From the comments, John Goodman on health insurance subsidies

Commenting on yesterday’s post, John Goodman writes:

Two points:

1. The premium the individual pays is not fixed as a percent of income. The subsidy is fixed, based on the second lowest silver plan premium and that amount is based on income. But the consumer is free to buy any plan. Remember, the second lowest priced silver plan may be a really lousy plan. It might have a very narrow network, for example. So, all the plans are competing against each other, with one fixed subsidy and an array of premiums. The premium an insurer charges will matter very much. 2. After 2018, the out-of-pocket premium for the second lowest priced silver plan will no longer be fixed as a percent of income. Premium subsidies as a whole will grow no faster than GDP + 0.5%, the same rate of growth that is in the Obama budget for Medicare.

The thread has some other good comments as well.


Don't forget to mention his excellent website. And his book, Priceless.

"Premium subsidies as a whole will grow no faster than GDP + 0.5%": my, how I laughed.

But, this time will be different. Those are the words that precede every liberal scheme.

Because Republicans will be so compassionate and willing to increase them for the votes??

You sell the plan with assumptions like "costs will increase at the rate of GDP plus .5%" then blame the the republicans for lack of compassion when the plan predicably results in disaster.

Sounds good to me!

That's why we have a two-party system: one party to do horrible things to the country, the other party to be blamed for them.

Can anyone explain more the import of the first point? What's the implication. Why is this important? Pardon my ignorance.

All plans will compete with each other, but the entire cost difference will rest on buyer so more expensive plans won't be able to charge much more without becoming so expensive no one will buy them while less expensive plans will cover more of the cost and can end up nearly free. Do you pay more for a plan that covers more, or not and apply the savings directly towards out of pocket costs? The latter becomes quite attractive and adverse selection should be less of a problem than traditional insurance.

But isn't that always true. The choice to go with a cheaper premium & apply the saved money to deductibles always exists? Even in a non-subsidized situation.

In a fixed subsidy situation, the cheapest options always get more subsidized as a percentage. The premium an insurer charges always matters, even in a perfectly free market.

I'm still not seeing the deeper point here.

A number of the commentators, including the one Tyler quoted in the blog post basically claimed that the subsidy structure insulates the the buyers from premium increases. Cohn wrote:

It means that rising premiums won’t affect the willingness of those people to enroll—which means, in turn, they’d still have incentive to sign up next year, as long as the technological bugs were gone and Obamacare online was working.

Goodman's point was summarized in the last sentence of point #1. And point #2 is still be ignored here in this thread, I might point out. I am of the opinion of Bakersfield, below- price controls will probably be tried almost immediately. The other option will be to buy the insurers off with bonuses.

The basic point is that if most Silver plans experience a cost spiral, but two plans in a region don't because they had various features making them unpopular to those in poor health who expect to use the plan a lot-- such as a higher deductible, a less expansive doctor network, then, contra a commonly made point, people will not have the same willingness to enroll.

If people sort into plans based on health status, then the end result is not terribly similar from having high risk pools and so forth. "Percentage of expected costs covered" (used for the Bronze/Silver/Gold/Platinum rating) isn't a sufficient metric to prevent this.

A different result occurs If sorting occurs on plan metal boundaries, so that the people sort so that the people in Silver plans are sicker than those in Bronze plans. That actually increases the subsidy. So there's a strong incentive from both the patient and insurance company perspective to arrange it so that Silver plans are for those in poor health, and Bronze plans for the healthy, in order to maximize federal subsidies.

Several unusual effects exist if a significant portion of the market is subsidy eligible. Many of the effects are dramatically decreased if a large enough proportion of the exchange market is unsubsidized.

One side effect is that someone who is old in many cases would actually pay less than someone who is young for the premiums Bronze plans in many markets, if both are at the same subsidy-eligible income level. It has a weird effect of reversing the modified community rating in some sense-- while normally you're worse off if you're particularly healthy in a rated age group where people are on average sicker than if in a healthier group, you'd actually be better off if subsidized.

Hypothetical numbers (based on 60% and 70% expectation):

Suppose the community rating (with max 3:1 age ratio) sets the premiums for a Bronze plan to $600 and a Silver plan for $700 for a certain young age, and then to $1800 and $2100 respectively for an older age, where people average being less healthy. (This assumes-- as may be unlikely, that health doesn't matter much for picking Bronze vs. Silver.) If you're not subsidized but in very good shape, you're a lot better off if young here (even though either way you're paying more than without community rating.)

