Bryan Caplan  

Health Insurance, Kids, and Cut-Rate Coverage

Why Bryan Caplan's CEA Veto Wo... Give One Point to Scott Sumner...
Many employer health plans cover kids for zero or near-zero marginal cost.  This anomaly came up years ago on my now-defunct Armchair Economists' Listserv, and no one had a really good explanation.  Now, as you might have heard, Obamacare has legally amplified the puzzle.  The NYT explains:
Under the rules, an employer-sponsored health plan or a company selling individual insurance policies must offer coverage to subscribers' children up to the age of 26, regardless of whether a child lives with his or her parents, attends college, is a dependent for income-tax purposes or receives financial support from the parents.


Under the rules, insurers and employers must provide young adults with a 30-day opportunity to enroll in their parents' coverage. Terms of coverage cannot vary based on the age of young adults under 26. Thus, the White House said, an insurer violates the law if it imposes a surcharge on premiums for children 19 to 25.

Umm, is any charge a surcharge, or what?  Here's what the Whitehouse says:

Same Benefits/Same Price. Any qualified young adult must be offered all of the benefit packages available to similarly situated individuals who did not lose coverage because of cessation of dependent status. The qualified individual cannot be required to pay more for coverage than those similarly situated individuals. The new policy applies only to health insurance plans that offer dependent coverage in the first place: while most insurers and employer-sponsored plans offer dependent coverage, there is no requirement to do so.

This seems like a classic case of, "If you're not confused, you don't understand what's going on."  The regulation sounds like it's forcing insurers to cover 19-25 year-old "children" at a loss.  But can insurance companies fiddle with their rates enough to avoid that?   It's hard to say.

Deeper questions: If the new regs actually make it unprofitable to cover 19-25 year-olds, how will insurers respond?  Will they stop covering all dependents?  Raise the rates on dependents up to marginal cost?  Or suck it up?  How does your theory of the original "cheap/free kid coverage" affect your answer?

HT: Daniel Lurker

P.S. Notice how like almost all health insurance regulation, the mandate for 19-25 year-olds has nothing to do with moral hazard or adverse selection, and everything to do with feel-good populism.  Adverse selection can rationalize mandatory insurance, but if the mandate fixes a genuine adverse selection problem, there's no need to cap rates.  Why not?  Because given the assumptions, mandates are supposed to make the rates fall!

Comments and Sharing

COMMENTS (12 to date)
Ted writes:

Insurers are already willing to give coverage to dependents under 18 (or whatever cut off age) at relatively low costs. This implies that they are probably very cheap to insure, which seems logical since most people under 18 require little medical care. The logic would probably similarly extend to 19-25 year olds. My age bracket is generally fairly healthy and we tend not to use particularly expensive treatments / surgeries / medications. So, if there is any net loss I suspect it would be relatively minor. So, I suspect, the insurance companies will just suck it up. They won't want to play funny games on minor losses since it might provoke political outrage causing some aggressive amendments to the bill that would undermine their bottom line much more.

On the whole though, I don't think this health care bill will effect the bottom line of the health care companies that much. Yes, it does impose some inconvenience from their perspective that minimize profit-maximization (inability to exclude pre-existing conditions, 19-25 y/o dependent status etc.), but I have to think that a substantial portion of the loss of advantages they have now will be offset by the public subsidization of millions of future customers.

As a side note, it's always confused me why insurance companies often allow 19-25 year olds to maintain dependent status if they are a college student. I suspect there is zero connection between going to college and short-term health, in fact, I suspect it would probably increase your likelihood of needing costly medical treatment.

Chris writes:

Maybe offering free insurance to kids is a way to encourage parents, who are more highly risk-averse (and hence profitable) than other adults, to select into the insurance pool.

MBP writes:

Totally agree that the language you quote is confusing. What I think it says is:

1) If an employer currently offer health plans that provide dependent coverage, that coverage must now be extended to cover dependents up to the age of 26. (currently many plans will cut off coverage at 22, i believe)

2) If a plan currently charges $x for a dependent and cuts off coverage at 22, the plan will not be allowed to charge > $x for older dependents (up to 26).

