Medicaid Isn’t Worth Its Cost

Medicaid isn’t worth its cost–that’s not my evaluation that’s what people who use the program think, at least as far as we can tell from their actions. Joshua D. Gottlieb and Mark Shepard review the evidence at Econofact, which aims to be a dispassionate and non-partisan review of the evidence on a variety of issues. We have also covered these issues before but seeing it all together is valuable.

The cost is large:

The Medicaid program cost about $532 billion in 2015 to cover 74 million people, or almost one in four Americans. The average full-benefit enrollee cost about $6,400 per year to cover in 2014.

People with access to the program use a lot more healthcare than other similar people

The Oregon Experiment found that gaining Medicaid uniformly increased health care use: including hospitalizations (by 30 percent), emergency room use (by 40 percent), physician office visits (by 50 percent), and prescription drugs (by 15 percent). This evidence stands in contrast to the conventional wisdom that providing health insurance could reduce costs by eliminating ER visits. Of course, understanding whether this additional care is worth it requires a comparison of these real costs to the benefits provided.

The health benefits appear to be real but modest:

The evidence is mixed on whether having Medicaid improves beneficiaries’ health. The Oregon Experiment did not find statistically significant evidence of improvements in physical health measures, such as blood pressure and blood sugar after two years of coverage. But it did find large improvements in mental health and self-reported health. Other studies examining the introduction of Medicaid or its expansion over time have found that Medicaid reduces mortality (of infants during the expansion of Medicaid eligibility for low-income children between 1984-1992; of adults during the expansion of Medicaid coverage for childless adults in Arizona, Maine and New York between 2000-2005; of teenagers who benefited from expansions of Medicaid to children during the early 1980s; and of infants and children in the 1960s and 1970s following the introduction of Medicaid) and improves health later in life (for instance among teenagers who benefited from the expansion of coverage as children). But these studies lack the gold-standard randomized design of the Oregon Experiment so should be interpreted with greater caution.

Health benefits may not be the most important benefits:

One important role for Medicaid is to provide risk protection, shielding enrollees from the financial impact of particularly adverse health events, which is the most fundamental role of an insurance product. Researchers seem to agree that access to Medicaid does improve financial security.

So how does one evaluate the tradeoffs? One way is to look at how users value the program.

Recent evidence indicates that beneficiaries value Medicaid at less than its full cost. One source of evidence comes from Massachusetts’ low-income health insurance exchange, where researchers could observe how much charging higher premiums for Medicaid-like coverage led enrollees to drop out: at least 70 percent of enrollees valued insurance at less than their own cost of coverage. A second source of evidence used economic models to quantify how much beneficiaries valued the benefits of Medicaid in the Oregon Experiment. In this case, the researchers found that beneficiaries valued Medicaid at about one-fifth of its cost.

Benefits are valued at only one-fifth the cost!  Why so low?

The literature suggests two explanations. First, Medicaid provides less complete choice of doctors and hospitals than other insurance, partly because of its low reimbursement rates (see this article for instance). Second, many of the benefits of Medicaid go to medical providers who would otherwise provide uncompensated or unpaid care to the same people.

The authors don’t mention this but if users don’t value the program highly because they would have gotten similar care for free in some other way, then the cost of Medicaid isn’t as high as it appears, because much of it is a transfer from taxpayers to medical providers or others who might otherwise foot the bill. Nevertheless we would probably design Medicaid very differently if we thought about it as (another) subsidy to medical providers rather than as a subsidy to the poor and sick.

It doesn’t follow from anything that has been said that Medicaid should be eliminated or even cut back (let alone that current efforts are the best way to do this). Nevertheless, if I told you that Program X costs $5 for every $1 in value transferred to recipients you would probably agree that Program X was in need of reform.

Addendum: Aaron Carroll and Austin Frakt offer a more optimistic review of the health evidence.


People eligible for medicaid are probably less afraid of bankruptcy than most and at that income level other expenses are often more immediately pressing.

But confounding all of this is that they can often be enrolled in medicaid after the occasion for any great need for it arises.

And maybe I'm missing something here, but (for a low-income enrollee) isn't willingness to pay constrained by ability to pay? So someone may "value" Medicaid at $6,400/yr but have to opt out around one-fifth that price.

If anything, the fact that poor people forgo coverage when priced at its "true" value is (in my view) a pretty good argument for artificially lowering its price.

If as group people making $14K-$23k/year decided health insurance was important enough to pay $6,400/year, rents would fall homes would get smaller, they would eat less meat and more beans and rice and they could do it.

I pray that you get the chance to lead them by example

Also another way to judge that is to ask, if you had health insurance how much would you require to give it up. See my comment below.

Don't bring up these kinds of issues. In LibertarianWorld these kinds of constraints don't exist.

If we measure "willingness to pay" just in dollar terms we miss the point. "Willingness to pay" means "willingness to give up something else for health insurance."

It's one thing if that something else is a new car or a nicer vacation. It's quite something else if it's rent or food.

What is the value of something that is free? It has no value. But for those who have to pay for it there is most certainly a cost. The solution is to require medicare recipients to pay for their care. Create a work requirement and a drug test for all welfare recipients. For Medicaid patients require a co-pay of perhaps $15 per visit.

End wealth distribution. It is unjust and unconstitutional. While the states "may" be able to constitutionally tax Peter to pay Paul the federal government cannot. End it and bring the federal budget back into balance.

Yep. And let's require corn and soybean farmers who take taxpayer funds to take a drug test too. And people who take advantage of the home mortgage interest deduction, and federal student loans, and the child tax credit (and have the kids take a drug test too while you're at it). That'll show EVERYBODY.

Read Frakt's piece first, whose headline reads: "Medicaid Worsens Your Health? That’s a Classic Misinterpretation of Research"

Post should read: Alex's Death Panels, or, Do You Have Room in the Basement for Grandma?

I also like the criticism that Medicaid is bad because it pays providers less, or that not all doctors take medicaid patients, as if they would take poor people who could not pay.

What was the world like before Medicaid? Here's the answer from Politifact:

"It doesn’t follow from anything that has been said that Medicaid should be eliminated"

There is a big logical fallacy in the valuation model presented in this post. The logic of the valuation model doesn't capture externalities, cost shifting to other insureds, and myopic behavior of rich and poor who believe they will never need insurance relative to their other more immediate needs (hyperbolic discounting of the probability of future adverse health events). Let's take an example that is not as extreme in its consequences of not getting it and see how the valuation model works. Education has externalities; low income persons (and children of the middle class who have no income) might value current consumption over spending on education that will increase income in the future. The valuation model cited in this paper would say the following: If a state subsidized higher education at state universities so that tuition were low, and the state made a change and raised tuition to full cost, with 70% of the students dropping out because they could not afford it, then that shows that students did not value the education they were previously receiving at the state school.

Well put. People's valuation is not necessarily an accurate proxy for a thing's actual value. In this case, the covered medical expenses per enrollee seem a better, if still imperfect, guess.

As I recall, willingness to pay for education on a near-subsistence income is about 0, which would indicate that my willingness to pay for education is about 0.

However, with access to government subsidized credit guarantees, my willingness to pay was whatever they were charging. (Which can itself be a problem if there is not control on what education providers are allowed to charge - extremely relevant for the case of the USA, but not much at all for the case of Canada.)

So yeah, piss poor methodology, to say the least.

Believing that you know better than others and that therefore you should be a privileged commisar ordering everyone else around is a cornerstone of progressive ideological economic policy. When they don't follow orders, the gulags come into play.

@Thomas: According to your warped standards the U.S. is the only place on earth not suffering Communism, being the only developed country without universal health care. But then again, even the U.S. levies taxes, so that must be Communism, too.

Thomas, We live in a democracy, or at least we did at one time, and we make collective choices--when everyone votes, and their votes are not suppressed, it is not the commissar speaking, it is the people.

The other important fact about MCD is that 75% of RECIPIENTS are poor young people, but 75% of the SPENDING is on nursing home care for elderly middle class people.

Good point. Financial advisors for years have been telling old persons in declining health to distribute their assets to heirs early, thus making themselves "indigent" and "unable" to afford their own end-of-life care.

I get that people want to leave money to their children, but it's hard to defend committing effective fraud to do it.

"Recent evidence indicates that beneficiaries value Medicaid at less than its full cost." How does one on Medicaid "value" it? Did someone ask her to guess the price of the services rendered that would have been charged by the providers in a fee for service system? Did she believe the providers charged too much for their services? Indeed, does anyone not believe providers charge too much for their services? In the ancient world faith healers were the health care providers. Did the patients of John the Baptist value his services at less than its full cost? Did John the Baptist charge his patients by ability to pay or did he have a standard charge for everyone? Did he offer his services for free to the poor? There were a lot of poor people in the ancient world, so I assume faith healers had to give away for fee a lot of their services, making up the difference by charging the non-poor more than the value of the services. America is a rich country with lots of poor people. Should today's health care providers offer services for the poor for free, as did the health care providers (faith healers) of the ancient world? [No, I don't know whether John the Baptist charged for his services (his diet consisted of locusts and wild honey, so I assume he didn't charge much), but it was common practice for healers to charge their patients.]

What would be the poll response if people were asked whether we are paying too much to members of Congress and the President given what we have been getting for the cost?