Suppose that you're at an income level so that the subsidy guarantees that the Silver plan costs no more than $500. Then the young get a $200 subsidy, and the old with the same income a $1600 subsidy. However, both are free to buy Bronze plans.

So then the young could buy a Bronze plans for $400 ($600-$200), while the old could buy a Silver plan for only $200 ($1800-$1600.) So, given the same amount of health, oddly you're actually better off in the less healthy group. Note that this effect actually gives the healthy old an even stronger incentive to switch to Bronze plans, so the Bronze plan premiums should actually go down (and Silver up) for the older age group in the example, since the original premiums were based on the Bronze and Silver recipients .

Couldn't your worry be fixed by tweaking the rules to match the subsidy to the second lowest silver plan premium with a certain minimum enrollment?

It doesn't entirely change it, since you can still have a plan that has a minimum enrollment, but only of the very healthy with low expected costs, because the plan is unattractive to people that already have a lot of doctors and high known costs (as opposed to the risk of unusual costs.)

I'm personally not entirely opposed to the end result, and I'm not certain it's something that needs "fixing," but I'm pretty sure that people who supported the ACA don't want a situation where the healthy all join low premium plans, and those in poor all join high premium plans. If that does happen, there's an enormous difference in the amount of subsidy in whether the plans for the healthy end up being all Bronze, or whether there are two such Silver plans in an area.

Insurance hasn't been subsidized in that manner though. Previously the coverage was pro-rata coinsurance giving strong incentives for the highest users to get the highest policy for coverage and have proportionally more covered by insurance. They will be paying for these costs themselves now rather than through sharing.

The import of the first point is that in a Vickery auction (2nd price auction) the bidders have no incentive other than bidding at their reservation price or true value. The second price sets the price, from which others can deviate, if they wish, but at a cost of fewer enrollees, or they can compete by varying the quality (plan size) at the same price, and compete on quality.

What is interesting to me is how Vickery auctions have crept into other government programs.

A while back, the general counsel of one of our clients, a non-profit student loan guarantor, asked me to explain to him what was this new auction thing he had to compete in for student loan guaranteee services. Lo and behold, it was a Vickery second price auction. I suspect, but have not confirmed, that the CEA and the Antitrust Division economists (Carl Shapiro and Dennis Carlton) may have had a hand in this as the Division had been doing work on auction markets for several years, including the time they were at the Division and the CEA.

Even in the absence of a second price subsidy match policy, providers with price higher than a reference always take fewer enrollees almost by definition of elasticity. How does a Vickery auction shape this?

Competing on non-price parameters (plan size, quality, deductible etc.) is also a strategy hardly unique to this subsidy model or Vickery auctions.

Well, if you consider that the plan which prices higher might have a larger network, which for some persons is more desirable, that plan could charge a higher price. Some people buy gasoline at the neighborhood gas station, even though they could purchase at a lower price elsewhere in the market. If you price above the Vickery, as the comment states, the subsidy is fixed-- you don't get more if you charge more.

The second point --competing on no price parameters -- is still competition. And, firms compete on non price and price at the same time. It's not either/or.

That thread was great. A nice back-and-forth between a number of well informed commenters. I think we're almost ready for Congress to put out the legislation for a vote!

Government price controls on all healthcare insurance premiums (and healthcare services themselves) will be the eventual "fix" dictated from the Potomac overlords.

The end-game is easy to predict -- widespread scarcity of healthcare services in the US. Cross-Border medical boutique services will thrive in the private market sectors of Mexico, Caribbean and elsewhere.

I'm still wondering how the scarcity endgame will result: For healthcare to be scarcer ultimately providers would have to exit these markets. Is that happening or likely to happen? Are we seeing doctors shut shop because their payments are too low?

I'm pretty unclear how that is going to happen with the ACA. For all the talk about the web site and exchanges, they are still but a sideshow in the healthcare economy. If you're a doctor most of the people who come through your door are going to be on Medicare or employer based health insurance, with some portion of people paying cash. If anything you might get a few more patients given that there will now be more people with insurance.

How is the scarcity of heath care services for Medicaid recipients arising then?

Pretty sucky, but I wasn't aware that the ACA established Medicaid. Last I checked it was established in the late 60's.

But that's not scarcity of health services, that's just a cheesy benefit. Lots of people on food stamps are only getting something like $100 a month or so. You can't buy much food with that. That hardly means food is scarce in the US where almost every demographic group is seeing obesity increasing every decade.

