Will the 'Young Invincibles' Join Obamacare?
It's no exaggeration to say that if young people don't show up to the insurance exchanges in the next few months, the Affordable Care Act probably won't survive. Young, healthy people paying more than they have previously -- either because they are buying more expensive insurance or because they are buying insurance for the first time -- are the financing mechanism that makes Obamacare's insurance markets work. The administration estimates that a little over a third of the people on the exchanges need to be in the 18-to-35 range to hold premiums down to reasonable levels.
During the run-up to its passage, experts frequently talked about the "three-legged stool" of the design -- guaranteed issue, community rating and the individual mandate. Take away one of those legs and the whole thing collapses, as the market for individual insurance basically has in New York state, which passed a law saying that insurers had to offer a policy to anyone who wanted to buy one, without regard to age or health. As premiums went up, the young and healthy people dropped out, leaving an older and sicker population in the insurance market, which meant that the average cost of health care for each customer went up, which meant that premiums went up, which meant that people who weren't that old or sick dropped their insurance, leaving an even older and sicker population in the insurance market.
It's called the insurance death spiral, and it's what the individual mandate is supposed to stop.
So how do we know whether the so-called "young invincibles" will show up in the next few months to buy insurance? The short answer is that we won't, until January. The long answer is to look at how much their insurance will cost versus how much benefit they'll get. This answer is not perfect, because people are not perfectly rational value maximizers. But neither are they perfectly irrational hyperbolic discounters, so looking at the numbers and seeing how they shake out is useful -- how much will young people be paying, and how much will that save them?
Warning: This post is about to get deeply wonky. If you'd rather not look at math, even basically easy math, the time to leave is now.
OK, the rest of us can get cracking on the arithmetic. Let's start with cost, which is relatively easy: We'll look at the premium data that are starting to come out. Here's the Wall Street Journal's list of the 36 states that are having the federal government run their exchanges. It shows the lowest-cost plan, or "bronze" level, for a 27-year-old single person. It looks like the median will be somewhere between $166 and $170 a month (not including subsidies). Call it $2,000 a year for the average young invincible.
We also have some data on the distribution of claims. According to the federal government, the top 50 percent of health spenders in the 18-to-44 age group had a mean expenditure of $4,804 a year. (This includes payments from third parties like Medicaid and private insurance.) So if you go uninsured, you have a 50 percent shot of spending $4,804 and a 50 percent shot of spending less than that. The expected value of your expenditures is somewhere north of $2,800, which makes $2,000 seem like a bargain -- especially when you factor in the penalty, which can cost those with incomes of more than $50,000 upward of $1,000.
But not so fast: That $2,000 doesn't cover everything. In fact, it only covers 65 percent of the "actuarial value" of your claims, which is to say that the rules are set up so that on average, you will pay a third of your bills and the insurer will pay two-thirds. Which means that the expected value of your insurance may well be negative -- the actuarial value of covering $2,800 of medical costs is that insurance will pay a little over $1,800 and you'll pay a little over $900.
And it's especially likely to be negative for young people. Folks in the 18-to-29 demographic (the usual age group cited as the "young invincibles") are the least likely to have a claim. Big claims are very skewed toward the oldest members of that cohort. How skewed? Let's look at the top 10 percent of spenders -- folks whose health claims have an average value of about $27,000. Just 2.9 percent of that group are in the 18-to-29 age bracket, while about 3 times as many are in the 30-to-44 group. So that mean expenditure is likely to be skewed toward people who are older and sicker than you.
If you dig into the health-expenditure tables, it looks like less than 15 percent of people in the 18-to-29 age bracket can expect to spend more than $3,000 on health care in a typical year. And that includes services such as vision and dental, which aren't covered by most insurance. For the 85 percent majority, their insurance will cost them much more than they get in services.
Doing some somewhat ham-fisted calculations based on the table above, I get an expected value of total health expenditures (again, including vision and dental) of about $2,000. But their expected spend, if they buy the low-cost bronze plan, seems to be about $3,000 -- or $2,000 for the insurance and $1,000 for their expected out-of-pocket expenditures. That's because they're subsidizing old folks.
And that's just the actuarial value for a normal young adult -- the predicted expenditure averaged over everyone. The structure that produces that actuarial value means some people will bear much more. I just priced out insurance on the Washington, D.C., exchange for the cheapest bronze policy for myself and my husband. The good news is that we can buy insurance for as little as $300 a month, even though I just turned 40. The bad news is that the deductible on that insurance is $12,000 for a family and $6,000 apiece, with almost nothing covered until you hit the deductible.
Of course, most years, my homeowners' insurance pays me nothing. I pay them $750 a year because if my house burned down or fell in an earthquake, I wouldn't be able to front the several hundred thousand dollars required to replace it. The value of insurance, as I have written before, comes from protecting against catastrophe, not paying for ordinary expenses. It is worth paying a good sum to make sure that you do not end up with a useless plot of land and no home on it.
On the other hand, one isn't willing to pay literally any amount to shield against catastrophe; if my homeowners' insurance cost me $20,000 a year, I might have to rethink it, since I'd probably pay the value of my home many times over before I lodged a catastrophic claim. Most people probably think of $300 a month as rather a lot to pay for what is basically quite catastrophic coverage.