What would be the response to whether the amount paid to the military in the 21st century is worth the cost given the results?

What would be the response to the question whether the huge cost over the past seven decades of building nuclear weapons and the means to deliver them was worth it?

"Recent evidence indicates that beneficiaries value Medicaid at less than its full cost"

Poor people's health has a lower $ value than rich people's health. That doesn't mean that it has lower human value.

I say this as a devout capitalist.

Define "human value"

What is the principled "devout capitalist" (free market) view of any Medicaid-type government welfare program?

> Define “human value”

Nah. That's a deep rabbit hole and the argument doesn't hinge on a precise definition.

> What is the principled “devout capitalist” (free market) view of any Medicaid-type government welfare program?What is the principled “devout capitalist” (free market) view of any Medicaid-type government welfare program?

Skeptical, of course. I wasn't defending medicaid, only pointing out a flaw (one of several) in using demand to determine value. That's always a problem but is particularly severe when discussing healthcare.

Obviously using supply to determine value is the best way to determine value? If the supply of $100,000 drugs increases while the supply of $10 drug cures shrinks, capitalists are increasing the value of human life.

"Capitalists" want to be paid.

So, without Medicaid, they do not get paid to do the work the law mandates.

But if that mandate is eliminated, they still don't get paid.

Would the US be better off if gdp were $1 trillion smaller? With $1 billion fewer workers?

Very few of the people who benefit from Medicare and Medicaid have the means to replace the government spending. And the "rich" who pay the taxes are mostly their younger selves who would not have saved the money because they would have been spending it on their parents, or had less income because someone was working caring for parents.

And if you do save money for old age, but then some problem occurs, then that cash must be spent now, your future old age be dawned. Few families go three decades without some economic crisis which taps into savings, sometimes very deeply.

What business tells customers "save your money"?

"Medicaid isn’t worth its cost–that’s not my evaluation that’s what people who use the program think, at least as far as we can tell from their actions."

I don't follow this. Suppose I make $12k a year. Most of that goes to rent and a car so I can get to work. How much would I pay for health insurance and the financial security it brings? Not very much. But does that mean that that I assign a low value to health insurance? We can define "value" that way, but it doesn't strike me as very interesting. All we would be saying is that poor people assign a low value to everything. For example, suppose I have a bar of gold and try to asses the value by selling it to a poor person. A poor person cannot offer you more than the cash they have, maybe $500, but we know it is worth 1000 times that on the open market. It is not clear that "what a poor person would pay" is a good way to do valuations.

I disagree. It is clear that it is a quasi-fraudulent way to do valuations.

"The only valid method of valuation is one that matches my preferences, and I intend to use guns to force my preferences on you."

What does relatively extreme application of libertarian propaganda have to do with the question of whether the valuation methodology is bad?

So, them guns took your taxes. OK.

Is the valuation method good?

Well, the poor go to the ER and get medical care which they can't pay for. Thus they get free medical care. Part is driven by EMTALA.

If EMTALA required selling a bar of gold to the poor, who could not pay more than $500 over the next decade, would the value of a bar of gold be $500.

That's Alex's argument.

"The authors don’t mention this but if users don’t value the program highly because they would have gotten similar care for free in some other way, then the cost of Medicaid isn’t as high as it appears, because much of it is a transfer from taxpayers to medical providers or others who might otherwise foot the bill. Nevertheless we would probably design Medicaid very differently if we thought about it as (another) subsidy to medical providers rather than as a subsidy to the poor and sick."

Ie, Medicaid is a subsidy to the gold mining industry which can be far cheaper by using EMTALA to mandate selling a bar of gold without payment in advance.

The fact that it is not **completely** free is what causes people not to like it. The natural comparison in that economic pool is completely free emergency room treatment.

Three generations of employer-provided health insurance taught people that healthcare IS free or, at most, that it costs only a $10 copay, even for an extended hospital stay.

When you change the rules of a deal like that, people resent it. They don't want to budget for health insurance because neither they nor their parents ever did it before. They believe that being held responsible for their own healthcare or health insurance is some kind of extortion or swindle.

It would be nice if we could get people to acknowledge, for starters, that healthcare isn't free. But when uninsured people are confronted with $175 bills for simple blood tests that should cost $20, as often happens, they are understandably resentful. And when you DO pay for your health insurance and your expenses are lower than your $3,000 deductible, you feel like a sucker.

(It also would be nice if we could get senior citizens to acknowledge that, on balance, they are drawing far more in benefits out of Social Security and Medicare than they ever contributed, but we're not good at facing simple facts that make us uncomfortable.)

If we adopt a single-payer regime maybe people will be a little happier, but they still will be angry if they have pay taxes for it.

This post proves libertarians are idiots.

All seven bullet points are inane. And, of course, the focus is completely misplaced. About 30% of Medicaid's budget is devoted to long-term care, a function handled historically by state asylums and sanitoriums, institutions which are no longer in existence. Bracketing out l/t care and provision for the elderly, the disabled, and dialysis patients, the gross output per person in the medical sector is about $4,800 per person. His complaint is that Medicaid beneficiaries receive $6,400 in benefits per year. Well, 70% of that is $4,500 per year, i.e. not terribly different from the average American not eligible for Medicare. Your complaint is that there is inefficient allocation in Medicaid? You fancy there is something resembling efficient allocation outside of Medicaid?

You can live with Medicaid, craft something which might provide equivalent service but be less suffused with the pathologies of contemporary medical finance, go full Ayn Rand, go full Ivan Ilich, or (like the normally sensible Thomas Sowell) subscribe to the notion that installment payments and philanthropy can handle medical finance (or could have if we hadn't mucked it all up in 1965). Or you can be an evasive twit.

Actually, historically long term care was overwhelmingly provided by kids, or parents, at the opportunity cost of not working and consuming. Gdp was lower because the work was done for "free" with net reduced production AND consumption. Long term care in the homes of children, or maybe by children living in the parent's home limited "innovation", ie, a young person engaging in new ventures by moving to and focusing on a job opportunity.

Social Security partially cut the dependency of the old on their children, and Medicare/Medicaid further cut that dependency.

My parents' three kids pay a lot in taxes to one degree or another, me more than my siblings did for several decades, so we were free to work without being forced to pick someone to house and then provide round the clock care for the last three years of our mother's life. Only three years because Medicare extended my dad's life for several years so he provided the round the clock care as dementia took her mind, until it wore him down and probably shortened his life by 6 months, then Medicare/Medicaid/State welfare.

In my youth, there were still three and four generation households with my classmates parents caring for their parent. But my dad was involved in developing independent and assisted living communities to provide better environments and care because changing housing and jobs, (lots of moving), made 3 generation households impossible.

Read something that indicated 14 or 18 States still have laws mandating kids care for parents unable to care for themselves. Obviously unenforceable for many modest to low income children renting with terms limiting household size and relationship to two generations.

"The average full-benefit enrollee cost about $6,400 per year to cover in 2014."

Damn. Isn't that a bargain? I know costs are growing quickly, but still.

I think if individuals could buy into Medicaid at that price, many would.

Kind of my immediate thought. I wouldn't be surprised if lower-income households could the use money better in other ways. If I were in massive credit card debt, I'd want the cash. $6.5k isn't insignificant if I am working with FHA to get a rehab loan, either.

Assuming you are judgment proof "massive credit card" debt is not really a problem. You ignore the dunning letters and collection agency calls. If worst comes to worst you declare bankruptcy.

Exactly. If we think that it's appropriate to help poor people access health care, and I think there are few these days that completely oppose it, then Medicaid is the most efficient way to do it. Aside from really big departures from the status quo (e.g. a nationwide investment in expanding public clinics that serve this population directly), it's by far the best approach. Giving them subsidies to buy insurance on the private market will cost a lot more and would be near useless for many poor people because the deductibles and other out of pocket costs would far exceed what they could pay (e.g. half of gross income for many ppl in the current Senate bill).

+1 for expanding public clinics. I think if it were possible to make it budget neutral (reduced medicaid costs for expanding clinics), this could fly. You'd end up with people well into middle class incomes at these places.

Many would buy in, but they would skew to those who need more care. The point of the study is that the average medicaid beneficiary wouldn't pay $1200 for what they get.

In fact, this is what happened to the ACA. One Pennsylvania insurance executive revealed in a hearing last year that in 2015 his company covered 200 individuals who racked up $100,000 or more in costs, each, and who then canceled their policies at the end of the year. The penalty for going uninsured was cheaper, almost certainly cheaper than the below-cost $1,200 figure cited in the article.

People respond to incentives.

But that is exactly what they were told to do by the Republican Party,

Who also say that no one died for lack of money on insurance because....

"The authors don’t mention this but if users don’t value the program highly because they would have gotten similar care for free in some other way, then the cost of Medicaid isn’t as high as it appears, because much of it is a transfer from taxpayers to medical providers or others who might otherwise foot the bill. Nevertheless we would probably design Medicaid very differently if we thought about it as (another) subsidy to medical providers rather than as a subsidy to the poor and sick."

They, like Alex, argue that sticking hospitals with uncompensated medical care is cost free with EMTALA.

Hey, let's bring back slavery, just this time the doctors, nurses, and other medical care workers will be the slaves, including the manufacturing and construction workers building the buildings, machines, and supplies. The health care will be free.

