Tuesday, April 28, 2015

How Vermont could have made single payer work, part 2: Green Frankenstein Care


Image: Frankenstein was grafted together of different parts of different cadavers and reanimated into a sentient being. Since single payer is apparently a political nonstarter in the United States, let's do the social policy equivalent and find a way to stitch together the component pieces of single payer.




This is a follow-up to an earlier post written in response to Vermont's decision to abandon their efforts to create a state-wide single payer health care system, which would have been called Green Mountain Care. I argued that that single payer in the United States isn't dead, though a pure single payer health care system might not be possible in the unique political and legal environment of the United States. That's not a problem, however, because single payer advocates don't care about single payer per se; they are only interested in attaining the universal coverage and mammoth administrative savings offered by a single payer system. Fortunately, there are ways to reach these goals without single payer.

As Vermont's single payer advocates unfortunately found out, any health care reform proposal will end in failure if it doesn't camouflage its cost as well as the current system. Americans generally don't have a good conception of how atrociously expensive the American health care system is--because the cost is well concealed. Since the American system splits the costs of health care so many ways, it's not immediately obvious how much health care truly costs. For example, employer-sponsored insurance premiums are paid for in part by generous federal subsidies most Americans don't even realize exist, and by employer and employee contributions. Further masking cost, private insurance doesn't cover the elderly or long term care. And, private insurers deliberately design cost sharing (deductibles, copays, and coinsurance) to be be difficult to understand and therefore obscure the true out-of-pocket cost of health care.

Due to it's simplicity, Green Mountain Care couldn't provide this level of cost camouflage. The entire program was financed by simple, direct payroll and income taxes. Because the cost of the premiums weren't split up several ways and cost sharing was simple and transparent, the extraordinary cost of health care in America became unbearably obvious.

Thus, Vermont abandoned Green Mountain Care because it was too expensive--even though it was less expensive than the system it was supposed to replace. The irrationality of this decision is impossible to overstate.

Clearly, any attempt to achieve universal health insurance coverage will fail if it doesn't conceal its cost at least as well as the current system.

I suggested that single payer advocates could look for inspiration in the German and Japanese health care systems because--as in the United States--health insurance is heavily linked to employment. Yet with a few straightforward regulations, the German and Japanese systems attain universal coverage and all of the administrative savings of single payer. As I outlined in the previous post, these regulations are:
  • Government regulations set prices of all types of office visits and procedures
  • Only a single insurance plan can be sold
  • All insurers must be not for profit
  • Redistribution of medical loss imbalances to spread risk across insurers
  • Employer mandate to provide all employees with health insurance (with employee contribution)
  • Health insurance companies must continue to cover enrollees who lose their job or stop working for any reason 
Together, these regulations would allow our employer-based system to attain the cost savings and universal coverage of single payer.

It's worth noting that a single payer system isn't automatically better than other systems. Norway, for example, has a single payer health care system, yet per capita health care spending is far higher than Germany and Japan. Clearly, single payer is a great model for health care reform, but it isn't the only model, nor is it necessarily the best model, either.

Of course, this last post was fairly abstract, so here I intend to ground it by explaining how these reforms could be implemented--step by step--at a state (or possibly city/county) level.

Green Mountain Care would have been a pure single payer system. It would have been elegant and simple. This new system would work equally well, but only after boorishly stitching together seemingly unrelated reforms into a functioning--if ungraceful and confusing--whole. Such a system does not deserve the elegant name of Green Mountain Care. Instead, meet Green Frankenstein Care.


Step 1:
Pass a law that appoints a group to set the prices for every type of office visit and procedure (for simplicity, we'll call this the Green Frankenstein Board). All payers and health care providers are legally bound to these prices. This should be done fairly and transparently. But it should also be done as simply as possible. Likely, the simplest way would be to index all prices to Medicare (ie, every insurer must reimburse double the Medicare reimbursement for every type of office visit and procedure).

We need a group of experts to set these prices because we don't want to accidentally put health care providers out of business. Ewe Reinhardt famously commented that there is a billing specialist--but not nurse--for every single bed of his hospital. This will not change overnight. As the system adjusts to the Green Frankenstein Care, administrative costs will fall and prices can be lowered. But special care needs to be taken to ensure this transition occurs sufficiently slowly that medical providers can adjust to it. Once administrative costs start falling, the Green Frankenstein Board can lower reimbursement rates to bend the cost curve.

This reform is unquestionably Constitutional, since it is the law of the land in Maryland and was actually once the law in 11 other states. This regulation has never been challenged as unconstitutional; furthermore, it has proven itself extremely effective wherever implemented (in addition to Germany and Japan, this reform is used in France, the Netherlands, and Switzerland). Obviously, this reform alone is insufficient to get all the benefits of single payer, so let's continue.