Boonton, I don't think he's saying that the ACA established Medicaid, merely saying that people with ACA exchange insurance could end up like Medicaid people.

Among people who were previously uninsured, it would be a "scarcity" in the sense that they couldn't actually have all the care that the insurance plan description seemed to promise, but would almost surely be more than they got before.

However, don't forget that a lot of people were previously insured with individual plans. It's certainly possible that a lot of them would experience a drop in care performed compared to their own status quo. They shouldn't be treated as just a "sideshow," they're similar in numbers to the percentage of people who are going to gain insurance through the exchange.


You're saying a person who was uninsured before might end up with more coverage and healthcare than he had before but not as great as if he had a really great insurance plan? Considering on the exchanges you have multiple tiers and have the option to buy a gold or platinium plan if you're willing to spend the money I'm having a hard time feeling all that bad here. Right now the system is more or less you get what you can pay for (whether directly if you're buying insurance OOP or indirectly if it's through your work's benefits dept) and if you can't pay for nothing you get more or less nothing. You're saying that it would be a bad thing if everyone at least had decent coverage with the option to do better if they work harder or put more of their own funds into it.

On average, Medicaid recipients receive more health care than if they didn't have insurance, even though they have a difficult time finding a doctor who will accept it.

However, the extra health care that they receive doesn't particularly seem to result in better health outcomes (though it does seem to reduce stress about not having insurance.) Partially this is because for many of the preventative type healthcare and wellness factors, compliance is more difficult and more of a barrier than expense.

Barriers to entry? Artificial Supply side restrictions? APA cartel throttles new doctor numbers & enjoys resultant high prices?

Doctor lobby sabotaging MD-alternative tracks (PAs / NPs etc.) and other technical innovations (e.g. dermatologists opposing skin cancer screening AI devices, anesthesiologists opposing automated devices, radiologists opposing remote interpretation) Bureaucratic board certification rules.

Seeing how in education subsidies seem to be captured by administration they kind of demands price controls.

This illustrates what I think is an advantage the ACA has, it's remarkably flexible and dynamic. Most people who complain about it being too complicated, or too radical, neglect to consider just how complicated and radical their own pet solutions would be. From the left perspective, a national single payer system entails abolishing all insurance and having gov't set payment levels for all existing and new medical services. You can argue back and forth whether this is a good idea, but clearly it would radically disrupt how almost every American pays for their health care. Likewise even though such a bill may be 'less complicated' as measured by some stupid metric like # of pages, it's not less complicated in its implementation. Just consider the transition to single payer along would probably entail thousands of pages of regulations and multiple court cases. Now single payer advocates can argue that on net it will eliminate the complexity of having doctors billing multiple insurance companies, navigating numerous payer systems etc. But whatever the merits of that argument the system itself would not be free of complexity.

Now consider right wing proposals. Abolish employer based insurance? You're talking disrupting how nearly 50%+ of working people have had insurance for generations now. Huge voucher schemes to buy private insurance? Err hello the exchanges are only expected to be covering 7M people, and they want to cover 300M+ with vouchers including the entire Medicare population?!!!!

While both types of ideas seem simplier when reduced to some talking points and powerpoint presentations, they really aren't. They both consist of risky gambles, betting the entire system on a single model and throwing away all other models the economy has on the assumption that they will not be needed anymore (if you abolish employer insurance and discover you screwed up big time, getting it back is going to be hard, so is pulling back Medicare-Voucher, or a single payer system).

Now consider the idea quoted above. It basically sets a cap on subsidy growth at slightly above GDP over the long run. If that results in people being very price conscious about medical expenses and 'bending the curve' downwards, that's fine. But what if people don't want the curve bent downwards? What if they find that they are willing to spend a larger share of income on healthcare because health care innovations are more worthy than other types of innovations (i.e. such as '3-d tvs' or the next generation of ipads)? Then they can. What if they think gov't should fund additional subsidies? Well that question can be taken up in the future and debated in light with what our budget situation looks like in the future. In the ACA you have a lot of flexibility for the system to evolve because it essentially let's the multiple systems we have in the US work and allows the ones that work best to expand and the ones that work less to contract. It may very well be that one set of systems is so great that they come to dominate the market (liberals are betting private insurance can't work in the long run, conservatives that the single payer-systems like Medicare/caid have to eventually convert to private insurance). The ACA could evolve in either of those extremes but it can also evolve into a mixed system (the elderly are on Medicare, the very poor on Medicaid and everyone else is in a robust private market where employer based coverage competes with individually purchased policies)

I don't think your strawman critiques of either the left or right viewpoint make sense. Fans of single payer can point out that Medicare and Medicaid *already* have thousands of pages of regulations setting payment levels for all existing and new medical services.