More needs to be said about the fact that Medicaid covers the costs for a lot of old folks living in nursing homes. That Medicaid spends only $6.4K per enrollee when it covers so much expensive care for old people near the end of their life is astounding. That said, this may also explain why many Medicaid enrollees don't appear to value Medicaid coverage that much per the studies. The people who are likely to respond are probably younger and healthier than the overall Medicaid population. I mean, how do you survey someone with alzheimer's or dementia on how much they value their Medicaid coverage? The appropriate people to ask would be their children and spouses, who I bet would consider Medicaid coverage very valuable, as there is likely no way the children or spouses could provide a comparable level of care on their own, or even if they could, it would be a huge strain both financially and emotionally.

Indiana's Medicaid (Healthy Indiana Plan HIP ) is not a conventional fee for service comprehensive insurance. It is more like a high deductible health plan. Enrollees have co pays. it treats the poor just like everyone else and expects them to act like everyone else.

I think if Trumpcare rolled out a HIP like plan to more states, with a work requirement for the able bodied, and allowed anyone to buy into it after a 6 month (or more) waiting period, we would have a very successful product.

Unless we are going to establish public jobs WPA-style along with that work requirement I would strongly oppose such a thing. What happens in a recession after all? And what about people who may be "able bodied" but are unemployable due to low intelligence or other deficiencies?

Actually, the States with work requirements end up creating WPA like programs. If your unable to work at a paying job, like you need to care for children, and the employers will not adjust schedules to childcare needs, then the work requirement is met by working for free in libraries or social service centers or schools during their kids school hours.

But according to Alex, picking up trash as a work requirement has no value because no one will pay people to pick up the trash they throw out their car windows. That means, everyone can save money by just throwing your trash out the window in other neighborhoods instead of paying to have it picked up and land filled or recycled or otherwise disposed of.

Just as car insurance doesn't make your car instantly better, nor does health insurance make your health instantly better. The concepts of having them are 1. Paying into a liability pool for large accidents (or catstrophic health events) 2. Not being able to afford fixing your own car in case of an accident (or afford more routine medical care.)

If allowed to I bet most people without any assets to attach in case of #1 would not buy car insurance (or health insurance). If you ask most people what they think car insurance is worth they would report a lower number than actual cost, as almost certianly they would value the liability risk component at zero, especially if they had no assets to attach.

Car insurance is an excellent analogy. Consider those with comprehensive coverage will most likely get their cars fixed after an accident. Those without it may or may not get their car fixed depending upon if they have the money and if the car is worth fixing.

So imagine on Jan 1 you get a list of 1000 cars with comp. coverage and 1000 cars without it. On Dec 31st some of those cars will be junked from each set. How much better will one be over the other? The difference may require some time to materialize. Of course you'd have to make some adjustments since full coverage cars are likely to be newer ones while non-covered cars may be close to the end of their lives to begin with.

Of course this ignores whether the cars being repaired is a desirable outcome. We could spend 100% of GDP on healthcare and there is no limiting principle in your pathos based argument against it. We could all starve but imagine the cars we could repair!

Why don't we have EMTALA for cars?

Alternatively, why don't we have mechanic liens for medical care? Why can't doctors take possession of patients and sell them as slaves, or chop them up to sell in pieces, like tow truck and mechanics get to do with cars they asset without being paid?

That's the bizarre thing about healthcare insurance. It doesn't actually function like insurance. It largely functions like a weird payment plan that shuffles costs around, ignoring how high they are.

Routine care should not be a part of insurance. If this happened nationwide, we'd have some sanity in healthcare prices real quick.

So, a couple of years ago, a State Medicaid system ranked all medical care by value, ie vaccinations were high value, organ transplants low value, and then based on budget forecasting denied medical care that fell below the line.

And in Britain, the NHS has determined the cost of life support for a functionally dead baby is too high.

So, what is the response of conservatives who argue Medicaid is too costly and needs to be eliminated? Trump is saying "bring the kid to the US" as if he is willing to spend millions on that dead baby.

The Federalist:

"While it’s not uncommon to involve the legal system in the United States—everyone remembers the awful case of Terri Schiavo—there hasn’t been a case here quite like that of baby Charlie. In addition to British law’s authoritarian bent against parental rights, socialized medicine operates in England via the British National Health Service (NHS). According to CNN Money, this sytem is “financed through tax and compulsory national insurance contributions, but faces serious financial problems.”

"Specifically, “The accounts of two-thirds of NHS providers were in the red in 2015, with a combined deficit of £2.5 billion in the last financial year. Prime Minister Theresa May has promised an extra £10 billion for the NHS by 2020, but lawmakers say the pledge is worth less than half that when rising costs are taken into account.”

"Given that socialism always “runs out of other people’s money,” as former British Prime Minister Margaret Thatcher noted, it makes sense NHS seeks to cut costs even at the expense of human life and despite parents’ ability to pay. For example, the cost to provide 24/7 care to a person, including life support, in the United States is anywhere from $2,000 to 4,000 per day. If Charlie was diagnosed around three months old, he’s been in the hospital approximately 151 days."

So, the Federalist promises a free lunch. Get rid of government single payer health care, which is the UK NHS and US Medicaid, and suddenly the money to pay thousands of dollars a day for care will rain down from heaven or suddenly become free because in the free market there is no rationing. A poor person needing to travel 100 miles a day to get to a minimum wage job will get a very reliable car that costs 3 cents per mile to operate for its entire ten year life.

To conservatives, the free market delivers free lunches.

No one who argues for this forgos health coverage for themselves or their family. I guarantee Alex has full health coverage even though I suspect he would be unable to point to health metrics about his body that would indicate he is healthier today than he was a year ago despite being well covered for 365 full days.

Regardless, I would consider this. Health interventions have the most dramatic impact when you are in an emergency. Having a heart attack or stroke, an emergency room makes a big difference in your health outcome as opposed to laying on your couch and having an aspirin. But if you are having a heart attack you are calling 911 regardless of whether or not you are covered.

More regular health interventions requires longer time to materialize. Taking BP medication, for example, may take years or decades to show up in terms of lower rates of heart attacks and death.

One idea, however, is paying for outcomes rather than services. Then you get cost control and quality outcomes.

I am a doc, if I could do so I would take the tax-free value of my healthcare as wages and buy a catastrophe only plan in a heartbeat. If it were legal in any state, I would immediately sign up for a lifetime healthcare plan that pays out if and only if my medical bills exceeded my expected lifetime medical bills to date (e.g. my "deductible" would *increase* by $3K per annum less a percentage of any health expenses for the previous year).

I want my health insurance to be insurance - something that protects me from unanticipatable losses. I do not want it to be a paperwork filing service that lays more hands on the transaction of moving money from my employer to a healthcare provider.

But as the law currently stands it is illegal to skip insurance, and if I am getting insurance I must pay for a bunch of completely predictable costs through a terribly inefficient process. Of course with taxation laws, the money I dump into insurance is easily 30% more valuable to me as I pay no taxes on it.

This sort of silliness induces all sorts of terrible behavior - if it costs the same to me I should use more of it (e.g. our hospital saw a dramatic rise in ER visits by a number of our "frequent flyers" once they got ACA subsidized plans). Paying for outcomes is laughably terrible. What outcomes exactly do you plan to pay for? Are you seriously expecting physicians who would pad services under a fee-for-service would not up-code a new patient and down-code them later on? Or that they might not terminate non-compliant patients who bring down their numbers or are particularly risky (e.g. as has already begun to happen for surgeons worried about their quality ratings)? Or heck, how do you plan to weight race and its known impact on disease outcome (e.g. are you going to penalize a physician who has more trouble controlling diabetes among a Pacific Islander population or will you reduce the health target for that population?)

It is ALWAYS going to be easier to corrupt the quality measure than to actually improve health, if you cannot trust your doctors under one system the problems will still be there under a different reimbursement scheme. I mean seriously, you could put all healthcare providers on standardized reimbursement and then you will see things like increased absenteeism, bed blockers, and work to rule. At some level you need to be able to trust the people you just trust with your life with your cash too.

+1. Somebody's been reading John Cochrane.

I do think it's worthwhile to track costs and outcomes - perhaps broadly, over a long horizon. Defensive medicine is one thing, but I've heard of doctors that ring completely unnecessary procedures to increase billing. I believe most people go into medicine for good reasons, but incentives are incentives. Maybe someone in law can tell us how this incentives is managed there.

I have never read him directly, I would be shocked if he hasn't just filtered out to people I do read.

In any event, I do not see how you get rid of bad incentives. Say you go for the Cleveland Clinic model, everyone gets a set salary. Okay, do they not then have the incentive to do as little work as possible before getting fired? Do they not have incentive to tell off and get annoying patients to leave? Do they not have incentive to game the system to get their relatives hired?

Capitation similarly removes some perverse incentives, but again creates new ones. Do you want your doctor to have incentive not to treat you?

Again, if we buy that incentives matter (which I do not doubt), it becomes awfully hard not create a situation wherein less than optimal care for the patient is in the physician's economic interest. After all, at the end of the day providing healthcare itself is expensive and playing with the numbers will always be cheaper.