Step 1A (optional):
Insert reform of choice. Maybe you think primary care needs to be reimbursed higher relative to speciality care, PA's or NP's need to be reimbursed higher, or that end of life counselling should be reimbursable. Or maybe you want to change the way chronic disease is treated. There are any number of reforms that could be inserted here.

Before moving on to step 2, it's important to remember that our primary hurdle is ERISA, which the courts have interpreted so broadly that it has become the grim reaper of state and local health care reform efforts.

Even leaving aside ERISA, only two things these days are unquestionably Constitutional: taxation and tax cuts (by the Supreme Court's logic, Obamacare's individual mandate is only Constitutional because it acts functionally like a tax). So let's do both...several times.

Step 2:
Institute a large tax on insurers. It doesn't matter where, as long as it's a hefty tax. There could be a tax on premium payments, a tax on billing transactions with medical providers, a tax on every dollar of coinsurance, deductible, or copay enrollees pay--there are any number of taxes that could be used here. The important thing is that insurers are taxed directly because we want the tax to be passed along to consumers as a premium or cost-sharing hike.

Step 3:
Exempt insurers from this tax, but only if they agree to:
  • Sell a single plan that is designed by the independent Green Frankenstein Board we created in step 1. For simplicity, let's call this health insurance plan Green Frankenstein Care.
  • Sell Green Frankenstein Care for a single price that is determined by the Green Frankenstein Board
  • Become not for profit
  • Submit to financial audits by the Green Frankenstein Board
  • Participate in redistribution of medical loss
  • Continue cover all enrollees even if they no longer work for the employer they originally received health insurance through, regardless of their ability to pay (if there is a Medicaid waiver from HHS, then insurers should be forced to accept everyone who applies no matter what)
This is a pretty complicated step, so let's take it in parts.

Single payer is administratively simple because there is a single payer for providers to interface with. When the Green Frankenstein Board set the price of all procedures and office visits, we recreated this aspect of single payer in our multiple payer system. But single payer derives further administrative efficiency by having a single health insurance plan--unlike private insurers, which may have dozens apiece. We want to recreate this aspect in Green Frankenstein Care. Obviously, the simplest way to for there to be a single insurance plan is to have a single payer. But since that's off the table, we'll do the next best thing and simply require all insurers to sell a single plan--which we're calling Green Frankenstein Care. In one simple regulation, we've recreated one of the most important aspects of single payer--a single plan--in our multiple payer system.

To make things even simpler, the Green Frankenstein Board should design Green Frankenstein Care to be easy to administer. For example, private insurance often has very complicated cost-sharing--ie, $30 copay for primary care and a $2000 deductible followed by 30% coinsurance for specialist and emergency care. Cost sharing should be designed to be simple: all primary care visits have $20 copay, all specialist procedures have 10% coinsurance, etc. (Obviously, anything related to chronic disease care should be without cost sharing immediately.)

Obviously, plans without cost-sharing have far simpler administration; furthermore, cost sharing is wrong and should be eliminated on principle. Unfortunately, eliminating cost sharing would result in higher premiums, and that's a political non-starter as we learned in Vermont (America's obsession with the "moral hazard" also makes eliminating cost sharing a political DOA).

The last three regulations are necessary to prevent insurance companies from trying to enroll only healthy enrollees.

In insurance, companies are only profitable if they get lucky (or cheat) and enroll more healthy individuals than their competitors. Insurance companies lose money when they get unlucky and enroll more sick individuals than their competitors. Companies with healthy individuals have less money going out to pay for medical services (the medical loss) than companies with sick individuals.

Obviously, this creates an incentive for insurers to try to enroll only healthy people and boot sick people, with predictably appalling practices like recissions and preexisting condition exclusions. Obamacare made these practices illegal, but it's such a toxic incentive that we should go beyond the individual practices and target the perverse incentive itself.

To do so, we have to remove the financial incentive to cheat by redistributing imbalances in medical loss. If company A runs a profit and company B runs a loss, it's because company A got lucky and had healthier enrollees than company B. Company A didn't do anything to deserve their profits, so their profits should be given to Company B so that Company A and Company B just break even. (This is part of the reason why there must be a single insurance plan sold for a single price; if insurance companies are free to set their own premiums, they might lower prices, knowing that other companies will have to foot the bill for any losses they incur.)

As should be obvious by now, the Green Frankenstein Board needs to be stacked with actuaries. They need to look at the cost of insuring the entire population, and set prices accordingly. This is why every company must sell the same plan at the same price. Since our actuaries will calculate the cost of insuring the entire population and build it into the cost of premiums and cost sharing, it must set the prices of premiums and cost sharing as well or the entire system could wind up owing more to providers than it takes in from premiums and cost sharing.

If the Green Frankenstein Board does their job correctly, we would expect about half of insurers to run a profit, and about half should run a loss, with the profits redistributed so everyone about breaks even.

