Why Roberts’ Tax Reasoning Ultimately Damns the Affordable Care Act (But Not in a Good Way)

Thom Lambert —  11 July 2012

There’s great irony in Chief Justice Roberts’ reasoning in the recent Affordable Care Act ruling.  In reading the ACA to impose a tax for failure to carry health insurance, thereby assuring the Act’s constitutionality, Justice Roberts also doomed the Act to failure.  Let me explain.

As the government repeatedly stressed, the individual mandate (now interpreted as a disjunctive order either to carry health insurance or to pay a “tax”) is necessary because of two “popular” provisions of the ACA: guaranteed issue (i.e., insurance companies are not allowed to deny or drop coverage because of preexisting conditions) and community rating (i.e., insurance companies must set common rates and can’t charge higher premiums to sick people or those susceptible to sickness). Taken together, those two provisions create a terribly perverse incentive for young, healthy people:  Don’t buy health insurance until you get sick!  After all, you can always sign up immediately upon becoming ill or injured (thanks to guaranteed issue), and (thanks to community rating) the insurer can’t charge you a higher price reflective of the greater likelihood — certainty, really — that you’ll make big claims.  To prevent young, healthy people from dropping their insurance, thereby leaving only the older and infirm in the pool of premium-paying insureds, the law must create incentives for them to buy insurance.  The penalty-backed individual mandate was ostensibly designed to do so.

But there’s a problem: penalties don’t deter if they’re set too low.  If a parking meter costs a dollar, but the penalty for not feeding the meter is only a quarter, who’s going to feed the meter?  Unless the expected penalty for an expired meter (the fine times the likelihood of detection) exceeds a buck, feeding the meter’s irrational.

What does this have to do with the ACA?  Well, the statutory penalty for not carrying health insurance is really low — way lower than the cost of insurance.  As Justice Roberts observed:

[I]ndividuals making $35,000 a year are expected to owe the IRS about $60 for any month in which they do not have health insurance. Someone with an annual income of $100,000 a year would likely owe about $200. The price of a qualifying insurance policy is projected to be around $400 per month.

So what is the young man or woman, fresh out of college and beginning a career, going to do — pay the $400/month or pay $60/month until he or she gets sick, at which point he/she can call up the insurance company and be assured of coverage (guaranteed issue) at rates not reflecting his/her impaired health (community rating)?  Surely a great many young people will take the latter tack, especially since — as Justice Roberts repeatedly emphasized — they’re not acting “unlawfully” in doing so.

Then we’ve got real problems.  Health insurance premiums are based on the likely health care expenditures of the pool of insureds.  The greater the percentage of young and healthy (low expenditure) folks in the pool, the lower the premiums.  Conversely, when the young and healthy drop out so that the pool of insureds is older and more infirm, premiums will rise.  And, of course, the higher insurance premiums rise, the more sensible it becomes for the relatively healthy to drop their insurance, pay the small “tax” instead, and wait to get sick before signing up for increasingly costly coverage.  It’s a pernicious cycle.

None of this is rocket science, and proponents of the ACA certainly understood these dangers when the statute was enacted.  They likely assumed, though, that the deficient penalties for failure to carry insurance were a “bug” that Congress would eventually fix once the Act was put in place and became operative.  Proponents needed for the penalties to be low so that they could get the statute through the political process; they figured they could fix the deficiencies later.

Justice Roberts’ opinion, though, will make it very hard for Congress to raise the penalty for not carrying health insurance.  The small size of the penalty was one of three factors that, according to the Chief Justice, transformed the penalty into a tax for constitutional purposes.  He explained:

[T]he shared responsibility payment may for constitutional purposes be considered a tax, not a penalty: First, for most Americans the amount due will be far less than the price of insurance, and, by statute, it can never be more.  It may often be a reasonable financial decision to make the payment rather than purchase insurance, unlike the “prohibitory” financial punishment in Drexel Furniture. Second, the individual mandate contains no scienter requirement. Third, the payment is collected solely by the IRS through the normal means of taxation — except that the Service is not allowed to use those means most suggestive of a punitive sanction, such as criminal prosecution.

This reasoning suggests that the penalty for failure to carry health insurance can count as a tax for constitutional purposes only if it is kept so small as to be ineffective.  Justice Roberts has thus transformed what was effectively a “bug” in the ACA into a “feature” of the statute — one that is required for the Act to constitute a valid exercise of congressional power.  He has damned the ACA in the process of saving it.

