differences in revenues and spending on Social Security, Medicare, Medicaid (all shown as percent of GDP, extracted from Table 1 in the CBO document). Here is the CBO analysis of the Roadmap, which it should be noted doesn't appear to contain coverage estimates....meaning how many people insured in 2020, etc. Further, it isn't a score of the bill or estimate of its cost over the next 10 years, but a 75 projection of the approach. Projections this far ahead are subject to a great deal of uncertainty, but they do provide a sense of what the proposal would entail.

% Gross Domestic Product
2020 2040 2060 2080
Revenue-Ryan 18.6 19.0 19.0 19.0
Revenue-CBO 18.6 19.4 20.6 21.9

Spending
Soc Security-Ryan 5.4 6.3 6.0 5.1
Soc Security-CBO 5.3 5.9 5.8 6.1

Medicare-Ryan 3.7 5.1 3.8 3.5
Medicare-CBO 4.2 8.1 10.9 14.3

Medicaid-Ryan 1.6 1.3 1.1 1.0
Medicaid-CBO 2.1 3.0 3.4 3.7

Tax credit-Ryan 0.6 0.6 0.5 0.4
Tax credit-CBO* 0.0 0.0 0.0 0.0
*This is new spending, and the amounts above are net of the difference between the refundable tax credit and the loss of the tax exclusion of employer paid insurance. So, in 2020, the tax credit results in $600 Billion more for private health insurance via tax credits than the current tax exclusion would provide in subsidy.

Big picture. Spending on Social Security would actually go up slightly under the roadmap from 2020 to 2060, then drop by 2080, while spending on Medicare and Medicaid would drop substantially from 2020 to 2080. Net spending by the federal government to subsidize the purchase of private health insurance would go up slightly, but the distribution of this expenditure would be equalized across the population with everyone getting the same amount.

Immediate Changes. The tax exclusion of employer paid insurance would be repealed and individuals would get refundable tax credits to purchase health insurance beginning in 2011, with each adult getting $2,300, children $1,700 up to a family maximum of $5,700. The tax credit amount would be indexed at the average between the Consumer Price Index all item Urban (CPI-U) and the Consumer Price Index Medical (CPI-M). This means the value of the tax credit would grow less slowly than medical inflation has grown recently. CBO estimates that the annual increase in the value of the vouchers will be 2.7% under the roadmap, as compared to a projected 5% per capita growth in health care spending under the CBO's alternative scenario. That is a big difference. I will comment more fully on the health exchange issues below, but the bottom line is that the bill would fix the federal government's contribution to private insurance over time and determine how fast it grows. This is like a governor on a bus accelerator. A similar approach is taken for Medicare, with vouchers indexed in the same manner eventually replacing Medicare as we know it. It is basically defined contribution instead of defined benefit.

Social Security. Social Security would be unchanged for persons who are 55 or older in 2010. For those 54 and younger this year, several changes will take place. First, a new and increased minimum benefit will be established for low wage workers, roughly persons who worked at or near minimum wage for more than 20 years. Second, private accounts will be available whereby workers can allocate some of their pay roll taxes into private accounts. The mix of investments will be regulated in some way. As people with private accounts begin to retire, the federal government will guarantee a return of equal to inflation (Consumer price index all item-urban?-not clear to me what definition of inflation is for this provision; very important). So, if you get a better return, you get to keep it. If you lose money, the government will guarantee you at least inflation-level returns. Gains above inflation reduce the flows from government to retirees.

My bottom line: This is privatization with a net, and with an expanded minimum benefit for very low income workers. This sounds about right to me.

Private insurance. This part of the Roadmap is just the Patients' Choice Act with a bit more detail on the actuarial aspects of the plan (notably, how the value of the credit will inflate over time) but there are MANY unanswered questions about the exchanges and how they will work. For example, the Patients' Choice Act as submitted May 20, 2009 has the following sentence about exchanges. (Exchanges)...."shall develop mechanisms to protect enrollees from the imposition of excessive premiums, reduce adverse selection, and share risk." One of the reasons the Senate (and the House) bills were so long is that they provided inordinate and necessary detail about these sorts of things. This sentence and others like it will have to be fleshed out in order to fully evaluate this aspect of the Roadmap. And you can't do it in 10 pages, by the way.

