Sunday, January 31, 2016

Why we need insurance networks

Here's one of the questions I often encounter when discussing health insurance markets and single payer: Why do we need insurance networks?

It's an understandable frustration. You get sick, find a doctor you like and BAM! your insurer says it's out of network and won't pay. Narrow network plans can be particularly problematic for a variety of reasons. I've seen plenty of cases where children with rare diseases were unable to get coverage for their doctor because the only specialist in the region that handles their condition is, of course, out of network. And there's a never-ending parade of stories in the media where individuals went to an in-network hospital with an in-network doctor only to find an out-of-network bill from an un-requested "drive-by doctor" while they were unconscious under anesthesia. Wouldn't it be nice if insurers just paid all health providers like in-network providers?

Unfortunately, that just doesn't work.

To understand why, it helps to be specific about the policy we are considering. To abolish out-of-network charges, we'd need a regulation that prohibits insurers from refusing to pay for out-of-network care. Now imagine that you are a healthcare provider in this environment. Insurers are obligated to pay your bills regardless of whether they've negotiated any prices with you. What stops you from raising your prices? Patients don't pay the prices, so they have no incentive not to continue to come to you for care, and insurers legally can't prevent their enrollees from coming to you for care, because of the new regulation. As a result, the new higher price has zero impact on demand for your service. Why not raise prices further? Why not charge infinity. Legally, insurers are obligated to pay you infinity.

Ok, actually healthcare providers wouldn't charge infinity because insured patients do still have to pay a portion of their prices. For a typical private insurance plan, patients pay 10 to 15 percent of total costs, though after the deductible the share is typically smaller than that. Of course, the same single-payer advocates who want to abolish insurance networks also want to abolish co-pays, co-insurance, and deductibles. But even if we keep them, with patients on the hook for only 10 percent of providers' price hikes, providers would be able to hike prices quite substantially before they'd start losing enough patients to harm profits. This in turn means insurers would have to hike premiums quite substantially, and most people would no longer be able to afford any insurance.

Insurance networks are the end result of negotiations between insurers and healthcare providers on your behalf. Insurers insulate you from the full price of healthcare; in exchange, they ask you to only use healthcare providers who have agreed to charge them reasonable prices. In the absence of this bargaining step, providers would have little incentive to charge reasonable prices, and the insurance plan would collapse.

Medicare-for-all, or single-payer, does not solve this problem. In the US, Medicare and Medicaid do have de facto insurance networks, even though they work slightly differently. Instead of the one-on-one negotiations with providers, Medicare instead promulgates a list of maximum prices it will pay for things, and the network forms around these rates as individual health providers decide whether or not to accept Medicare patients. It is then up to Medicare patients to make sure they see only Medicare-accepting doctors. And medicaid is way more complicated than that, since most states run tons of different medicaid plans that each pay different rates for different kinds of care—I often see cases where kids have to be dropped from one medicaid plan and moved to another medicaid plan as their medical condition changes overnight in the hospital.

I definitely think the way we regulate insurance networks could be vastly improved, and the financial liability for non-consensual out-of-network care should never fall on patients (as in case where out-of-network doctor drops in while patient is unconscious). But still, unless we commit to a system where the government directly dictates the prices providers are allowed to charge—with all of the waiting lines and misallocation of resources that government price-setting typically causes—insurance networks will remain a necessary part of the health insurance system.

Wednesday, January 20, 2016

Bernie Sanders doesn't seem to know what Medicare is

In my last post I analyzed whether Bernie Sanders's claim that medicare-for-all would save the average family $5,000 a year was realistic, and ended up concluding that it was at least plausible, and that some amount of savings is likely, due to the lower administrative costs and in particular the lower rates Medicare pays to healthcare providers.

Ezra Klein has some tough criticism of Sanders's plan though. Here's the part that stands out:
Sanders calls his plan "Medicare for all." But it actually has nothing to do with Medicare. He's not simply expanding Medicare coverage to the broader population — he makes that clear when he says his plan means "no more copays, no more deductibles"; Medicare includes copays and deductibles. The list of what Sanders's plan would cover far exceeds what Medicare offers, suggesting, more or less, that pretty much everything will be covered, under all circumstances.
Klein concludes that Sanders has done a bait-and-switch, calling his plan a simple expansion of Medicare, while actually proposing something that works very differently than Medicare. I think the truth is worse than that.