When people compare the American health system to those in Canada or the U.K., they often look at things such as how long it takes to get hip-replacement surgery. That probably doesn't impact life expectancy but it is certainly a quality of life issue and it is only fair we consider similar metrics when comparing people who have Medicaid to the uninsured.

Uh, no.

One year mortality after a fall with fracture in the elderly is 25.2%. Among falls without fractures one year mortality risk among the elderly is 4%. Among the most significant risks of fall - failing joints. Delaying joint replacement directly kills people, every week of delay is a non-negligible increase in risk of fall and associated mortality. According to the CDC decreasing fall risk would save 27,000 Americans per annum, which the the CDC describes at the "leading cause of injury and death in older Americans". In the elderly falls lead to broken bones which fail to heal, become infected, or develop avascular necrosis. And of course the risk of delay and fall means that many procedures will go from simple surgeries to emergent ones with trauma. Pretty much any way you slice it, delaying needed surgery is deadly. If your joints are bad enough to replace, aggregate delays cause aggregate deaths directly through falls.

But let's ignore falls.

"Quality of life" includes things like exercising. Joint pain is one of the leading causes for patients diminishing their activity and exercise. This makes it that much harder to manage oh, ever single chronic condition ever. Diabetes? Vastly worse with poor mobility. Coronary artery disease? More likely to kill you if you are less mobile. Depression? Poor mobility correlates with higher suicide rates. Hypertension, renal disease, arthritis, COPD, and even Alzhemier's are harder to manage with poor mobility. Your ability to walk well is so idiotically important to your health that we routinely use it as a primary marker of clinical disease progression and even the NICE recommends spending thousands on maintaining that ability for a single year.

You can make all manner of arguments about cost effectiveness, appropriateness, etc. but mobility is one of the highest risk factors for all cause mortality that is known to exists. Delays have an obvious impact on life expectancy.

The picture is not complete until we examine the impact on the other people who pay for it. Total costs are only 3% of the economy, slightly less than interest paid out on the national debt (600 billion and climbing straight up).

But who is paying for it? If the tax impact hits middle class who do not use the service, then in that smaller group, the cost is about 6% of their income, and that leaves them with severe health impacts, mainly the inability to afford a family. Example, Illinouis is behind on medicaid payments, got sued. But Illinois is losing its middle class as fast as Detroit lost its middle class.

Aside from your GDP*percentage calculation, most of your post is a product of imagination.

Also, if your concern is about affordability for the middle class, consider Canadian or European models which, weighted for size of the US economy, would save about a trillion dollars a year and lead to superior results.

I didn't imagine anything. I proposed the standard two sided flow of funds to isolate effects of medicaid, then suggested some possible views to find it.

Many of the other comments are single entry accounting, the money arrives out of this air. No economist, nor any mathematician would accept that, they would tell you folks to get back into statistics class, you are not contributing anything with single entry accounting, it went away in the 1500, and we don't use it. Assets and liabilities. Show me both or get back to school.

Nathan doesn't give one f about any analysis of medicaid. The dude is an ideological communist.

You're the one willing to shrug off a trillion dollar annual savings and accept worse results to fit your ideology.

At what point in time did the middle class become the sole bearer of taxes for any public expenditure question?

Otherwise, is it not in vogue to emphasize how much the 1% pay as a share of total personal income taxes?

(Admittedly, the middle class has a tough deal, receiving few special benefits and not having investments of the right size or type fo benefit from various special tax benefits. And hence, the ability to believe, or easily follow, the truthiness of the statement. If your concern is about the middle class, perhaps consider the much better value for money they get in relation to health services in basically every other wealthy country on the planet.)

America's Middle Class have become little more than glorified poverty.
Living standards are falling or stagnating, drug addiction is widespread, violence is rampant, real wages are frozen, manufacture has been hollowed out by NAFTA and Communist China. Good jobs have been shipped to Asia and Americans are supposed to make a meager living by giving each other haircuts and selling each other currency swaps and apps. Healthcare and education are out of reach for the proverbial man of the streets. Americans have lost faith on their system and their leaders. America's leading i tellectuals agree America have been losing ground fast and may be collapsing into lawlessness.

"manufacture has been hollowed out by NAFTA "

The primary manufacturing industry hurt by NAFTA was making tiles and bricks. Literally what Moses liberated slaves in Egypt from is now presented as the 'solution' to the good life. Sigh.

I'd like to hear from people who harp on manufacturing as the go to place for jobs exactly how many manufacturing jobs have they held in life? Have they ever worked on an assembly line, for example.

I believe your are wrong. Bricks and tiles are still very much manufactured locally, in the USA. A cursory search brings up an industry site page that lists many brick manufacturers that exist in the US (

the LEED rating awards points if the building materials are produced closer to the building site, which incentivizes more local brick manufacturing.

That may be but they were the lead manufacturing industry hurt by NAFTA in terms of jobs.

I believe the Textile industry is usually considered the worst hurt by NAFTA.

"I’d like to hear from people who harp on manufacturing as the go to place for jobs exactly how many manufacturing jobs have they held in life?"
You are right, make American proles into trade consultants, pundits, Congressmen, Donald Trump and blog commentators. That will be the ticket to good life for Americans. In fact, I am not sure what is stopping people who cheer at America's losing its manufacture jobs (e.g. all administrarions since 1981) from applying that solution right now.

Aside from letting the "slaves" go, I am not sure how you deal with native brick-makers after you fool the Mexicans into doing your bricks for you (those fools never read the Bible and don't know manufacture will eat the product of your fields and kill your cattle and your firstborn).

How about plumbers, electricians and auto mechanics? Those are service jobs. They don't require college and they are jobs that work for people who like to feel they are doing something physical rather than sitting behind a desk. They ARE NOT min. wage jobs that offer a low standard of living. In fact they offer a pretty decent standard of living and the potential to get rich if you branch out into your own business with it.

I remember the irony of a Kurt Vonnegut's character (in Player Piano) after all manufacture jobs were automatized: of couse, I will open a workshop, we all will open workshops.

Sorry it's a myth that manufacturing jobs are nice wonderful things that provide good livings and service jobs are burger flippers and toilet cleaners.

An assembly line can be a mind killing, soul crushing job where adult men have to beg for permission to use the bathroom. This is why I keep asking those who idealize manufacturing just how much experience they really have with it? The bulk of manufacturing is not being a huge brawny coal miner or making $80K a year on an auto assembly line with union benefits. There's a lot more service jobs and no the list is a lot longer than 'trade consultant' or upper crust white collar jobs like lawyer or accountant.

Those foolish Chinese should just go back to their wonderful communes and the Mexicans should go back to their tomato picking before manufacture kills their firstborns as it did in Egypt.
"An assembly line can be a mind killing, soul crushing job where adult men have to beg for permission to use the bathroom."
As opposed to a plumber forcing his way into my bathroom whether I like it or not? I am not sure it works like that.

"the list is a lot longer than ‘trade consultant’ or upper crust white collar jobs like lawyer or accountant"
But trade consultants are the ones "who harp on manufacturing as the go to place for jobs" The others just think America needs jobs, vote Trump and hope for the best. Again, I am still expecting the plan Republicans and Democrats have to provide Americans with the jobs they need. Since the 80's Americans are selling their birthright for a mess of pottage and cheaper trinkets. It is time, while you still have a country, to make the Chinese spit out what they took instead of ordering the peasants to eat cake.

"As opposed to a plumber forcing his way into my bathroom whether I like it or not? I am not sure it works like that"

I don't recall a plumber ever coming to my house except when I've called him and agreed to pay him what is often a pretty hefty sum.

Do you agree beforehand to let him use your bathroom or does he have to beg? Or don't you hire adult men so it is OK if they have to beg?

So your concern is if you call a plumber to fix something in your house he may need to use the bathroom and you are hence in a socially awkward position of not being able to say no?

This does not sound like the concern of someone who really thinks they are living in a collapsing nation.

Relative decline of the manufacutring sector is generally considered as a primary indicator of a more advanced economy.

As industrial economies advance, the relative share of services increases and the relative share of industry decreases. Germany is an exception in this applying to a lesser extent, due to the competitiveness of its extremely high quality industrial outputs.

Some Americans have been advanced into destitution.

I have not read the source, but my first instinct is to wonder if there is data on the age/race/sex of Medicaid users and whether we have some estimates on cost by those categories. For instance, Hispanics make up 31% of medicaid. It would be interesting to see if those costs track or if some groups account for a disproportionate share of the expenses.


Because only stupid people think age/race/sex are unimportant in public policy discussions.

Would the answers change your solutions? If so, how?

You mentioned race specifically, which seems a weird way to build Medicaid policy. But would love to know what your thoughts are.


Since we have no solutions, your question is akin to asking me if I prefer to ride my unicorn or my flying carpet. If, on the other hand, you want to measure the impact of public policy, you better break it down by race/age/sex. Otherwise, you get the wrong answer. Crime is the most obvious example. In health care, we know disease rates vary significantly by population. We know age is a factor in health costs. Men have different health care costs than women.

Someone said "unicorn", and therefore all other perspectives are wrong and the argument is over.

A.k.a., "win by unicorning".

Jan asked a specific question. "And then what?" And all you had was to divert to unicorns.