As I've imagined it, participating in the redistribution of medical loss is voluntary, in that companies choose to participate so that they can be exempted from a tax hike and benefit from knowing they will never operate at a loss. But this is actually a part of Obamacare, and it's not optional. For health care reformers who decide to go in another direction, it's worth noting that this piece of Obamacare--forcing profitable insurers to redistribute their profits to unprofitable insurers--has never been challenged as unconstitutional.

Finally, this system only works if the government guarantees that no insurance company that plays by the rules ever operates at a loss. And that's why the government needs to have robust power to audit insurers. Since they're guaranteeing the solvency of these companies, they have to be able to ensure these companies are doing their job correctly.

Step 3A (optional):
This is a great place to insert other reforms. We could stipulate that these plans can only reimburse medical providers that salary their physicians, change the way chronic disease is managed, force providers not to pick and choose their patients, for providers to use an integrated state-wide electronic medical record, etc.

Step 4:
Apply for Medicaid and Obamacare waiver from HHS.

Green Mountain Care would have covered everyone in the state of Vermont. In this way, Green Mountain Care would have made Medicaid and Obamacare unnecessary--who needs Medicaid or Obamacare if you've got Green Mountain Care? So, the federal government agreed to pay Vermont directly the money they otherwise would have spent on Obamacare subsidies and Medicaid.

Were Vermont to institute Green Frankenstein Care, they would still apply for these Medicaid and Obamacare waivers from the federal government (see *footnote for how this could work at a city or county level). As I've imagined it, Green Frankenstein Care would eventually result in universal coverage, as it does in Germany and Japan. Thus, waivers for Medicaid and Obamacare are clearly justified.

But even if these waivers can't be obtained, that's really not that big of a problem. For Medicaid, we might stipulate that someone who loses their Medicaid must be accepted back to their previous Green Frankenstein Care insurer regardless of their ability to pay. Health care reform without a Medicaid waiver might be more complicated, but for health care administrators to sort out just two plans (Medicaid and Green Frankenstein Care) is a massive improvement over the current system.

And an Obamacare waiver would be great, but missing out on this waiver only complicates Green Frankenstein Care a little. Remember, our regulations from steps 2 and 3 ensure that only one type of plan can be practically sold (because any health insurance plan other than Green Frankenstein Care is subject to a massive tax); thus, anyone who utilizes an Obamacare subsidy will have only one plan to purchase, and the federal subsidies all will go to a non-profit Green Frankenstein Care insurer anyway. Once per year, everyone would have to reapply for their Obamacare subsidy. That's not ideal, but it's not terribly onerous, either.

Logistically, it wouldn't be difficult for a city or county to ensure that only Green Frankenstein Care is sold over the Obamacare exchanges. Legally, Obamacare premium subsidies only cover the cost of the second lowest cost silver plan, but because we're charging a large tax on insurers that don't do our bidding, they would be easily priced out of the Obamacare exchanges. As long as there are at least two Green Frankenstein Care insurers on the Obamacare exchanges, the second lowest cost silver plan will always be Green Frankenstein Care (because step 3 requires that all Green Frankenstein Care plans be sold at the same price). No one would buy any insurance plan that wasn't Green Frankenstein Care because the out of pocket costs would be far higher.

A sliver plan is one where the enrollee can expect that premiums will cover 70% of their medical costs and the enrollee will pay the remaining 30% out of pocket, via some kind of cost sharing (copays, deductibles, or coinsurance). However, in addition to heavily subsidizing premiums, Obamacare also heavily subsidizes cost sharing. While this unfortunately adds to the administrative complexity of the system, it is nevertheless a vast improvement.

In the end, Frankenstein is practical at a city or county level, though it would be slightly more complicated without an Obamacare or Medicaid waiver.

Step 5:
Leave Medicare alone

Vermont had planned on integrating Medicare into their single payer system. I see no reason to mess with Medicare. It's already a single payer system. Ideally, we would have a single health insurance plan for all, but because of the constraints imposed by federal policy, we'll have to have two (Green Frankenstein Care and Medicare). Trying to integrate Medicare is more trouble than it's worth. Remember, single payer advocates don't care about single payer per se; they only care about attaining the universal coverage and administrative cost savings of single payer. If we can attain universal coverage and 98% of administrative savings, that's certainly better than insisting on a pure single payer system and losing every political fight.

Step 6:
Institute a payroll tax--paid only by the employer, not the employee--that costs more than the cost of our insurance plan. In Vermont, this was estimated to be 11.5%.

Step 7:
Exempt employers from this payroll tax if they offer all their employees Green Frankenstein Care. Remember, any insurer can choose to sell Green Frankenstein Care, regardless of whether or not they actually meet the requirements to receive the tax cut of step 3.