The cap on subsidy growth is no more or less flexible than caps on Medicare payments (such as the SGR), or caps in the growth rate of subsidies offered through vouchers. It's not a special flexible feature of the ACA; it's subject to the same political and budgetary pressures.

The right wing idea of converting from a deduction to a refundable tax credit that could be spent on either employer based policies or individual policies would not "abolish employer based insurance." It would give flexibility by treating employer based insurance and allow individually purchased policies to compete with it. In such a situation, just like what with you claim with the ACA, the system could evolve with the multiple systems. Employer based insurance could stick around in such a situation, or it could contract. Just like under the ACA-- note the offering of subsidies under the ACA that are not available under employer insurance can push both employees and employers away from employer insurance (and note that the ACA's "Cadillac tax" discourages certain employer based plans as well.)

Similarly, fans of single payer would point out that they were talking about the "public option" of expanding Medicare and Medicaid to more people. Again, they would say that the system could evolve; if people wanted to buy into Medicare then they could, otherwise they could keep their current option.

In all three of the systems, there would be a lot of arguments about the details of the regulation pushing people towards one answer or another-- the ACA having one method, deductions, for employer insurance, whereas another method, subsidies, for individual insurance gives a lot of room to radically affect the market not via a "natural evolution," but by putting a thumb on the scale. Since deductions are worth more at higher income rates, but the subsidies are worth much more at lower income, the ACA inevitably pushes things towards a two-tier solution even more than the existing deduction for premiums alone did. (Even with the employer mandate, at the low end both employer and employee can be better off by dropping coverage.) The same sort of arguments would come up in the other plans-- is the "public option" being subsidized too much or not enough to fairly compete? Are things like Medicare Advantage having too high payments, pushing people towards private plans? Etc.

You exaggerate the differences.


A very interesting comment. A few points:

1. "The cap on subsidy growth is no more or less flexible than caps on Medicare payments..."

No, if Medicare declares it will only pay $75 for an office visit it is not legal for patients or doctors to offer additional cash OOP. If that's sufficient then doctors are happy and patients have lots of choices. If it isn't then few doctors accept Medicare and patients experience long waits and few choices.

A subsidy cap basically means if today the plan costs $400 a month with a $300 subsidy tomorrow it can be $550 with the same subsidy. If consumers want to spend more they are free too, otherwise insurance companies are competiting with each other to supply insurance for the lowest possible cost which puts pressure on them to control costs.

2. "The right wing idea of converting from a deduction to a refundable tax credit that could be spent on either employer based policies ..."

The problem is employer based policies are essentially set on 'community rating' so all pay the same price. To keep this you'd essentially end up with requiring everyone to go on the exchange. And let's face it, while in theory you can say "if you like your employer provided insurance you can keep it", you really can't. Many insurers would drop insurance, give their workers a raise and tell them to go buy insurance on their own.

Perhaps that's a good thing but why not start with the uninsured and businesses under 50 buying insurance with help where income is a problem and see how that works before disrupting the majority of people who have no problem with getting insurance from work? And if that is the best direction to go in, it's not like the ACA would prevent that. You could convert employee/er tax deductions into credits expanding the method of individuals buying insurance just as easily as one could expand single payer by increasing Medicaid or letting people buy into Medicare.

"No, if Medicare declares it will only pay $75 for an office visit it is not legal for patients or doctors to offer additional cash OOP."

I don't doubt that you are right; but how do such clauses ever stand the test of constitutionality. i.e. Isn't it highly restrictive to legally criminalize my giving a private doctor extra money entirely voluntarily? Or is this considered a bribe in some highly contrived reasoning.

Is there a legal precedent here? I'd be curious to know. Has such a case ever been litigated? IMHO the worst the government ought to be able to threaten is to just de-roster the doc from government programs.

IMHO the worst the government ought to be able to threaten is to just de-roster the doc from government programs.

That's already the case. The payment caps for Medicare are conditional on your hospital accepting Medicare funds, which is a government program. Likewise EMTALA, the act which famously compels hospitals to provide emergency care regardless of ability-to-pay; it too technically only applies to Medicare-accepting institutions.