@Troll Me,

I fully answered Jan's question. You should always leave open the possibility that the problem is you. Maybe you're just too dumb to understand the material and are easily distracted by certain words and phrases. That is where the evidence points.

You diverted to another topic and then made some unspecific statements about the first topic.

Not sure I get what you're driving at. Is there something you'd propose in response to a particular finding?

I see zero specifics and insults.

How little respect to you have for your fellow man to believe that anyone could see your non-argumentation as being persuasive in any manner whatsoever? Please, assume that your interlocuter has greater intellect than a five year old.

They certainly can be unimportant. You can do your study based on zip code and get all the take away you need for concentrations of wealth, poverty, health, woe.


Health outcomes are mostly genetic. That's what makes race and sex important. Health costs are mostly age related. Young people need less and old people need more.

But health professionals say that self-reported "race" is useless for health screening. They need at a minimum more detailed ethnic ancestry, but even then it only screens. It is a precursor to individual genomic screening in cases of high risk.

tl;dr: no.

In contrast, here is a clever page that demonstrates geographic correlations

It lets you flip male/female. Very similar.

Health outcomes cannot simultaneously be "mostly genetic" and "mostly age related".

You may retort that this is just a semantic distinction and you're point is that health outcomes are "in some part" related to both of those things, but the specificity is important when discussing public policy like this. I would say that's especially true in light of Alex's bizarre conclusions in the OP.


What you're telling me is "health professionals" are just as afraid of social justice warriors as everyone else. That's my default assumption. But, it does mean everyone going into hospital should have their ethnicity verified and logged as a part of their medical data. We don't get smarter by trying to know less.


Perhaps you should read my posts more carefully. Health outcomes are genetic. Health costs are age driven. Two different things.

ABM, did you even read that link? Explain this, and how it relates to health outcomes by race:

In one example that demonstrated genetic differences were not fixed along racial lines, the full genomes of James Watson and Craig Venter, two famous American scientists of European ancestry, were compared to that of a Korean scientist, Seong-Jin Kim. It turned out that Watson (who, ironically, became ostracized in the scientific community after making racist remarks) and Venter shared fewer variations in their genetic sequences than they each shared with Kim.

(The more I read MR the more I become convinced that people fixed on "race" don't care about genetics. They may say they do, but when push comes to shove, the genetics are abandoned for the old starting place.)

"That’s my default assumption."
It problably says more about you than about them.


Look up Lewontin's Paradox.

A German Shepherd looks more like a coyote than a beagle, but the shepherd and beagle are much closer genetically. Two species that are similar will often have less diversity between them than is observed within each species. When two populations diverge and eventually become separate species, they usually do so because they have adapted to different environments with different selection pressures. These selection pressures, however, do not act on the entire genome. In fact, they act only on a small set of genes—the ones that need to function differently in either environment.

Human groups have split into separate species, but they have split into races that generally follow the continents. It's why sickle cell, for example, is primarily a sub-Saharan disease. It's also why the French test their African populations and not the natives.

This is a good illustration of why we should both read more Scientific American, and less Marginal Revolution.

Medicaid covers a disproportionate percentage of younger cohorts:,%22sort%22:%22asc%22%7D

One would expect the cost of services provided to be lower than that for a cohort of persons of average age.

Also, as far as the "cost" of Medicaid, the baseline is a bit murky. Absent Medicaid, some of that medical care would be provided by providers without any compensation and the full cost shifted to insured patients. While Medicaid may transfer some of that to taxpayers (the cost on the books), due the increased usage of medical services combined with Medicaid's price controls means that there is (still) a significant element of cost shifting that is not included in the cost estimate of $532 billion.

Given the $6,400 annual cost for lower-age cohorts and inevitable cost-shifting not included in that number, it does not appear to me that Medicaid is a bargain.

Again, 30% of the budget is devoted to l/t care, not medical expenses.

I don't think by race. But we know a lot about how Medicaid dollars ate spent.

60% of long term care is paid by Medicaid.
50% of all births are paid by Medicaid.


Alex has never lived with Medicaid as his only option.

The economic freedom of not being able to access things that you do not have money to pay for.

As opposed to the economic freedom of having access to things that you require to thrive.

The poor and the rich are both equally free to sleep under bridges.

The freedom of social engineers to extort money from some people to give stuff to other people who don't seem to value the stuff nearly as much as the social engineers think that they should.

Do you support equal rights of labour unions and shareholder unions to represent their respective collective interests?

The poor would rather have the cash, but white urbanites with master's degrees know better.

Not because they "know better", but because it is difficult to prioritize health situations 10 years in advance when you're not sure yet how you're going to put food on the table in a week's time.

If providers who do uncompensated care are able to pass that cost to other customers as part of their general overhead, then Medicaid is not so much a transfer from taxpayers to providers of otherwise uncompensated care, as it is a transfer from taxpayers to private insureds, who are generally the same people albeit at different distributions, so a mildly progressive redistribution.

I'm not sure if the fact that per this study poor people don't accurately understand the value/price of the welfare provided to them is reason enough for them to not receive medical care. Sure there is a pricing problem, but that's not because of government involvement in the healtchare market. If they drop out then there is no coverage for a large portion of the population. There will never be a functioning market for health insurance, unless you are ok with 1/4 of the pop being uninsured, because that is a functioning market, where poor people can't afford to be in the market.

What if anyone who was eligible for Medicaid signed a waiver at the beginning of the year. They receive the NPV of their premiums up front in cash. In exchange they promise to pay for any received medical care. Including forfeiting the right to ER visits if they dont pay upfront.

My guess is that at least half of Medicaid recipients take this deal. So the question is why are we being so paternalistic to not let them make their own choice?

Because we don't want them to die on the street, even if they make wrong decisions?

Not to sound too condenscending but I don't think the access to healtcare is paternalistic. I'm pretty sure those that have access to Medicaid still have the option of not going to get healthcare, right? Regardless of how many people you think would take that deal (I agree with you about the amount that would take the deal by the way), this hardly leads to better health outcomes. People don't think they need insurance until they need it. And god forbid someone that takes the NPV amount has an accident that costs more than the cash subsidy received. So say someone breaks their leg and they were responsible and saved their annual premium check, that might still not be enough to cover emergency care.

Also just curious which premiums are you refering to? Private care equivalent? I thought that Medicaid didn't have premiums? So what's the benchmark premiums?

Public health policy has always been about providing minimum services to the people who cannot afford these services. One option is to let charity handle it and the other is to have a social welfare system for it. Neither is perfect. What we have in America is a vast skimming operation run by rich people against the middle class. The system is going to seize up in the next decade and we end up with a single payer format, with rich people having private hospitals and doctors.

I think the charity scenario is naive. Government is pushing out all these charities from providing healthcare to the poor? Where were the charities before government involvement? I don't have data on it, but I'm fairly certain there isn't enough charity dollars to replace Government sponsered medical programs.

People with access to the program use a lot more healthcare than other similar people The Oregon Experiment found that gaining Medicaid uniformly increased health care use: including hospitalizations (by 30 percent), emergency room use (by 40 percent), physician office visits (by 50 percent), and prescription drugs (by 15 percent). This evidence stands in contrast to the conventional wisdom that providing health insurance could reduce costs by eliminating ER visits. Of course, understanding whether this additional care is worth it requires a comparison of these real costs to the benefits provided.

Something here doesn't ring right. If they aren't going to the ER for emergencies or doctors visits for useful interventions then all Medicare has to do is say it will only pay a small reimbursement for ER visits or doctors visits that are very close to previous visits where no new diagnosis code or procedure was actually performed. Likewise prescription drugs can be analyzed and challenged if they are not for clear diagnoses or have questionable benefits.

If Medicaid is made up of relatively healthy people who don't really need coverage but use it to hound the ER whenever they have a headache or mild cold....well then there's no reason for it to cost $6400 a year. You can indulge in multiple non-necessary visits for a lot less than $6400 a year. On the other hand if they do really need more visits and more drugs that implies something more is going on.

The archetyical cost driver is someone getting a quarter million in treatment for uterine cancer. Despite the fact that there's zero evidence that chemotherapy improves uterine cancer survival rates.

Source? The 5 year survival rate for uterine cancer is 82%, even higher if it hasn't spread regionally when detected. But if it has spread distantly the rate is only 17%. The sad irony is diseases that kill fast cost less because you can't give a drug to a dead patient. The money comes when they find things that extend survival. indicates carboplatium and avastin, not new drugs, had 40% of women progression free at 6 months. So it certainly doesn't seem sensible for you to tell your mom if she gets uterine cancer she's better off doing nothing and letting it take her.

Which leads to why would huge spending come for uterine cancer if there's no benefit? Carbo is pretty mild as far as chemotherapies go but a person still wouldn't opt to have it for fun. Absent money if therapies really had no impact patients wouldn't take them and the costs associated would be lower.

Amphetamines and opoids are a cash income for anyone who can get a prescription. Of course, you don't know any actual poor people, but you certainly know what's best for them.

In that case why is Medicaid coverage important? Clearly you can make profit by selling your pills on the street and still have enough cash to pay doctor and pharmacy in cash rather than coverage.

So how is this supposed to work?