Sanders commits the classic liberal fallacy of assuming that Medicare is great health insurance. He doesn't seem to realize that Medicare has copays, co-insurance, and deductibles, that it doesn't cover everything or every provider, that it still denies a portion of all claims.

Medicare is actually fairly crappy insurance. Here's a break down: Medicare Part A is insurance for hospital in-patient stays, Part B is insurance for "outpatient" care (which, confusingly, includes a lot of in-hospital care, even short stays and many kinds of surgery), while Part D covers prescription drugs. Part C is Medicare Advantage, which is an alternative to Parts A and B.

On deductibles, Medicare is pretty reasonable: $1,288 deductible for Part A and merely $166 for Part B. It's everything after the deductible that's crappy. For Part A, after the deductible your out-of-pocket costs actually grow as the severity of your condition worsens: after 60 days of hospitalization you pay $322 per day, and after 90 days you pay full cost (there are an additional 90 "lifetime reserve days" you can apply there though). Part B is actually worse: you pay 20 percent of all costs after the deductible.

This made some amount of sense when Medicare was designed. Back then doctors typically didn't even bother charging patients for outpatient care, because almost everything was done inpatient and outpatient stuff was just insignificant. But because of technological improvements, healthcare overall is shifting from inpatient to outpatient. Currently, Medicare spends about as much on outpatient care as inpatient, and that ratio is growing—CBO projects that by the end of this decade, outpatient care will exceed inpatient care. My point here is that while you could argue that hospitalizations longer than 90 days are very rare, that 20 percent on the other half of your health costs is extremely expensive. Seniors can easily find themselves owing $20,000 on a $100,000 outpatient surgery. Considering that the average social security benefit is only $14,169.60 per year, I don't know many seniors who can bear that kind of medical bill.

The simple truth is that Medicare is no longer sufficient insurance because it does not cap seniors' outpatient expenses. The program has totally failed to modernize in the face of dramatic changes to healthcare delivery. As a result, the vast majority of seniors—86 percent to be exact—purchase additional insurance to cover the rest of what Medicare won't pay. Some seniors qualify for both medicare and medicaid, so medicaid does actually cover these additional costs. But the rest, a solid majority, purchase private insurance one way or another. This includes Medicare Advantage (Part C), which is a private alternative to Medicare and generally has much better coverage, as well as employer-sponsored insurance (the most common option, typically offered as a retirement benefit), or "medigap" policies which are simply private supplemental insurance coordinated through federal exchanges.

Sanders undoubtedly has supplemental insurance as it is an employment benefit he would have received as a result of his years in the US Senate. He doesn't seem to realize that that's not medicare, and that he in fact relies on private insurers for much of his medical coverage.

Tuesday, January 19, 2016

My rating of BernieCare claims: plausible

In my previous post, I criticized Bernie Sanders for having a policy specific—always a vote-losing thing to have in an election—without actually having any of the specifics needed to defend it from attacks. As an example, I cited Clinton's charge that the plan raises taxes on the middle class, to which Sanders was only able to respond that it actually saves middle class families an unspecified "thousands of dollars."

At long last, the Sanders campaign has released some more detail on his plan. It would, he now claims, save the average family $5,000 per year after taxes.

That estimate is roughly what you'd get if you naively apply Medicare payment rates and administrative costs to the average existing private insurance plan. Medicare pays hospitals and doctors about 80 percent of what private insurance does, and has substantially lower administrative costs of 3 percent compared to 17 percent for private insurers. Average health spending is $10,000 per person per year, and the average health plan (ESI) has an actuarial value of around 85 percent. Applying the 80 percent lower Medicare prices and the 14 percent administrative savings, I get roughly $7,000 in health spending per person per year, a $3,000 savings per person (that is, after paying Sanders's payroll tax, you'd still be $3,000 ahead in savings because of the lack of private insurance premiums and lower out-of-pocket prices). So for a family of four, Sanders's $5,000 estimate actually has a bit of wiggle room—it's smaller than the most optimistic possible scenario.