But market-oriented folks shouldn’t rejoice.  It’s true that Justice Roberts’ reasoning has assured that the ACA, if not repealed, will implode. That will occur because the combination of guaranteed issue, community rating, and constitutionally required low penalties will drive young and healthy folks from the pool of insureds, causing health insurance premiums to spiral upward and inducing even more people to opt for the “tax” over costly coverage.  Congress will eventually have no choice but to restructure or repeal the Act.  But its replacement won’t be pretty.  Look out for the argument, “We tried a solution involving private insurance.  It failed.  Now our only option is a single-payer system.”

Justice Roberts’ decision may turn out to have been an even bigger gift to the left than anyone initially thought.

Thom Lambert


I am a law professor at the University of Missouri Law School. I teach antitrust law, business organizations, and contracts. My scholarship focuses on regulatory theory, with a particular emphasis on antitrust.

11 responses to Why Roberts’ Tax Reasoning Ultimately Damns the Affordable Care Act (But Not in a Good Way)



    Good point. I did leave out the effect of the subsidies, which will certainly encourage many to buy insurance rather than pay the penalty. I don’t think it will be enough, though. For example, for an individual earning $32,000 in 2016, about 237% of the federal poverty level, the penalty for failure to carry insurance would be $695. Under the subsidies, the person would be expected to pay 7.6% of income, or $2,430, on insurance. For this person, it still looks like an awfully good deal to pay the fine and forego insurance until you really need it. (BTW–I got these numbers from this post, which also highlights the many exemptions from the individual mandate: http://www.forbes.com/sites/aroy/2012/07/09/obamacares-dark-secret-the-individual-mandate-is-too-weak/.)

    Thanks for the comment.


    There is at least one major problem with this — it completely ignores the SUBSIDY portion of the law. The law employs both a carrot and a stick approach to get people to opt into the market. And while you are correct that the stick is quite low for low income people, this also where the carrot is very high (i.e. where maximum subsidies kick in). So the question is really would a young/poor person would rather (i) pay a tax of 60 dollars a month and have no health insurance (which can be valuable even for young/health folks who stay young and health) or (ii) pay 400 a month MINUS any applicable subsidies. For someone making 35,000 a year, the subsidies are projected to be very large (don’t have a precise number on hand)….


      But the subsidies decline pretty aggressively with increases in income. I can’t find the post right now, but Mankiw (using admittedly back of the envelope math) put them at something like a 90% implied marginal tax on income for certain wage ranges. So, for some the subsidies will make buying insurance a better deal, but for others they won’t. And they provide a significant disincentive to improving your salary at certain ranges.

      But the argument doesn’t need to be theoretical. The Massachusetts health care law has similar conditions, and they have this exact problem, though it’s not as strong or fast growing as Thom Lambert’s post would imply; there is a significant norm to follow the law, even if you know you can get insurance, not having insurance is scary for the risk averse, and the MA fines are significantly higher than the federal tax. Massachusetts still has the highest per capita health expenditure in the nation and their costs are among the fastest growing.

      A couple years ago I found something from a Massachusetts agency estimating the cost of people who went uninsured until they got sick, but I’m not finding any specific numbers now. The closest, arguably objective source I’ve found in a quick search was a CNNMoney article: http://money.cnn.com/2010/06/15/news/economy/massachusetts_healthcare_reform.fortune/index.htm


    Unfortunately, your basic premise is flat wrong. Escalation of the penalty is already built into the ACA, at an interval designed – along with other “features” of the legislation – to mean the end of private health insurors, so government is the only place to turn.

    As for alleged “market failure”, the real failure is the result of, and to the extent of, government regulation. For instance, Atate and (through the ACA) Federal governments regulate what coverages must be included in you insurance policy, whether you want them or not, and you cannot buy health insurance across state lines, in order to obtain coverage that better fits your needs.

    Were these barriers to actual, free-market competition not in place, I have no doubt that people could obtain insurance for catastrophic events – such as they already can for their cars – for MUCH less than is anticipated for the fines under the ACA for not carrying insurance.

    northfork investor 11 July 2012 at 3:18 pm

    1. Many qualifying high deductible plans which still pay for the basic preventative care without a deductible cost way less than $400. I have one and have found high deductible plans under $285/month consistent with ACA. I don’t know if the number Robert used was given him in the case materials from the plaintiff or defendant or his clerk googled it somewhere.