A recurrent patient comes to my ED. I suspect, but do not know, that his "chest pain" is heartburn. Should I tell him that he was here last month so he should not get a troponin screen? Or that the EKG we did then is good enough for today? If he comes in with "the worst headache of his life" do I only give him a head CT the first time or do I burn a few hundred bucks every visit making sure he does not have a subarachnoid bleed?

How exactly should I reduce the cost of his care? If you are paying me less I should also be doing less. Most of the ways I "do less" involve accepting a risk that the patient will be harmed. I am already exercising my professional judgment about are these test needed for his health, so how should I triage for cost here? If you are on Medicaid should double my risk thresholds before administering interventions?

This really makes no sense to me. We are already paying for ED visits by components and have ICD codes coming out our ears.

And seriously, CODING is your magic bullet for cost containment? Do you know how blindingly easy it is to up code - "earwax impaction" becomes "sudden left sided hearing loss", or heck you make subtle changes to the more than 50 ICD codes for ear infections and it can be years before you repeat. Do you have any idea how much the chart reviews to ensure coding accuracy would cost? And God help you if you end up in court, ICD codes are notoriously hard to pin down as improper - good luck convincing a jury that the government and not the ED doc should code an illness.

Maybe their capacity to pay is less than the benefits?

Maybe they could try an experiment where they were proposed to be hypothetically earning twice as much income, and then asked how much they would pay.

Personally, on 10 or 20k a year income, I doubt I would be willing to pay out of pocket $6k for health insurance, considering that it would leave me without a roof over my head and hungry most of the time.

"Benefits are valued at only one-fifth the cost! Why so low?Benefits are valued at only one-fifth the cost! Why so low?"

What's missing was asking them to estimate how much they think they would be spending if:

A. They paid out of pocket for their medical costs and prescriptions.
B. They purchased a private insurance policy that covered the things Medicaid covered.

I suspect you would find many of them wildly underestimate the actual costs they would face. Hence they underestimate the 'value' provided by coverage.

If I walked around thinking an auto body shop could fix up my car for $500 after a serious crash, I too might dramatically undervalue the worth of having collision coverage.

On a 10 or 20k income you would qualify for housing and ebt you ignoramus.

Yea ok and the waiting list on low income housing is 2-3 days.

Like Cyrus, I believe that Medicaid is mostly a transfer from taxpayers to private insureds. But I don't see that as progressive; being a net taxpayer begins at a lower income level than (typically) having private insurance when there isn't any mandate to have it.

I suggest that a more efficient way to achieve the same transfer would be simply to institute a general tax that subsidizes emergency room service providers, and let Medicaid recipients go back to getting their services free from the ER. This would have the added benefit that semi-poor people who go uninsured but pay for services out of pocket would no longer be subsidizing the welfare recipients by paying the service providers higher prices.

So emergency wards have to deal with even larger numbers of the poor rather than focus on actual emergencies? Also, ERs are a very expensive way to provide primary care. In summary, this proposal achieves worse care at a higher price.
How about the opposite: Have Medicaid cover all poor, ensuring everyone has access to at least basic healthcare, with deductibles and co-pays increasing with income to prevent overuse and implement a graceful degradation into one of the other insurance systems.

"Also, ERs are a very expensive way to provide primary care"

Why? They are open 24 hours, always have at least one doc there. Often have a lot of downtime.

You're confusing a high bill with high cost.

If things were that simple we should provide all health care in ERs for optimum staff utilization. Handling all poor patients alone would require additional staff working odd hours. Do you really think that the factor 3--4 between an ER bill and urgent care has nothing to do with actual cost?

It's important to note that the sentence "But it [the Oregon study] did find large improvements in mental health and self-reported health" as written is false. Those self reported improvements were right at the boundary of statistical significance, and 2/3rds of the reported improvement occurred after enrollment but before any Medicaid funded care had been provided.

That's not how to interpret statistical significance thresholds, which reject results which are deemed 94% likely to be true. 94% is much greater than 50/50, for example.

And that's not how to interpret statistical significance either - a p-value of 0.06 isn't a "result... deemed 94% likely to be true." Even more so if one hasn't corrected for multiple hypothesis testing.

Well ... yeah, but you don't reject something at p = 0.06 because it's untrue, you reject it because you're cautious.

Aaron Carroll and Austin Frakt have pointed out that the Oregon Medicaid study was underpowered:

That means the sample size was small enough that a hypothesized real outcome would be classified as statistically insignificant or "at the boundary of statistical significance" a non-trivial percentage of the time.

Re: 70% figure, it's a given that the most healthy are going to value insurance less. How much of that 70% are people who are generally healthy and don't feel they "need" insurance for that reason?

They also can sign up after they get sick. So they really don't "need" Medicaid.

Same thing with the ACA: you can game the mandate, and find a way to sign up after you get cancer pretty easily.

If they can sign up when they get sick then they still need Medicaid to be available as a thing one can sign up for when one gets sick. They just don't "need" to be on it at all times.

You can sign on to a bronze plan, pay the $7,000 deductible, and let the insurer pay for everything else.

Alex - have you read yesterday's piece "The Conservative Case for Medicaid"? I'd be interested in hearing your thoughts on it.

My sister is an urgent care specialist. She says that her patients fall into four groups in terms of price-sensitivity. Group 1 has excellent insurance; price plays little role in their medical decisions. Group 2 has lousy insurance; they may ask how much a test will cost. Group 3 has no insurance; they don't use urgent care unless absolutely necessary. Group 4 has Medicaid, and they are "more Group 1 than Group 1".

The example she gave was of entire families coming to urgent care together. Most patients come singly. When two members of a family both come for treatment, that's called a twofer. But at least once a week they see a four-fer or a five-fer. One of the kids will have symptoms consistent with, say, strep throat. A couple other family members have some sniffles, and the rest don't show symptoms but Mom feels they all ought to be checked because they've been exposed. According to my sister, this is exclusively a Medicaid behavior. Nobody with even a nominal co-pay brings asymptomatic kids to urgent care.

She was quick to add that she doesn't think they're intentionally milking the system. Getting to urgent care may be logistically difficult and cost work hours, so they're being proactive about trying to avoid future trips. But the result is that Medicaid is getting billed for 4-5 visits where only 1 was medically justified.

I support Medicaid in general as an efficient way to provide care for poor and often medically needy (e.g. disabled, long-term care) people, but I am totally fine with people having some nominal co-pays in the program. Make it a sliding scale even, up to $5 per visit. Or even a dollar makes a difference with prescription fills.

And you would withhold medical care if the patient couldn't pay? If not, the copay if meaningless.

That's how it'd work, yes. Except in emergencies.

I asked low income friend (I think he makes about $15k-$20k/year) 2 questions:

If without insurance if you got something like cancer they would just give you a prescription for pain killers and send you home.

1. How much would you pay for insurance, he said $90/month but not $100.

2. If you had Government provided insurance how much would give it up for, He said he would give it up for $110 but not for $100.

He is in his mid 50's. I would think to insure him costs much more than that.

So, about a cell phone bill. Obama had that right.

I recall people thinking it was outrageous that my employer went from charging $80/month to $130/month for insurance. They were almost entirely unaware that what is coming out of the paycheck is only the tip of the iceberg for what their insurance costs....those who go on COBRA are often rudely awakened.

The change that forces workers to see how much their employer is spending to subsidize their health insurance was one of the best things to come out of the PPACA. I wish they would report them differently, though, so that they show up as a deduction from the employee's paycheck. Just to ram home the point that it's money they could be receiving as salary instead of as a health insurance subsidy.

I knew A guy how declined employer health insurance because his part was $40/month and the deductible was $1,000.00.

There are certain hospitals that will treat uninsured people with cancer, but you have to call around to find which, if any, are in your area. These are usually public or non-profit hospitals. Sometimes non-charity care hospitals do have separate programs for the indigent. The kicker is that almost all of this care is funded through federal and state subsidies and programs. Also the treatment is often uncoordinated and these programs are often unable to accommodate someone who has to for example get some of the services at a different facility or see a doctor that is not part of the uncompensated care program at the hospital.

This is why we don't eliminate all social welfare programs and just have a guaranteed basic income. If you gave everyone on medicaid $6,000 a year and said "would you like to buy a policy?" you wouldn't sell a hell of a lot of policies. We don't give them stuff to make them happy, we give them stuff to make us happy.

+1 Yes, a UBI or wage subsidy of about $200/week for the bottom earners.

Then, because I think medical care could cost less that half of what it now costs, get the states to focus on that.

People in Utah (arguably the least corrupt state) already spend just $5,982/year to Massachusetts' $10,559/year, so it seems doable.,%22sort%22:%22asc%22%7D

I wonder whether if -- factored into these calculations -- there is any consideration that Medicaid is covering bills that would other wise show up in higher premiums in everybody else's plans -- perhaps even costing more because the cost shifts will be at a higher rate than Medicaid pays.

Costs don't change. If Medicaid reimburses less, the costs don't go down.

Sounds like an argument for actual healthcare reform - nothing to do with insurance or shuffling around costs.

Easiest solutions would be to fund cost-effectiveness research at the national level, bundle all payments to docs or provider groups and allow a lot more highly-qualified physicians to immigrate to the US.

If Medicaid pays providers less those otherwise would be higher charges wont end up in everyone else's premiums.