Big questions I have:
*will states continue to regulate insurance?
*will the Roadmap specify benefit packages? If no, then what is their a strategy to help consumers shop/compare policies in a reasonable way?
*will private insurers still be able to offer policies outside of exchanges that employ underwriting?
*will pre-existing condition bans and other insurance market reforms apply only within exchanges? If yes, why won't the exchanges end up in death spiral?
*What about affordability? The mean annual insurance premium for a family in the U.S. is about $13,500, alot less than $5,700.
*Is there any sort of a soft mandate? Auto -enroll, etc....while I get politically individual mandates are not liked by Republicans, that sentence above about adverse selection is most easily fixed via an individual mandate with teeth. I don't see how you make it work without some sort of a mandate.

I favor ending the tax exclusion of employer paid insurance, so am willing to listen to anyone who wants to do that. However, alot more detail is needed to understand how this will work.

I understand the general idea here to be a cousin of the Senate plan. However, I think what Roadmap is really proposing is a catastrophic level of coverage to everyone, and a new cultural statement that people are going to be responsible for the rest with after tax dollars. That message does not come through in how the Roadmap is described. With the questions above addressed and with premium support for underneath cover for low income folks, this would begin to look alot like the Senate bill with only catastrophic cover being guaranteed, with people buying other cover if they wish with after tax dollars. The Senate bill says everyone will be covered with comprehensive benefits and a much lower level of out of pocket cost once insured. Those are more differences in degree than in approach. So, I see this as similar to the Senate bill if you answer the questions above with the regulations that I see as necessary to make this work.

My bottom line: More detail needed, but this is an approach with which a compromise should be possible with what is proposed in the Senate bill. The big difference is whether goal is guarantee comprehensive coverage to everyone, or to guarnatee catastrophic coverage to everyone. I can live with the latter if everyone is in. I don't see how you do it without a mandate of some sort...

Medicare. I understand Medicare to be the most practical example (along with Social Security) of our culture making the moral precept 'Honor Thy Father and Mother' a reality (and grandfather and grandmother). I am much more willing to try more individual based/market notions with younger persons than with the elderly. As I say that, I am thinking of my 85 year old grandmother, but of course she will not be affected by this proposal. And there is a 10 year 'warning' of changes coming to Medicare, on the backs of individual's being much more involved in purchasing their own insurance via the tax credit/exchange regime noted above for the decade prior to Medicare eligibility age (which also increases over time which makes good sense). But, I am skeptical of this part of the proposal.

The proposal is as follows: For those age 65 and older in 2020 (55 and older in 2010) you will not experience a difference in Medicare. It will be fee for service Medicare. Apparently Medicare Advantage will also continue, with I think reduced premium rates, but that which is often a hot topic is small potatoes here. For those turning 65 in 2021 (54 this year), you will not have the option of being in fee for service Medicare. Instead, you will get a voucher with which to purchase private health insurance. 65 year olds would get $5,900 (in 2010 dollars). The in 2010 dollars means that the voucher amount is set at $5,900 this year and then is inflate in by the average of the CPI-U and CPI-M, as with private insurance tax credits, but the vouchers would not be used until 2021. The amount of the voucher that one would recieve would vary based on their age and health. The average voucher amount in 2021 based on projected age/health breakdowns per CBO of 65 year olds then would be $11,000 (in 2010 dollars). The CBO notes that Rep. Ryan's staff provided extra info to allow these projections to be done, so they will follow up with more written detail soon, I assume.

The voucher would be means tested. Those in top 2% of income would get 30% of the voucher; those in next 6% (92%-97.9%ile of income) would get 50% of voucher amount. [I think the 90th percentile income is about $120,000 last year...but I need to check that.]. Everyone else would get 100% of the voucher amount.

The federal government would then provide funding for a Medical Savings Account (MSA) for low income elders (not clear how low income defined) Attachment B of the CBO report on Medicare (p. 27-28) is important here. It says "The amount of the Medicare voucher would equal average per capita Medicare expenditures in 2010 under current law." Is this average as simple arithmetic mean, or average as defined by 'average adjusted per capita cost' which varies by county and is used to set premium levels for current Medicare Advantage plans? If this voucher is a straight mean, then doctors practicing in the Miami area (which has very high Medicare expenditures per capita, nearly 3 times as high as some areas) will remember today as the Halcyon days in terms of how much Medicare pays them!