That's the cost side. What about the benefit side? What is the expected value of those catastrophic events? By which I mean, if I look at the likely expense of a catastrophic health-care cost and multiply that by the probability of that kind of event, then what benefit can I expect to get from buying this insurance each year?
Well, that kind of depends on what you think of as "catastrophic." Defining a health-care expenditure above $10,000 as a catastrophic event -- one that would be extremely difficult for someone on modest wages to pay off even on a lengthy payment plan -- someone in this age cohort has a 3.8 percent chance of having this happen. If you define it as higher than $50,000 -- a bill that most people would have great difficulty paying even with great sacrifice -- the odds are just 0.1 percent.
Now again with some more ham-fisted math. I took the table above and assumed that all the charges in each basket occurred at the exact midpoint between the top and bottom of the range. This does not produce very exact estimates, because in real life, the charges are not evenly and symmetrically distributed. Instead, they're weighted toward the bottom of the range -- basically, it's very rare to spend a lot of money on health care, and the more you spend, the rarer it is. That means that my method is off: It somewhat overstates the value of the insurance.
Using this method, I get an expected value for catastrophic events of about $880. Which is why you can buy a catastrophic policy for less than $1,000 -- policies that our "young invincibles" among the uninsured, the very people we are hoping will sign up to make the market affordable, are currently declining to buy.
I'm not saying that young invincibles are going to head into the Medical Expenditure Panel Surveys and do the math; they're not. The risk is that they'll do the intuitive equivalent: look around and notice that almost no one they know has had a catastrophic health event with a huge bill. Remember, our calculation method actually overstates the expected value of catastrophic coverage; it's probably at least somewhat lower. Which most of them can see in the fact that none of their friends has gotten leukemia.
Most of them could already be insuring against truly catastrophic risks, the kind you can't afford to bear even with a really aggressive savings plan, for $1,000 a year or less. They haven't, because they don't feel they need it as much as a mobile phone and a car.
Of course, that doesn't factor in two parts of the law that are designed to encourage people to buy: the mandate penalty if you don't purchase insurance and the subsidies, which could considerably cut their cost.
I don't think the mandate penalty is going to make much difference: There are multiple exemptions, and it's $95 in the first year. And it never will get high enough to offset the fact that young invincibles are being asked to pay more for insurance than most of them can expect to get out of it in the next decade or so.
Of course, laws require people to buy insurance for cars, and mostly they do -- so why wouldn't that carry over? That's a good question. But health insurance probably won't work like car insurance, for a few reasons.
The first is that a shocking number of people actually drive without car insurance -- in 2009, the numbers ranged from 4.5 percent of drivers in Massachusetts to 28 percent of drivers in Mississippi. And who is most likely to drive without insurance? The same people who go without health insurance: poorer people for whom it's too big a portion of their budget.
And yet, the state has very effective tools for making you buy insurance. If you don't have current insurance, you can't register your car or renew your plates. If you don't register your car or renew your plates, cops will pull you over and then ask you for your insurance card. Then they will issue you a huge ticket for driving with outdated plates plus no insurance -- the penalties for getting caught once can be higher than the mandate penalty if you make less than $40,000 a year, as more than 90 percent of the young uninsured do.
The mandate is not only less expensive than getting caught driving without insurance, but also harder to enforce. The Internal Revenue Service cannot use any of its usual tricks to collect: garnishing paychecks or seizing bank accounts. The only thing they can do is take it out of monies you are due from the federal government -- say, your tax refund. As long as you arrange not to have a big tax refund, you don't have to pay it.
So I don't think the mandate is going to have all that much effect -- except possibly the psychological one: "It's the law." (Don't underestimate that effect, either; it can be quite powerful.) On the other hand, the subsidies could make a substantial difference. If you're relatively close to the poverty line -- say, making less than $15,000 a year -- you're going to get a very large subsidy. Someone making $16,000 can expect to pay less than $50 a month.
But according to MEPS, about 50 percent of the single uninsured people in this age group make more than $25,000 a year -- the point at which the subsidies drop off rapidly. At $25,000 a year, you get a subsidy of more than $900. By $27,000, it's dropping toward $600. At $30,000, you get just $145. You can play with the numbers yourself using the Kaiser Family Foundation's calculator. (The premium data on this calculator are old, and premiums have come in below the projections I believe that they were using. But smaller premiums also mean smaller subsidies -- the subsidy is based on the second-lowest cost of a moderately generous plan -- so their figures make my Obamacare calculations look slightly more attractive than they will actually be when the real subsidies are applied.) What you'll see is that for young single people, the subsidies decline at not-very-high income levels. If those people don't participate, there probably won't be enough subsidized customers to keep the insurance market from going into a death spiral.
So here's the question: Are enough of those relatively affluent young invincibles going to pay $150 or more a month for insurance rather than pay their ordinary expenses out of pocket and hope they don't have a medical bill that drives them into bankruptcy? They aren't doing so now, and Obamacare is probably a slightly worse deal for them financially than the insurance they're not buying. Young invincibles seem to think that insurance should cost less than their mobile-phone contract and substantially less than their car payment.
But Obamacare is the law of the land, and maybe that will make a difference; maybe they will be lured by the prospect of more comprehensive insurance. The only way to find out is to see what happens over the next few months.
Corrects description of age group in 28th paragraph.
This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.
To contact the author on this story:
Megan McArdle at firstname.lastname@example.org