It's just that hospitals which refused Medicare subsidies would be rapidly insolvent, since these account for nearly half of all healthcare spending, public and private both.

Niche boutique hospitals that do not accept Medicare funds at all are not bound by the conditions.

The problem is employer based policies are essentially set on ‘community rating’ so all pay the same price. To keep this you’d essentially end up with requiring everyone to go on the exchange. And let’s face it, while in theory you can say “if you like your employer provided insurance you can keep it”, you really can’t. Many insurers would drop insurance, give their workers a raise and tell them to go buy insurance on their own.

Not particularly more so than insurers right now have incentive to drop insurance under the ACA and tell people to go on the exchange, particularly if their employees tend to be at the subsidized level. One big problem with the ACA scheme is that for a company like Trader Joe's with lot of fairly low income employees, dropping coverage is a win-win for employees who are single, with Trader Joe's their only job, but it's a bad deal for, say, married couples for whom the part time work or lower pay work is a second job in the family. The tax credit effects would be smoother.

With a tax credit, I suspect that there might still be good reasons for employees to seek group coverage, and thus for employers to offer it as a benefit. (Employers offer a variety of benefits that aren't mandated, such as cafeterias or other food, offered by some but not others.) It may be that most plans get dropped without a bias towards employer insurance, or it may be that people prefer the status quo. Wouldn't the process of finding out be part of the same sort of flexible evolution that you praise in the ACA?

It seems to me that you're praising the flexibility in the ACA while attacking other plans (of both left and right) precisely for that same flexibility. The incentives for employers to drop coverage are in the ACA as well. As I said, the ACA is a little bit different because it preserves this two-tier (or three-tier) nature, where deductions are more valuable to people with high income and high marginal tax rates, but subsidies are better for the low end. (People with lower middle class incomes a little too high for subsidies but with lower marginal tax rates so that deductions are worth less for them end up getting squeezed.) Thus the ACA isn't really a "natural evolution," it's a system that highly biases the higher income in favor of employer insurance while highly pushing the lower income towards subsidies on the exchanges. Such a two-tier system could result in individual plans mostly being second-class inferior plans, even.

Note that the employer mandate is being delayed, but even with it, a large company whose employees make a small enough income on average would still find it beneficial for both employer and employee to push them onto the exchanges. (They would be right morally to do so, since it would benefit the employee as well.)

It's true that the ACA is designed to have a lesser effect on people (largely at higher income) getting employer based insurance by keeping the deduction and adding low income subsidies, than a plan to convert the deduction to a universal credit aimed at both employer and individual health insurance.

However, by doing so, the ACA imposes more costs on those currently buying individual health insurance in order to avoid putting costs on people currently getting employer insurance. People in the lower middle class on the wrong side of the subsidy cliff get stuck with a huge part of the bill as our price for exempting the upper middle class.

Yes, all of Amerikan health care, as well as health insurance, is complicated as hell. Try calling around the USSA for an all-inclusive pricing for a cataract surgery, for example. You will have to negotiate a cash discount with surgeon, facility, anesthesiologist, and drugs separately, making numerous phone calls and providing details about duration of surgery and CPT code, etc.

One call to a clinic in Thailand, India, Prague, Budapest, Costa Rica, or Mexico will get you am all-inclusive quote that is 1/3 what you pay in Amerika. No muss, no fuss.

A possibly wacky idea I've had is to introduce some sort of voucher system akin to school choice. What if your insurer actually let you spend your money between a list of hospitals in Thailand, India, Prague, Budapest, Costa Rica, or Mexico? Possibly incentive by some carrot like a lower deductible or premium perhaps.

For a lot of elective & non-critical procedures (e.g. cataracts, hip replacements) the economics are likely to be very very attractive. It needn't even be one of these "cheap" nations: done even in a place like Canada or Germany I suspect there's a fairly attractive cost calculus.

Dean Baker actually mentioned something similar, if I recall right, when he was on Econtalk in January of 2012. I remember Russ Roberts kind of brushing the idea off because he preferred that the government got rid of the subsidies/voucher as opposed to finding a way to make them work better.

Not surprised by Russ Roberts. He's highly annoying at times, letting his fantasy conceptions of the ideal be totally ignorant of ground reality.
s/b rebranded as

May I simply say what a comfort to discover someone who genuinely knows what they're talking
about on the internet. You actually understand how to
bring an issue to light and make it important. More and more people should look at this and understand this side of your story.

I can't believe you are not more popular since you surely possess the gift.

Comments for this post are closed