If it is average meaning based on average adjusted per capita cost, then of course it serves to lock in geographical variations that do not appear to be linked to quality. The unexplained geographical variation in Medicare spending is a big problem. I am just not clear which of these they are using.

In any event, moving to a voucher based system in which elderly persons get a defined contribution will result in Seniors having less generous benefit packages and paying more in extra premiums for similar benefits. It is unclear to me whether there is a ban on pre-existing conditions for elderly purchase? Perhaps in the exchange but not outside? The CBO report says on page 28 "To the extent that risk adjustment did not adequately reflect expected costs, however, less healthy people would either have to pay an additional premium for the same coverage or accept some limits on their coverage, and healthier individuals would be able to buy more extensive coverage or pocket the difference between their premium and their voucher."

Now obviously the status quo planned level of Medicare expenditures is fscal insanity, and the only way to reduce spending is for people to systematically get less care (price x quantity) than they would get under the scenario of no change. We are talking about rationing care, with not rationing meaning everyone gets whatever they want. We have rationing now, and I give Rep. Ryan credit for honestly noting this in his interview with Ezra Klein yesterday. He says, "Rationing happens today. The question is who will do it. The government. Or you and your doctors."

I think he overstates the difference in the choice of how to ration.

It is a question of how do we ration, or apply limited resources in the best way. I prefer an approach that is based on systematically asking the questions: (1) do the benefits of the care outweigh the costs? If no, we shouldn't pay for it via Medicare. and (2) Is the care worth it? Meaning, at some point, there is a marginal benefit that is not worth the extra cost.

The qeustions I suggest are hard questions to answer. They are laiden with value judements. It will be hard. Mistakes will be made. I would prefer a system that relied on an Independent Medicare Advisory Commission applying cost effectiveness research and having guidelines that mixed guidance of what is covered and how much is paid for care with patient choice and a bigger role for docs who were less incentivized to do more by default, in conjunction with their patients. Government has a role in interpreting the research evidence, and keeping incentives straight. But it is not a choice of gov't v. doctors deciding. There are many steps to what I write above, but I think we can work out the over spending and reduce the cost of Medicare in this way.

Rep. Ryan's approach fixes the cost to the federal government and then assumes that the market will work it out. It says we will fix an amount the government will pay for your health care in old age, but you are responsible for the rest. In the end, society will have to decide whether they would rather muddle through on changing what is covered, when and how it is paid for as I prefer, or whether we would rather fix the public costs and assume it will work out via the market. In fact, you have now created a payer (private insurance) with a clear incentive to provide less.

But, we need to get straight the market for what. Insurance? or Care? They aren't the same thing, but they are certainly related. Meaning markets work when consumers understand how much they want something and sellers know what they are willing to take to give something up. I actually don't think that people can do a very good job of shopping for health care. If I go to the doctor and am in pain and scared, how am I going to inquire about price? Thus, I would prefer more expert driven guidelines with docs helping patients decide after some of the incentives now in the system have been changed. I think people can do a better job of shopping for insurance, so long as you have some basic things set such as benefit packages so they know what they are buying.

It is true that you first start with only 65 year olds in 2021. In 2022 there are 65 and 66 year olds, and so on. So, you are easing into this as opposed to trying to do it with a broad cross section of the elderly at once. Still, I have the following BIG questions about this approach to transitioning Medicare to a voucher based private insurance system.

1. Why would any insurance company want to be involved in this market? Everyone dies. Before you die you are tpically sick, and costs rise tremendously before death. The older you are, the closer you are to death (you heard it here first). Again, why in the world would a private insurance company go into this business unless you could figure out how to cherry pick. But, if so, the non cherry pickers will blow up and exit. I know there is some premium at which this will work, but the voucher suggested is far below the amount of money planned to flow into the system to care for the elderly. That is the point of all this, I just don't see how private insurance companies are going to want to jump into this market.

All of the experience we have seen with Medicare Advantage/HMOs has been one in which the healthies sign up and the sickest stay in FFS Medicare. If insurance companies had to insure a certain proportion of the elderly, with some sort of rebalancing of risk to account for adverse selection that occurred, then maybe.

2. The above will be exacerbated if you keep a system with exchange based policies with no pre-existing conditions, but outside the exchanges they can underwrite. With elderly individuals, I would think you just sell outside the exchange and aggressively underwrite. Who ends up with the sick folks....remembering that eventually almost all of us are sick.

