Affordable Care Act, Part III: When is a "penalty" actually a "tax"?

Is the “penalty” that people who fail to get health insurance must pay under the Affordable Care Act actually a “tax”?

Oddly enough, the Supreme Court’s answer is not that the Act did in fact create such a tax, but only that it is “fairly possible” to interpret the Act that way. Poor Mitt Romney – it’s easy to see how he could get confused, since the justices are in such disarray on the point too. Four of them (the “joint dissenters,” Scalia, Kennedy, Thomas and Alito) insist that the Act’s penalty is not a tax. Chief Justice Roberts, meanwhile, says that “[t]he most straightforward reading of the mandate is that it commands individuals to purchase insurance” (opinion at 31) – or in other words that it is a regulation of behavior, to be judged under the Commerce Clause’s authorization of the regulation of interstate commerce, and (for Roberts plus the joint dissenters) found invalid on that score.

But Roberts goes on to explore whether the law can reasonably be read instead as imposing a tax. As to this possibility, he tells us that “[t]he question is not whether that is the most natural interpretation of the mandate, but only whether it is a ‘fairly possible’ one.” (32). This inquiry is based on a longstanding principle of statutory interpretation, the “canon of constitutional avoidance,” which essentially tells courts that to avoid the risk of holding a federal statute unconstitutional, they should find a way, if one is “fairly possible,” to interpret the statute that avoids the reading – even if it was the more natural reading – that posed serious constitutional problems. Ultimately Roberts says (for himself alone) that the statute can be read this way, “[g]ranting the Act the full measure of deference owed to federal statutes” (id.), and the four liberals concur in the argument he then develops to justify this conclusion (33-44). In the course of that argument, Roberts observes that “[w]e see no insurmountable obstacle” to the interpretation being defended – hardly a vigorous declaration that it is in fact the most plausible reading! (38) So actually no one seems to think the likeliest reading of the law is as a tax.

For fans of statutory interpretation (such as me), this decision is an important instance of how much difference this form of legal reasoning can make. It is fair to say that Roberts has to work to find a way to read the law this way. After all, as he points out, the statute “states that individuals ‘shall’ maintain health insurance” (id.), language which certainly sounds like a mandate of behavior or, in other words, a regulation, with the breach of the regulation punished by a “penalty,” another word from the statute itself. In fact, the dissenters point out that the word “penalty” is used to describe this payment eighteen times in the law. (Joint dissent at 21.) They also cite repeated instances of the statute referring to the “requirement” of purchasing insurance. (Id. at 19.)

Moreover, and quite remarkably, the question of whether this payment was a “penalty” or a “tax” comes up twice in the case. The very first issue in the case is whether the Court can hear the case at all, given the existence of another statute called the Anti-Injunction Act, 26 U.S.C. § 7421(a), which forbids suits to challenge taxes before they are paid. The majority position ultimately is that for purposes of this statute, the payment is not a tax, but that for purposes of assessing its constitutionality, it is. The dissent says that this result “carries verbal wizardry too far, deep into the forbidden land of the sophists.” (Joint dissent at 28.)

How does Justice Roberts manage this? The answer is that the Anti-Injunction Act is simply another statute (rather than part of the Constitution); so long as doing so doesn’t somehow violate the Constitution, Congress can define terms in statutes any way it wants. If it doesn’t want the Affordable Care Act payment to count as a tax for purposes of the Anti-Injunction Act, that’s entirely Congress’ call. So here the majority is persuaded that Congress in the Affordable Care Act made clear that the “penalty” was not meant to count as a “tax” under the Anti-Injunction Act. (Roberts at 11-15.)

But for purposes of judging the constitutionality of the statute, the question of whether the payment is a “tax” is not entirely in Congress’ hands. For purposes of the Constitution, Chief Justice Roberts and the four liberals who join him on this point declare (and maintain that precedent supports them), Congress’ “choice of label” is not controlling. (Roberts at 33) Something may be a tax even though Congress called it a penalty, Roberts writes (34-35); and he adds, quoting an earlier case, that “the ‘question of the constitutionality of action taken by Congress does not depend on recitals of the power which it undertakes to exercise.’” (39) But the fact that Congress’ labels aren’t controlling just means it’s possible that something labeled a “penalty” could be something else; we still need affirmative reasons for concluding that it is something else. What are those reasons?