Also, you get into an interesting question: is it unprofitable to cover 19-25 year olds? There may be 2 different answers to this: it may be profitable to cover this group as dependents to their parents' policies since the incremental cost can be spread to the entire insured population and it may not be profitable to cover them as individuals (due to adverse selection and the high cost of mandates in some areas). In fact, the individual market for young people is almost non-existent in many states. I think WellPoint has been quoted as saying that one of their individual products has 1 member in the entire state of New York, because the state imposes so many mandates that drive up cost of coverage.

Lastly, as to why many employer plans do not increase premiums based on number of dependents: Good question. It's a fact that a family of 3 and a fmaily of 9 likely have the same premium if they work at the same company. Why? I've wondered if it was a way to increase the compensation of employees who are viewed as more loyal and have longer tenure at the expense of younger employees who have higher turnover, but this is not a very satisfying answer...Perhaps it's simply administratively easier to have a single "family rate" instead of trying to prove how many dependents each employee has??

David writes:

Why should the cut-off be age 25? There's no shortage of deadbeats in their 30's (and older) who could probably benefit from being on their parents' employer-provided policies. In fact, maybe we should change Medicare so that once Mom and Dad reach 65, Jr. gets Medicare coverage too! On the other hand, if the gov't insists on treating 19-25 year olds like children for medical insurance purposes, how about we revisit the case that was made for lowering the voting age? If you're on Mom and Dad's medical insurance, you can't vote. Let's raise the legal drinking age to 26 while we're at it.

Blakeney writes:

I wouldn't say that the new law has nothing to do with adverse selection - it's going to create plenty of it. The adult offspring who elect to be covered under their parents' plan will be disproportionately those who expect to get a lot of benefit out of the coverage.

As far as Bryan's original comment about employers covering kids at zero marginal cost (after the first one, i.e. an employee-plus-child tier covering any number of kids), I think that's about to go the way of the Dodo. This traditional arrangement made some sense when the kids in question were a fairly stable and homogeneous group. That is, an actuary like me could reason that kids under 21 each cost about 50% as much as an adult, and that 40% of employees covering kids would have one child, 30% would have 2 children, etc, so that the cost for the average brood of kids was about the same as that for one adult. The important economic assumption here is that the number of covered kids is pretty much exogenous to the actuary's decision about the price of coverage. Nobody decides to have another child just because they won't have to pay anything extra for its health insurance. Adverse selection among the covered kids wasn't really a problem, because it wasn't like the healthiest 9-year olds were more likely to go out and buy their own insurance, leaving the sicker kids in Dad's plan. So overall, it's usually just been simpler to charge one rate for any number of kids. Sure, the families with only one child were subsidizing those with three, but the employees didn't really mind. They're typically paying a small enough percentage of the cost anyway that those one-child families are still getting a subsidized benefit for their kid.

Now, though, there's going to be both a higher variability in the number of kids to be covered, and adverse selection among a subgroup of those kids, which makes it much more problematic to develop a single rate to cover one or more kids. In response to this change, I know that many employers are considering a change in the way they charge their employees for child coverage - going from the old "all the kids you can eat" system to one that charges an additional fixed contribution per child. That fixed contribution will almost certainly be higher than it would have been without the mandatory coverage of adult offspring, because of the adverse selection I mentioned earlier. The end result, I think, is that instead of smaller families subsidizing larger ones, we'll effectively have younger children subsidizing older ones, and eventually we'll have more uninsured children as a result (actual children that is, not the age 22-26 ones).

Yancey Ward writes:

I was going to comment,but Blakeney covered everything I wanted to say and more.

mark writes:

In my experience, plans are basically "family" plans; there are no premium variations tied to number of members in "family". (I have never understood why as a matter of logic, but anyway). So I read the language to say it is increasing the definition of family but no increase in premiums can be imposed for that reason. Since premiums are going to rise, how you ascertain causation of the increase is a mystery. Doubtless it affords an opportunity for politicians to allege the insurers have broken the law, the political benefits from which may have been one reason for the law.