For those who are skeptical of the evaluation, saying people just don't have that money to spend, would you claim that if we handed the Medicaid people $6,000 that they would definitely spend all of that on the great, great Medicaid program or would they spend it somewhere else?

We all know the vast majority wouldn't spend it all on Medicaid.

Why would they? If your car got smashed up tomorrow and you had no insurance and no money, that's too bad. If you have a heart attack in the street and fall into a coma, you are going to the hospital. If you want to give them $6K and make buying Medicaid optional you also got to say actual health care will be denied for those who opt for the $6K as cash.

The fact that medicaid beneficiaries do not on average value the program at its cost does not imply that society as a whole does not value the program at its cost.

We don't like to see people suffer, even if it that suffering were due to ignorant naivety about the possibility of getting ill. And, we value that more than giving such a person $6K/year to spend as they wish, even if that is what they prefer.

Exactly. But it's more about detached wisdom than knowledge. Above all poverty and risk tolerance are closely related. Also helps explain why the poor smoke more, put low importance on occupational safety, play a lot of lotteries etc

I have never posted on MR, and its one of my favorite blogs out there. But Alex, cmon, these aren't well researched sources. Even Vox has covered this more extensively []; I always expect more from MR.

Bottom line, even if Medicaid isn't the answer, I'm of the opinion we should come up with *some answer* for "poor people who get really sick" that isn't merely "no money = you die". I'm well aware that some disagree on this point.

How much money per patient? The standard progressive answer is "all of it".

@Thomas: I don't see how your comment helps. You can probably convince a lot of folks that a bare minimum, low cost single-payer option is the optimal near-term solution. I'm referring to low-cost medical practices from the last generation available in developing nations, with mostly nurse practitioners, importing doctors from developing nations, etc. That's certainly the direction I think it's headed.

@buddyglass: It isn't the case that "no money = you die". Currently, you get healthcare in emergency situations even without the ability to pay. It could be the case that if you are poor, you can't go to a doctor that catches rare cancers early enough to save your life, or you get a condition that is so prohibitively expensive to treat that only a billionaire could keep you alive for a few more years. But them's the breaks.

Seems like inability afford regular checkups that would catch conditions that can be correct before they become "emergencies" implies "no money = you die". I'm also not cool with forcing parents to wait until their kids' fevers are high enough that they start having seizures before taking them to the emergency room because they can't afford to "non-emergency" (read: non-free) care. Even if we're not going to cover poor adults, state-provided (or subsidies to purchase private) coverage for minors, non-means-tested, seems somewhat reasonable.

I'm agree with most of this. A great many things get treated at the ED - "drug seeking" is a popular term because folks show up complaining of pain when they just want vicodin/oxycodone. Part of the cost reduction argument of the ACA was to reduce ED usage by inducing people to get routine care/screening.

But that does nothing about the actual costs of healthcare (reshuffling of costs/incentives may make it worse). If we can't have a market in healthcare where costs/benefits are transparently evaluated, then second best may be budget public clinics with strict cost controls (free third world health care paid for by taxpayers at third world prices). That will inevitably be less pleasant, of lower quality, with more complications.

You should be careful with "no money = you die". Perhaps there are frequent tests that are very expensive ($10,000 each time) but improve chances of catching cancer. Mandating that everyone gets it is clearly suboptimal. But allowing the rich to pay for it will 1) bring the technology to market and 2) allow broader populations to reap the benefits in the future. Standard profit motive arguments. Improvements in health care is neither a right nor free.

Given a choice to pay $6,400 a year for Medicare like insurance, I would in a heartbeat. Currently my employer and I pay much more than that - and I haven't been to the doctor in a long time.

Sounds like an argument for a public option for me! I'd buy in if it meant I got to keep some of the difference.

Ditto. Just realize that if the costs balance, the public option is likely to be unpleasant. Lower skill doctors, longer wait times, more liability waivers, no cutting edge techniques/technology.

Maybe it's addressed in the link, but humans are notoriously terrible at probabilities and risk evaluation. And there are huge informational asymmetries in health care as a whole.

I don't think you can really talk about how much enrollees value coverage without addressing the fact that enrollees are practically certain to be wrong about their level of exposure to risk without coverage or their inability to have a good understanding about the real costs of care.

Medicaid is simply too many things to talk about as a single program. And it is not just that each state has its own program. Medicare serves many diverse populations with a variety of benefits that are not all medical. There are over 50 eligibility categories (see: ) including some that are not necessarily "poor" such as disabled children, and it covers housing and education-related benefits (see ) that to think of it simply as health insurance misses quite a lot of the bigger picture. The devil is in the details here. Examples of questions that would promote a more interesting debate would be "Should the Early and Periodic Screening, Diagnostic, and Treatment Services (EPSDT) have copays?" or "Are the administrative costs associated with CHIP and Medicaid managed for efficiency?" or "Is there excessive overhead associated with the administration of the nursing facility services benefit?" or "Which states' home and community-based services waivers (HCBS Waivers) program have the better outcomes?" Just because Medicaid as health insurance may not be appreciated by many beneficiaries simply because of the lack of access to participating providers, does not mean that the you can dump the whole program and not see important ramifications elsewhere.

On any given issue Alex Tabarrok always starts with his libertarian priors and then assembles arguments to support that viewpoint. I suspect he doesn’t even realize how much his philosophical view filters the information he absorbs. His ideologically rigid libertarian convictions makes much of his writing pure derp.

He's not always off, but it's close to a broken clock at this point. I don't think he passes the ideological turing test.

"The average full-benefit enrollee cost about $6,400 per year to cover in 2014."

Hang on a second! Many people in nursing homes are on Medicaid so they are part of that $6400 per year per person figure. If you expand Medicaid beyond those who are in nursing homes you are almost certainly taking on lower cost patients.

Re: The average full-benefit enrollee cost about $6,400 per year to cover in 2014.

That is not too far afield of what it costs for a year's health insurance from private sources (The average individual monthly premium was $534 last year). And bear in mind that Medicaid has low or zero deductible and copays, and as such pays a larger fraction of the bill that almost any market plan or even Medicare does now. Shall we also insist that health insurance on the private market is "not worth the cost"?

Indeed, however I think you need to consider the elderly population in Medicaid. Medicaid will pay the elderly person's Medicare premium (maybe $60/month) and then only worry about paying for the stuff Medicare *doesn't*. So on one hand Medicaid's costs might be inflated by those in nursing homes and underinflated by offloading a lot of cost to Medicare.

In expanding Medicaid then, you are probably going to have healthier populations who aren't in nursing homes so that's a positive (a big one) but most will not have Medicare so Medicaid would have to cover those costs.

For that matter, most people get back far less than they pay into any form of insurance-- health, life, disability, auto, home etc. This of course suggests that insurance in general is "not worth the cost".

"Second, many of the benefits of Medicaid go to medical providers who would otherwise provide uncompensated or unpaid care to the same people."

This part should be explained better than what has been stated so far. There should be a discussion of effects on the supply of medical providers and how costs are distributed among other sources of income; at least an acknowledgement of how such considerations may change the math. As written they're basically claiming Medicaid recipients would receive a free lunch in the absence of Medicaid, and the commentary on that quote doesn't correct this issue. Where is the money coming from in the absence of Medicaid?


The obvious answer is that the rest of us pay unnecessarily high costs (both pecuniary and economic) to make up for it. ED's are money losers for almost every hospital.

I think the right answer is not necessarily medicaid, but govt provider option with even more cost minimization.

It would be more precise for the headline to say "Medicaid recipients would be better off if they were instead given the cash value of their Medicaid in lieu of health coverage." 'Medicaid isn't worth the cost" is a shocking headline that suggests that things would be better if Medicaid were eliminated. But the evidence (if you believe it or interpret it this way) suggests that Medicaid should be replaced with cash transfers. That is a very different (and much more preogressive) policy than eliminating Medicaid altogether. Neglecting, or at least failing to emphasize, this point is dangerous.

Imagine if everyone was given $6400 a year instead of Medicaid *but* those really sick, in nursing homes etc. would not be kicked out on the street.

Of course most people would be better off. If you are healthy you get $6400 more in that year. If you aren't you are either going to die (if we kick you out) or you'll get care anyway and $6400. Because of the 80-20 rule, $6400 in cash will work for most people at the expense of the minority who get really, really sick.

But just because 20% of the sickest make 80% of the costs doesn't mean 80% of us would win. Each year we take another shot at falling in the unlucky 20% so a majority could still end up harmed by getting $6400 rather than Medicaid.

I find it pretty funny that we still have a problem determining even IF healthcare access leads to better health outcomes on average.

If nothing else it's a big jobs program right?

It's probably positive but the effect is small and difficult to measure. I buy that conventional measures (blood pressure, rates of disease) aren't improved by medicaid. This is often driven by lifestyle choices.

In true emergencies (the largest cliff in outcomes if denied access to healthcare), people have access. It doesn't necessarily have to be life-saving. If you've broken a bone, the ED will treat you. Will probably give you oxy that you can later re-sell, lol.