    2. Many low income healthy people will choose to pay the $720 tax/year rather than buy insurance. Those dollars ostensibly will be useful in subsidizing lower cost policies for in health exchanges. That is an improvement versus the pre-ACA days ie paying nothing, getting hit by a car and letting the emergency room pick up the cost (and eventually society picking it up).

    3. Some of those potentially non-participating persons will be carried by mumsy and popsy on their policies until they are 27. Mom and Dad may be in a large employee pool and find incremental insureds far less costly than Robert’s number (perhaps because the employer subsidizes the family add-ons.) Mom and Dad may be more risk averse than their children (after all there may be grandchildren involved already). This is also an improvement over what was the status quo in getting healthier people into the insurance pool.

    4. In analyzing this you better look at tax effects. When you start looking at non-deductible tax penalties at $150-200 a month or so, assuming a 30-36% combined state and federal tax, it might cost as much $300/month (after tax) to pay the penalty. On the presumption that health insurance is either deducible by the employee (or deductible to employer/not income to the employee) a basic high deductlble policy costing $400/month costs after tax about that same $300/month. and for that you get insurance and some “free” preventative health care. This is what is so important about the ACA structure–getting persons who can afford to be in the pool to find a policy they can afford net-net and reduce the externality of going without and offloading the risk to society.

    5. I’m self employed and know the tax treatment of my health insurance. It’s an offset to AGI putting me on equal footing with the employer who gets a deduction and the employee who doesn’t have to take the employer subsidy as income. But supposing I was quite a bit younger–say 30– just a retired young guy (a trust fund kid or Facebook winner) with no earned income just investment income. Can that guy deduct his health insurance premium? Cause if we want him in the pool we should treat his tax situation the same way.

    6. So Robert’s opinion which says the penalty is way lower than the cost of insurance is just not necessarily right. He has not taken into account that there are policies that are structured to cost lower than average and he did not take into account taxation.

    7. Hence your conclusion that the Act is damned is flawed. Health insurance marketers will devise ACA-compliant plans that will entice some (but not all) marginal buyers into the market and even the clear non-joiners will probably particpate via the penalty whereas they do not participate at all today.


    I followed you until the last paragraph. You have set out the adverse selection market failure rationale for the ACA. You then explain why the penalty might be too low to deter free riding. Why would market oriented folks even consider rejoicing over this shortcoming in the law, whether bug or feature? Aren’t market oriented folks in favor of solving free rider market failure problems with collective action as efficiently as possible?

    Putting aside the issue that the single payer “solution” may be a less efficient fix, are you also suggesting that this market failure is not worth solving? Are you suggesting that the free riding market failures from (say) the inability of manufacturers to set resale prices is worse than the free riding market failures from adverse selection in the health insurance market?



      The reason I suggest that “market-oriented folks” would rejoice if the ACA were repealed is that the statute does nothing to alleviate — and, in fact, exacerbates — one of the key problems driving health care costs through the roof, namely, the lack of price competition among health care providers. Gold-plated health insurance policies that have essentially turned insurance into pre-paid health care have eliminated virtually all price competition by assuring that no one pays for health care out-of-pocket (and thus no one shops on, and providers don’t compete on, price). The federal tax code encourages these gold-plated policies by exempting employer-provided health insurance from taxation (thus encouraging employers to provide a greater proportion of benefits in the form of pre-paid health care). If we could repeal the ACA, deal with the fundamental problem in the tax code, and encourage health insurance that looks more like insurance than pre-paid health care (perhaps coupled with tax exempt health savings accounts), we could actually address the cost of health care itself, as opposed to the cost of health insurance only.

      I probably shouldn’t have used the term “market-oriented” folks. I realize there are lots of “market-oriented” folks who support the ACA (e.g., Austan Goolsbee). I originally wrote “conservatives,” but I’m not one myself, so I decided to change the term.