At this point, we've maintained the employer-based system, but we've also created a de facto employer mandate. Of course, because we're using a mix of tax increases and tax cuts, we're not forcing anyone to do anything--employers are choosing to provide their employees with insurance to benefit from the tax cut. Any taxes actually collected will be used to finance Green Frankenstein Care; hence, any employer doesn't offer their employees health insurance still pays their fair share. This tax could be indexed to company size to protect small businesses (Obamacare uses a simple cutoff of 50 employees, but there should be a sliding scale; cutoffs are terrible policy).

Again, this is business as usual for employers that already offer insurance to their employees--they'll just have to switch to Green Frankenstein Care--and this is a major strength of Green Frankenstein Care. Employers still benefit from the federal tax break for employer-sponsored health insurance, and they can still split the cost of the premium between employer and employee however they see fit. To repeat: there are many aspects of the American health care system that I think are terrible policy, like cost sharing and employee contributions to health insurance premiums. Unfortunately--as the failure of Green Mountain Care taught us--any successful plan for health care reform must conceal its true cost at least as effectively as the current system. Maybe these things can be eliminated in the future, but not now. Our political system is too irrational for that.

Step 9:
Raise a large tax on health care providers.

Step 10:
Exempt providers from that tax if they agree to open their books to government audits. We need to make sure that cost savings from administrative simplicity benefit everybody, not just providers. As providers save money from the greater administrative simplicity, this should result in a lowering of prices. But the Green Frankenstein Board can't know to lower prices if they don't know what savings each health care provider is actually attaining.

Conclusion
This series of oddball reforms are very disjointed, so let's conclude by trying to grasp the big picture.

In Green Frankenstein Care, health care providers operate as though they are in a single payer system. Every patient has the same insurance plan, and the government will set all prices in order to slow--and then reverse--price inflation.

Though there are multiple payers, because insurers can only sell a single plan at the same premium price--and because we redistribute any imbalance in medical loss--they act as though a single payer. This allows the system to harness all of the actuarial strength and nearly all of the administrative simplicity of single payer.

Typically, single payer systems are financed through payroll taxes paid both by employer and employee. Functionally, we have created the exact same thing: employer and employee contributions to insurance premiums get paid into an insurance system that functions like a single payer system. And, we have created a de-facto employer mandate to offer insurance, meaning we no longer have employers that can choose not to pay--in essence, a payroll tax.

Finally, other reforms--including making insurers not for profit--we have removed the profit motive from health insurance.

For most people, there will be very little change. They will still get employer-sponsored health insurance, albeit with the security of knowing they will never lose that insurance. And businesses that were already providing health insurance to their employees will notice very little change as well.

The one major advantage of single payer is automatic enrollment. As I've imagined Green Frankenstein Care, you can never lose your coverage if you lose the status that made you eligible for it in the first place (ie, you got it through your employer, you were the dependent of an enrollee); if there was a Medicaid waiver in place, then insurers can never turn down anyone's application, period. But--because enrollment is not automatic as in single payer--it will certainly take longer to achieve universal coverage. Unfortunately--as Green Mountain Care proves--a perfect system will never be politically palatable, not even in deep blue Vermont.

And that's how you get all the advantages of single payer while maintaining the cost camouflage of the American employer-based system. Green Frankenstein Care is a functional single payer system that has the key advantage of being politically palatable.

Most importantly, we have slain ERISA. By making everything voluntary--insurers and providers chose to obey our reforms in order to benefit from tax cuts--ERISA becomes a non issue. Again, this is the exact logic by which the Supreme Court upheld Obamcare's individual mandate.

I've ignored a few details. For example, I didn't explore premium payment structure for people who don't get Green Frankenstein Care from their employer, cost sharing structure, or address minimum employer contributions to employer-sponsored Green Frankenstein Care. Nevertheless, these problems can be simply worked out, and the structure of the system ensures a functioning single payer imposter that fits into the unique legal and political environment of the United States.

Finally, let's not forget that Germany, Japan, and a couple other countries have universal health care systems very similar to what I've envisioned here. This is a model that really does work; we certainly need to adapt these policies to our unique legal and political environment, but the underlying model has already proven itself.



*Obviously, these reforms would be no problem at the state level. A city would have a considerably more difficult time. HHS probably wouldn't approve a Medicaid or Obamcare waiver to a city, but it wouldn't hurt to ask. If it wasn't approved, we'd still only have three payers--Medicare, Medicaid, and Green Frankenstein Care which would likely be indexed to Medicare anyway. That should be administratively very simple, albeit more complicated than with a Medicaid waiver.

1 comment:

  1. Decent to see online journals like this...got a more helpful informations..spent a pleasant time with this blog.keep more updates this way..
    Self Employment Tax
    Tax Preparation Services
    Cpa Tax Accountant
    Tax Return Services

    ReplyDelete