I wonder if there is research using data from other countries that could be used to answer the question. I think that the US is an outlier among developed nations in terms of health conditions related to "lifestyle choices," such that the impact of health coverage may be dwarfed by the prevalence of chronic health conditions that are much less prevalent in nations that are less auto dependent (and hence overall much more physically active) and have healthier diets. I know that Israel expanded their health insurance program to something universal in the nineties. Maybe there is some research on that? It does seem logical to me that if you have a population that is overwhelmingly fat, sedentary, and eats a poor diet, access to health coverage probably isn't going to do much to improve their health outcomes.

I fully agree!
Not a lot a doctor can do in this case and yet the focus is still on medical interventions.
Remember how controversial the NYC soda tax was? Denying poor people the right to happiness and all that. Very difficult situation to address, particularly when 30% or the employment sector relies on the status quo (and perhaps even needs growth) for their economic stability.

People are jumping all over Alex - perhaps your ideological bent leads you to quibble with title he says at the end that the program is in need of reform. He doesn't say make a policy implication other than reform.

I do disagree with Alex in that I don't think this is a full or direct subsidy to medical providers.

This is utterly not surprising.

Exercising directly saves $2,500 a year in medical costs and has far fewer barriers to use than Medicaid. Getting that takes about 2.5 hrs per week of exercise.

Figuring that at average hourly income in the US, you need to spend about $3200 in time to get it. Figuring that at minimum wage is $940. And if you are unemployed that cost should drop even more. Yet I cannot recall the last time I saw medicaid patient over the age of 24 who got 2.5 hrs of exercise a week.

And exercise is more beneficial to your health than curing cancer (literally, even if you are morbidly obese, exercising adds more life expectancy than never having cancer).

And please do not suggest that patients just do not know. I have yet to find a patient who is unaware that exercise makes them healthier. Certainly I have oodles of patients who came to me for something nasty (e.g. broken bones), got the exercise lecture (and sometimes even the colorful brochures), and then promptly come back for something else without exercising.

Exercise has far more demonstrated health benefit, costs less, and is still not done. I believe it is safe to assume that innate value of healthcare would not be that different from exercise so the value placed on one is at least indicative of the value placed on the other.

Or is it simply that people with more energy and fewer health issues find it easier to exercise and hence do so? Exercise is not something 'natural' for humans to fact it is very artificial.

As opposed to most of the jobs people do? It takes plenty of energy to take jobs that keep you on your feet all day and I am certain that these work environments are highly artificial. Certainly standing on a street corner selling drugs is highly artificial and takes far more of a toll than say walking three miles - something that is literally one of our defining natural characteristics. Or standing in line for things like Amusement Park rides, concert tickets, etc.

I get that exercise is hard, so is holding a job, and so are many recreational activities. The fact that people (rich or poor) are unwilling to make a tradeoff of exercise for an amount of time that works out to a few thousand dollars per annum in exchange for years of life expectancy and much more quality of life suggests to me that future health is HEAVILY discounted and most healthy things - exercise, insurance, diet, etc. - are worth much less to people than economists estimate.

Exercise has had mixed results when studied for weight loss. The human body conspires to return to it's desired 'set point'. Exercise for one hour when you're not normally inclined to do that sort of thing and your appetite hours later may jump causing you eat something you wouldn't normally have eaten...or if you're the type that figits, burning calories in the 'background' you aren't aware of, maybe you figit less.

Note that since the 80's we've gone through a fitness revolution with lots of regular people joining gyms, taking classes etc. Previous to that you wouldn't do that sort of thing unless you were a professional boxer or other athlete. Despite this rise in 'fitness' we've only grown fatter and more prone to diabetes, heart disease and other issues. I'm doubtful paying people to job or 'spin' is going to be an effective way to improve overall health.

"As opposed to most of the jobs people do? It takes plenty of energy to take jobs that keep you on your feet all day and I am certain that these work environments are highly artificial. Certainly standing on a street corner selling drugs is highly artificial and takes far more of a toll than say walking three miles – something that is literally one of our defining natural characteristics. Or standing in line for things like Amusement Park rides, concert tickets, etc."

Yes being on your feet all day burns calories. I guess standing in line does too, although for most of us a concert or amusement park is something we do a few times a year....not something on a weekly basis. Nonetheless the jobs we had 100 years ago or more had plenty of 'low activity time' and it is easy to find plenty of examples of people who have highly active jobs who are nonetheless not in great shape

Funny, I never mentioned weight loss anywhere, precisely because the evidence suggests that exercise does not lead to weight loss. However exercise does show effect on such things as:
1. A1C and blood sugar. AKA helping manage/prevent diabetes.
2. Cancer risk. E.g. exercise for more than an hour per week is correlated with a one third reduction in breast cancer risk.
3. T-score. AKA bone density, fracture risk, etc.
4. Blood pressure. And a host of of other cardiac function markers.

30 minutes of exercise a day is more than sufficient to achieve dramatic health improvement in the long term. Regardless of weight, BMI, or anything along that axis.

Yet I see all the time where people are able to do *prolonged* standing for things they enjoy (e.g. concerts, amusement parks, grilling) yet are not willing to walk (an activity which is much less painful) a mile a day.

Exactly how bad do you think exercise feels to an unhealthy person? How much would you say they would pay to avoid it if it was otherwise forced upon them? Unless that number is stratospherically high, it is exceedingly unlike that they actually have low discount rate on future health.

Or forget about exercise. Take something like marijuana. Assuming that is not addicting, why would people smoke it? It has known health associations with pyschosis and schizophrenia, it is known to be associated with weight gain, and it is an increased risk of lung cancer. Marijuana is bad for your health (though less bad than cigarettes). Yet massive numbers of people smoke it. Obviously they value the high, the comraderie, etc. more than the future health loss.

This is again consistent with the idea that people have very high discount rates for future health value. And this fits for all manner activities: risky sexual behaviors, seat belt use, extreme sports, most fast food, not eating vegetables/fruits/fiber, etc. People do all manner of health sabotaging things that are generally known (and almost always known by my patients) to be unhealthy. The enjoyment people get from this sort of thing has to be astronomically better than the next best alternative to them or they have to value future health with heavy discounting.

Good info, good argument!

"It doesn’t follow from anything that has been said that Medicaid should be eliminated or even cut back (let alone that current efforts are the best way to do this). INevertheless, if I told you that Program X costs $5 for every $1 in value transferred to recipients you would probably agree that Program X was in need of reform."
Is there some way the author could rephrase this in English?

When surveyed about how much they would be willing to pay for services that they already receive at zero cost to them, the amount that they indicated being willing to pay was 5 times smaller than its actual cost.

if I told you that Program X costs $5 for every $1 in value transferred to recipients you would probably agree that Program X was in need of reform.

Or maybe your measurements are terrible.

The Oregon Experiment did not find statistically significant evidence of improvements in physical health measures, such as blood pressure and blood sugar after two years of coverage.

It turns out that when you take a bunch of people whose blood pressure is just fine, and average them in with some whose BP is not fine, and needs to be treated, you find out that treating those who need treatment produces no significant change fro the entire group. So what? That's a stupid way to measure it.

Since when does ones wealth value more than ones health? Now I understand how the money value of Medicaid might be too high but the demand for Medicaid and health is there. A lot of the people who use Medicaid's benefits would not be getting the medical attention they need without. You can argue that Medicaid is worth the money cost but understand that you cannot take away the medical attention that these people need away from them. That is murder. The cost of Medicaid may seem to high to me or you but understand from a low income perspective that those benefits are invaluable. Medicaid covers more than just healthcare but the argument that people need to pay for their own healthcare does not work because some people simply don't have the health to work. Also Medicaid will pay its self off in time. As more people receive its benefits, families will have less issues and more opportunities. Low income family sizes will decrease do to the fact of a higher rate of survival for kids. Those families will see more opportunities in life than they would have without medicaid. Those opportunities can help get them out of poverty and once they are out of poverty and able to stand on their own feet, they will no longer use medicaid but rather help pay for medicaid through taxes. In the long run, medicaid can decrease the population growth as well as help families out of poverty. This in the long run will lower the cost of medicaid while also raising tax revenue. Our generation may not taste the benefits of the fruit but generations down the line will.

Additionally, I believe gyms should be included in Medicaid as gyms relate directly to ones health. Gyms are also a boat load cheaper than Healthcare.

All these studies of "value delivered to recipients" tell us is that the Medicaid recipients would probably prefer to be given cash then Medicaid. Also one has to make sure the population being surveyed matches the population from which the expenditures are measured. This is critical because a huge part of Medicaid goes to treat elderly and disabled and children--many who are in no state to make decisions on their medical care.

But who cares, it is the taxpayers money and we would rather not watch people suffer due to being unable to obtain care. We prefer preventing this outcome to giving the poorest more discretionary spending.

The part of this that I found most interesting was that The Oregon Experiment found that those with Medicaid visited hospitals and Doctors more. This apparently contrasts conventional wisdom, which apparently dictates that the opposite should happen. However, this actually makes perfect sense to me. People without health insurance are going to try to avoid hospitals, and doctors as they don’t want, or can’t afford the bill because of low finance condition that should be keep in check as suggested here . When they have health insurance, and they know that the can be treated, they’re seeking medical help. Some countries have a system where those who can afford to pay for their medical care do so, either privately or through health insurance, and those on welfare receive completely free medical care. The waiting lists are longer for those receiving free medical care, but overall, it could be a better system than the US or British systems.

Comments for this post are closed