Where this is going to get really weird or ugly is in plans where the insurer is merely the administrator but the employer really pays the bills ultimately. In those, the premium is merely the amount needed to pay the bills plus the insurer's margin for processing claims. I don't know how you apply this language in such cases, except to say the insurer must cover the claims of people between 19 and 25 for free. If the insurers challenged that in court, as Bryan points out, what possible rationale could be advanced to support that policy?

Blakeney writes:

@ mark - The regulations describing how the age-26 mandate will work just came out a few days ago, and I haven't read them in detail yet, so I could be wrong about this. Anyway, here is my understanding of how things will work for self-insured, employer-sponsored plans:

First, the actual adjudication/payment of claims can't be any different for a 25-year-old dependent than it is for a 12-year-old. For example, if you have to meet a $500 deductible for your toddler's medical care, your employer can't require you to meet a $1,000 deductible for your college student's medical care. I don't know of any employers who were seriously considering such a plan design, but I think it could have been allowed under the language of the original bill. Anyway, the regulations published this week prohibit this sort of attempt to limit benefits for adult offspring. This does not mean, however, that the extra kids have to be covered "for free"; they just have to be covered at the same level as other kids.

Second, the contributions you pay for coverage (usually through a deduction from your paycheck) cannot be any different for the adult kids than for the youngsters. This means that, if you're in a plan with the traditional "employee-plus-one-or-more-kids" contribution structure, then your employer cannot charge you anything extra if one of your one-or-more kids happens to be age 24.

On the other hand, if your contributions are of the less-common "fixed amount per child" type, then the law requires that the "fixed amount" be the same regardless of the child's age or student status. Again, the law doesn't require the adult kids to be covered without any payroll contribution at all; it requires that the contribution be the same as it would be for a minor child. This is the specific portion of the regulation that caught many employers (and their consulting actuaries) by surprise.

Finally, the law doesn't require that any child be covered. If a company says "screw you guys, we're not gonna cover any of your children, no matter what their age", then that's fine under the law. It's also employee-relations suicide, so I don't expect to see many companies going this route. Except maybe the really depressed ones.

John Thacker writes:


If that were the case, why wouldn't the insurance companies be willing to do that without the force of law?

mark writes:


Thank you.

"if you're in a plan with the traditional "employee-plus-one-or-more-kids" contribution structure, then your employer cannot charge you anything extra if one of your one-or-more kids happens to be age 24."

Yes, this is the kind of plan I have (I am a partner in a firm so I am an employer for these purposes by the way). This, unless I am missing something subtle, supports my point about them being covered for free. Pre legislation, they disappeared out of the pool at whatever age used to be the rule. Now they stay in the pool. But the cost can't change. That to me means they are being covered for free.

Brian C writes:

My company Self Insures (but pays a HMO to help manage the plan). End of last year when I asked are benefit coordinator about charging for additional children she said it takes almost 4 kids to = one female. She said spouses were the majority of the cost to the plan. My company is a Manufacturing/Construction/Maintenance firm so 80% to 90% of the employees are male. Due to the administrative cost and the fact the one additional child adds relatively small incremental cost to the plan she said it really was not worth it. I assume that is why most business have plans which cover college students, they really are not more expensive than a teenager. The cost is so small that they do not see the point for a fairly small saving to tick off there employees.

I still hate the fact most Colleges require students to pay for the on campus medical service even students on a parents plan.

The real issue is why does the federal government now require parental coverage of college students when almost all colleges require you to pay for their medical services?

So assuming the bill is not repelled, there will be no more medical fee/tuition adders for free medical services at any college because all students will have insurance right? They might have an on-sight clinic but it should charge the students insurance not be covered by student fees/tuition. I wonder the chances of that actually happening on a majority of campuses in the next 20 years?

Dan Weber writes:
I suspect there is zero connection between going to college and short-term health, in fact, I suspect it would probably increase your likelihood of needing costly medical treatment.
I would expect a correlation between college enrollment and lower health-care costs.

1. College students come from higher SES families, who are healthier.

2. College students are less likely to face workplace injuries.

I could see that college students are more likely to have certain health care costs, like skiing accidents. Anyway this is all my armchair expectation with no data.

Comments for this entry have been closed
Return to top