      Steve: The ACA does far more than alleviate the so-called “free rider” problem in this context. If the mandate was about solving “free riding,” it would simply require individuals to purchase a policy that covers catastrophic expenses that uninsured individuals impose on the system when they incur a large expense they cannot afford. But of course the mandated policy covers far more than unantipacted expenses; it also requires individuals to purchase “insurance” for garden variety modest expenses, including those that are fully anticipated. Moreover, as I explained in a post in early 2011, the ACA Community Rating feature encourages free riding by mandating premiums for young and healthy individuals that are significantly higher than their expected (run-of-the mill plus catastrophic) medical expenses. (Imagine if, instead, the State mandated that good drivers must pay the same for car insurance as bad drivers.) It’s not surprising that such individuals would choose to self-insure in response to the ACA, particularly in light of the guaranteed-issue component that Thom highlights in his insightful post. Put another way, the amount of free riding is not entirely exogenous but instead depends in part on the existence or not of community rating and other features of the ACA (e.g., what sort of benefits the mandated policy must include). By mandating community rating, Congress made the free riding problem worse, by increasing the incentive to self-insure, as Thom explains, despite the tax for doing so. In short, I don’t think one can justify the coercive individual mandate to purchase a full-blown health insurance policy by invoking a problem, free riding, that the authors of the ACA themselves have exacerbated, especially when the policies mandated by the ACA far exceeds any effort to combat free riding.

      I’d also note that minimum rpm, like other similar restraints, is voluntary and thus not a good analogy for the coercive mandate sought by the Obama Administration but rejected by the Supreme Court as beyond the power of Congress. Of course, sometimes coercive collective action is appropriate.

      Anyhow, here (at the risk of undue self-promotion) is the post I mentioned above:





        Sorry for my delay in responding to Alan’s interesting post, and my only partial response here.

        There are “twin evils” in insurance markets — adverse selection and moral hazard. It is no surprise that healthy people would choose to do more self-insurance, but that is caused by and leads to the adverse selection problem. The rationale for community rating and the mandate is to deal with adverse selection. Moral hazard is usually dealt with by managed care contracts that require a certain amount of screening of covered services on an ongoing basis. In auto insurance, there is more of a concern with moral hazard, rather than adverse selection.

        Having said that, while community ratings prevent companies from giving lower health insurance rates to young individuals (who are generally healthier), no one seems to object to making teenagers pay more for auto insurance. I think that one explanation for the difference is the relative importance of adverse selection in the two insurance markets. The auto insurance market has not unraveled at all, whereas the health insurance market has suffered greater “failures” (in an economic sense of market failure). There also is believed to be a larger moral hazard component in driving than in living, though it appears that it is also a matter of what individual behavior it is considered legitimate to tax.

        I thought that Alan’s observation that ACA is not the “least restrictive alternative,” in that it mandates insurance for everyday as well as catastrophic illness, was pretty interesting. I expect that the rationales are (i) that it is difficult to draw the line; and (ii) lack of everyday care causes more serious illness and the need for more expensive care. It might be analogized to a required subsidy for driver education classes.

        I need to think some more about Alan’s interesting comment about RPM.


    I damn Roberts.

    The complete text of my letter to his is below:




Trackbacks and Pingbacks:

  1. Why Premium Subsidies and the “Employer Mandate” Won’t Solve the ACA’s Adverse Selection Problem « Truth on the Market - August 3, 2012

    […] Recent Comments FTC sacrifices the rule of law for more flexibility; Commissioner Ohlhausen wisely dissents on FTC sacrifices the rule of law for more flexibility; Commissioner Ohlhausen wisely dissentsWalter Sobchak on Epstein on Posner’s “Patent Adventurism” in Apple v. MotorolaFT Alphaville » Further reading on Happy 10th Birthday Sarbanes-OxleyRandy Picker on Chicago Continues to Thwart Food TrucksAndrew Clearfield on Chicago Continues to Thwart Food TrucksConference Honoring Larry Ribstein « Organizations and Markets on GMU Law & Economics Center Presents “Unlocking the Law: Building on the Work of Professor Larry Ribstein”GMU Law & Economics Center Presents “Unlocking the Law: Building on the Work of Professor Larry Ribstein” « Truth on the Market on Unlocking the Law SymposiumInformation Overload on Macey Throws Some Cold Water on the CFPB’s New Mortgage Disclosuresnorthfork investor on Macey Throws Some Cold Water on the CFPB’s New Mortgage DisclosuresSteve Salop on Why Roberts’ Tax Reasoning Ultimately Damns the Affordable Care Act (But Not in a Good Wa… […]