However, even though Medicare pays lower prices and has substantially lower administrative costs, it is not a slam-dunk case to argue that Medicare-for-all represents a cost savings. For the full story, you also have to look at utilization. It would be nice if you could just compare average Medicare spending per beneficiary to average private spending, but obviously Medicare beneficiaries are on average quite a bit older, sicker, higher cost patients so we'd expect Medicare spending to be higher even under the null hypothesis that it is identical to private insurance. Indeed, that's exactly what we find: Medicare spends about $11,200 per beneficiary per year (including administrative and other costs), compared to the average private insurance premium of $6,010 per person per year (note: Medicare spending per beneficiary is less than the total health spending per beneficiary due to cost sharing; this figure is most comparable to insurer's revenue per beneficiary).

So the raw numbers aren't conclusive at all—they're consistent with both the claim that medicare is cheaper but current medicare enrollees are higher-cost patients (yes, they are), and the alternative claim that medicare is more expensive, because of higher unnecessary or fraudulent use, in addition to serving an older higher-cost group of patients.

There are two ways in which Medicare can still be more expensive than private insurance inspite of lower administrative costs and lower payment rates: more fraud and more "moral hazard" (by which I mean beneficiaries' non-fraudulent but voluntary choice to use more healthcare than they would with other kinds of insurance). The FBI puts the rate of fraudulent health insurance payouts at 3 to 10 percent, but does not differentiate between private insurance and Medicare (see also). Even assuming no private insurance fraud, this is simply not large enough to even offset the higher administrative costs of private insurance (It appears that private insurers' fraud prevention programs are cost-ineffective!).

Maybe there's a lot more fraud going on, in spite of the lack of evidence. Or maybe private insurers really are getting enrollees to avoid unnecessary costly care. In either case, those effects would have to be quite massive to offset the lower prices and administrative costs of Medicare. So, in the end I do believe Sanders's medicare-for-all plan would represent a net cost reduction on average—the opposite hypothesis, while not impossible by the numbers, is pretty far fetched. Moreover, I think Sanders's $5,000 figure is a reasonable (but highly uncertain) estimate of the average family savings, though it should be noted that a large share of "middle class" people with above average incomes will in fact pay more because the payroll tax is a flat tax while the health benefit is lump-sum.

Monday, December 21, 2015

Bernie's Medicare-for-all and the risk of policy specifics

I watched parts of the democratic debate on Saturday, and the part that really stood out for me was the Sanders-Clinton exchange on Sanders's Medicare-for-all plan and then his paid family leave plan.

I've written before that a big lesson of the 2008 and 2012 is that offering policy specifics always costs votes. It hardly matters how great the policy idea is—any policy change will harm someone, and even if those someones are almost no one, any policy proposal will be beaten down with an endless parade of sob stories from the few who benefited from the status quo. Elections punish policy innovators. The winning strategy, as we saw in 2008 and 2012, is to offer vague platitudes instead of policy specifics. Offer just enough to signal your supporters that you'd pull their favorite policy levers, but never commit to a specific policy in the midst of an election campaign.

This is excruciating to folks like me who really want to debate policy specifics. Unfortunately, policy specifics won't be on the ballot.

In part because Clinton has absorbed the lesson of her campaign with Obama, Sanders's Medicare-for-all is one of the most specific policy proposals so far in the 2016. It would expand Medicare to all Americans under 65, and raise the payroll tax to pay for it.

Almost immediately, critics charged Sanders with advocating a massive tax hike to pay for his new "spending plan" to expand Medicare to all. Of course these critics ignore the fact that with Medicare, everyone would be able to dump their legally-required private insurance. Under the Roberts doctrine that mandatory private insurance is a tax, the Sanders proposal could constitute a net reduction in taxes by "thousands of dollars" for the average person, to use Sanders's words.

And now we witness the destruction of Sanders's rhetorical ground. Not only has Sanders offered a policy specific which can be directly criticized, his rebuttal to that criticism is just the made-up-on-the-spot-sounding unspecified number of "thousands of dollars" that people will allegedly save under his plan.

Not only did Sanders lose votes by offering a policy specific for opponents to attack, but he's losing even more votes by not having the research to back up his claim. That's just bad staff work.

Clinton has in the past criticized Sanders's health plan as a tax hike on the middle class—she's promised not to raise taxes on families making less than $250,000 a year—but danced around it this time by pivoting to his paid family leave plan, which raises payroll taxes on everyone slightly to cover a new benefit, in contrast to her own version of the plan which raises taxes only on those earning over $250,000 a year. Because most people don't currently pay into a paid-family-leave fund, Sanders can't even claim to be offsetting an existing cost here.