3. Is the average premium a straight national average, or a county based average like average adjusted per capita cost used to set Medicare advantage premiums?

My bottom line: the reduction in Medicare cost as percent of GDP that the CBO says the Roadmap would generate are tremendous. Medicare is expected to be 8.1% of GDP in 2040 under CBO alternative scenario, and 5.3% under the Roadmap. I appreciate the fact that there is extra resources for low income persons, but I don't see how practically this market will work without some form of forced risk pooling. I am skeptical and need to hear more. But, this is a serious proposal and I am willing to listen.

Medicaid. The Medicaid changes called for an analysis that CBO was unable to perform....meaning there are so many changes to a complicated subsystem--particularly the long term care parts-- that it is hard to predict exactly what will happen. The letter from CBO to Rep. Ryan says they agreed with his staff to simply alter the rate of Medicaid growth, with most persons in Medicaid being shifted into private insurance using the tax credit plus a Medicaid-paid subsidy (premium support) to purchase private insurance. This essentially does with the Medicaid program what the Senate bill envisions with persons just above Medicaid eligibility (133% poverty + in Senate bill).

I would be open to this goal of moving income-eligible persons into private insurance, from the standpoint of removing the stigma of Medicaid and removing a payer that is not beloved by doctors and hospitals (again, we will see what they think about the old days if this actually became law because every dollar saved is a dollar of somoene's income dropping). These would mostly be children and pregnant women. Others will say they are more vulnerable, etc. so the marketplace won't be so good for them, but I think it is worth the risk to try and mainstream them into the system, so to speak.

Persons who were disabled and/or receiving long term care services would remain a part of Medicaid, with the federal government block granting Medicaid payments to states. This is actually a big deal, as around 6 in 10 Medicaid beneficiaries are children or pregnant women, but they account for less than 30% of Medicaid spending. So, the real big money in Medicaid is for the disabled and long term care, and the Roadmap has the federal government fixing their costs for this portion of the costs, block granting and thereby shifting the rest of the responsibility to states. The most obvious group to be opposed to this will likely be bi-partisan upset from Governors. I am unclear whether the block granting is population based only, or whether it also accounts for the relative poverty level of the state, as does the basic Medicaid funding formula?

I think there is lots of evidence that long term care needs to be greatly reformed, and particularly Medicaid programs are notorious for insitutionalization bias, meaning they pay for the most expensive setting after spend down (nursing homes), but lag behind in less expensive and likely preferable community arrangements and settings (that are assumed to have more moral harzard or wood work effect). So, a change in how we fund long term care is needed. Would block granting force a change? Probably so. But, it bothers me to be making changes that affect the most vulnerable members of society in such a blunt manner.

As Rep. Ryan's roadmap could be viewed as a clear statement of what government will provide and what you will be responsible for, you could imagine the CLASS provisions in the Senate bill fitting into his framework. Here is what I wrote about CLASS in December. And this recent paper I have with colleagues in Health Affairs on the use of genetic markers to underwrite private Long Term Care Insurance lists reasons why people don't purchase private insurance, and discusses issues related generally to trying to increase rates of purchase. Adding a self-financing provision such as CLASS could stimulte the development of private LTC policies that would be more likely to be purchased, which could potentially reduce the future costs to Medicaid. Any riskiness of the CLASS provisions in terms of not being self financing pale in comparison to transitioning Social Security to private accounts with a federally guaranteed rate of return equal to inflation.

My bottom line: I could live with trying to shift Medicaid beneficiaries into private insurance, but have concerns about block granting Medicaid for the most vulnerable members of our society. Especially since the Roadmap is an approach that is trying to spur more individual role in addressing health insurance needs, the CLASS provisions in the Democratic bills would seem to be something Rep. Ryan would support.

In the end, the Roadmap clearly defines a vision for what the federal government will provide in terms of retirement and health care, and makes the individual responsible for the rest. We need such a clarifying societal agreement going forward. I like some aspects of this proposal (Social Security), think others have some good ideas but with *many* questions and details that need to be answered (private health insurance), and am worried that the proposed vision will not work well for other parts of the system (Medicare, Medicaid). However, this is a serious and wide ranging proposal that at least says how one guy would do it.

Update: What others are saying about Rep. Ryan's Roadmap. Here, here, here, here, here, and here. Even more and more