Clinton's attack on Sanders is just a version of the same attack republicans will eventually pull on her. When the republican candidate repeats the charge he'll neglect to mention that her tax is only on income over $250,000. If republicans are forced to acknowledge this at all, it will suddenly become a tax on "job creators" that somehow also hurts middle class workers.

Unfortunately, in an election campaign, policy specifics are just vote-losers. It pains me to admit this.

Friday, December 4, 2015

Sex differences and probability distributions

Yesterday Twitter pointed me to this post highlighting what David P Schmitt argues are established, methodologically sound findings on the differences between the sexes, arguing that people who try to claim there are no differences are misrepresenting the evidence. Here's the established findings he cites:
d = -0.20 has been observed for sex differences in trust (Feingold, 1994). This size of sex difference is considered “small” and indicates 58% of women are higher than average man in trust (based on Cohen's U3).

d = +0.50 has been observed for sex differences in spatial rotation skills (Silverman et al., 2007). This size of sex difference is considered “moderate” and indicates 69% of men are higher than average woman in spatial rotation skills.

d = +0.80 has been observed for sex differences in physical aggression (Archer, 2004). This size of sex difference is considered “large” and indicates 79% of men are higher than average woman in physical aggression.

d = -1.00 has been observed for sex differences in tender-mindedness (Feingold, 1994). This size of sex difference indicates 84% of women are higher than average man in tender-mindedness.

d = +2.00 has been observed for sex differences in throwing distance among children (Thomas & French, 1985). This size sex difference indicates 98% of boys throw farther than the average girl.
But here's the important caveat:
As I have noted in earlier posts (see here and here), sex differences with larger d values are not “more real” than smaller-sized sex differences. All men do not have to be taller than all women for a sex difference in average height to be “real” and have important social consequences (Prentice & Miller, 1992). Sex differences with larger d values also are not necessarily more attributable to evolution or biology, and smaller sex differences are not necessarily more cultural or due to learning than larger sex differences. Nature does not work that way.
One issue I have is the way he presents the data. A statement like "69% of men are higher than average woman in spatial rotation skills" make it sound like he's saying there's a 69 percent probability that a man is better at spatial rotation than women, which would be a misinterpretation—in fact for most of these examples, the two distributions overlap more than they separate, and the headline statistic overstates the relevant probabilities.

Schmitt's post doesn't quite give us enough info to calculate the relevant probabilities, but if we assume everything is normally distributed and that the standard deviation is always the same for men and women, we can crunch some numbers. So let's do that:

"58% of women are higher than average man in trust"

The relevant probability here is this: if we randomly sample one man and one woman, what is the probability that the woman is more trusting? Crunching the numbers with our added assumptions yields a 56 percent probability that the woman is more trusting. Ok, that's not that different. Either way, about even odds. Here's what the two distributions look like graphed:
Men (red) less trusting on average than women (blue)


"69% of men are higher than average woman in spatial rotation skills"

That's a 64 percent probability that the man has better spatial rotation skills than the woman.
Men (blue) have better spactial rotation skills on average than women (red).


"79% of men are higher than average woman in physical aggression"

That's a 72 percent probability that the man is more aggressive.
Men (blue) are more aggressive on average than women (red)


"84% of women are higher than average man in tender-mindedness"

The difference between the headline figure and the relevant probabilities gets a bit more pronounced in the tails. This stat actually corresponds to a 76 percent probability that the woman is more tender-minded than the man.
Women (blue) are more tender-minded on average than men (red).


"98% of boys throw farther than the average girl"

This makes it sound like a man is all-but-certain to through farther than any woman, but in fact there's a 7 percent chance our randomly sampled woman will throw farther than our randomly sampled man. And as in the rest of the examples, this still understates the degree of overlap, as lots of women can throw farther than lots of men:
Men (blue) throw farther on average than women (red).


The statistics overstate the case for sex differences. Even the computed probabilities comparing a randomly sampled man and woman understate the degree of overlap between the two distributions, where, even though the averages differ, lots of men and women fall to the left and right of each other. And that's to say nothing of the fact that most of these differences are socially constructed rather than innate. If Schmitt's point was that men and women, as groups, are not the same as each other, then my point is this: we need to make sure we aren't making the same mistake about the people within our arbitrarily defined groups. No actual woman is identical to the average across all women. No man is the average man. With very few exceptions, the differences between the two distributions are dwarfed by the overlaps.