One reason has to do with the definition of a penalty. Roberts writes that it is a central feature of “penalties” that they “‘mean punishment for an unlawful act or omission’” (37), and so in determining whether the required payment is a tax or a penalty it becomes important to determine whether the failure to buy health insurance (which triggers the duty to pay) is or isn’t unlawful. In other words, is the “individual mandate” not actually a “mandate” at all? Apparently that is indeed the case. As Roberts puts it (id.), “[w]hile the individual mandate clearly aims to induce the purchase of health insurance, it need not be read to declare that failing to do so is unlawful.” Instead, it appears that “if someone chooses to pay rather than obtain health insurance they have fully complied with the law.” (As one of my sons pointed out to me, on this reading of the law the "penalty" for not buying health insurance is the equivalent of the baseball "luxury tax," the payment individual baseball teams must make to Major League Baseball when their payrolls grow too extravagant.) Congress, Roberts rather cogently points out, likely “did not think it was creating four million outlaws” (38) – the number of people predicted to choose to pay the penalty rather than buy insurance. (37)

(Here I have to add an aside on an odd feature of the case. My impression is that normally the canon of constitutional avoidance is brought to bear to determine what a statute commands or forbids. Here, the immediate question before the Court with respect to the penalty or tax payment is not what the statute directs people to do or not do as what to call those directions. People must pay money to the government if they do not buy health insurance, and nothing in the Court’s decision changes that. All that happens is that the Court determines that these payments can be interpreted as “taxes” for purposes of the Constitution. But when the Court bolsters that conclusion by arguing that the statute does not actually mandate that people purchase health insurance, it does affect the meaning of the statute in a somewhat concrete way. We now know that those who do not buy health insurance are not lawbreakers. If the duty to buy health insurance was a legal mandate, then failure to comply would have been a violation of the law – and it’s of some importance to people to know that they are, or are not, “lawbreakers.” But one last odd feature: though Roberts seems to view the question of whether it is lawful not to buy health insurance as an important point, he discusses it in somewhat tentative language. For instance, he writes, referring to the large number of people who it is predicted will pay the tax rather than buy health insurance: “That Congress apparently regards such extensive failure to comply with the mandate as tolerable suggests that Congress did not think it was creating four million outlaws.” (37-38) I think one can rely on this case as firm legal authority for the proposition that those who don’t purchase health insurance are not lawbreakers – but the Court’s language on the point is a shade short of absolute. )

Roberts is able to invoke several other arguments (not all of which I’ll retrace here) to support his conclusion that the payment is actually a tax. For one thing, the “penalty” – which does have one other statutory name, the “shared responsibility payment” (33) – is collected through the income tax system. Interestingly, the IRS is barred from using its heaviest enforcement weapons, such as criminal prosecution, to collect it. (7-8) For another, the penalty is predicted to generate quite a lot of money, $4 billion by 2017, as taxes are supposed to. (33) Moreover, the penalty amount is decidedly less than the cost the penalized person would have incurred in purchasing the insurance – suggesting that it’s not much of a penalty, and therefore that it can be read as not being a penalty in the first place. (35) As already mentioned, “it is estimated that four million people each year will choose to pay the IRS rather than buy insurance.” (37) Roberts acknowledges that this tax is certainly meant to affect behavior (that is, to encourage people to buy health insurance), but “taxes that seek to influence conduct are nothing new” (36). At some point, a tax can be so punitive that it can’t any longer be upheld as a tax, but Roberts is confident that this tax doesn’t cross that line (43) – and the large number of people who apparently are prepared to pay the penalty (oops, the tax) seems to support that conclusion. (43)  

To all of this the joint dissenters respond emphatically. They write that (joint dissent at 18):

[W]e have never held – never – that a penalty imposed for violation of the law was so trivial as to be in effect a tax. We have never held that any exaction imposed for violation of the law is an exercise of Congress’ taxing power – even when the statute calls it a tax, much less when (as here) the statute repeatedly calls it a penalty.

And, they say, this payment is “unquestionably” “imposed for violation of the law. Citing the statute’s many references to the “requirement” of purchasing insurance and the “penalty” for not doing so, they write (at 21):

[W]e have never – never – treated as a tax an exaction which faces up to the critical difference between a tax and a penalty, and explicitly denominates the exaction a “penalty.”

These “never’s” may be overstated, by the way. Roberts emphasizes one case “in particular” that did interpret a payment, labeled by statute as a “penalty,” to actually be a “tax.” That earlier decision was not about the constitutional character of the payment but rather about its status under the bankruptcy statute – but it’s not clear why Congress’ label could be overridden for purposes of another statute but not for purposes of understanding the nature of the payment under the Constitution. (Roberts at 35 n.7; the joint dissenters respond in their opinion at 17 n.5.)

In any event, based on precedent and other considerations, the dissenters conclude that “to say that the Individual Mandate merely imposes a tax is not to interpret the statute but to rewrite it.” (Joint dissent at 24.) Judicial legislation is always problematic in our system, but here the dissenters see two special concerns as well. First, the last thing courts should be doing is imposing taxes, a matter of special democratic concern. Second, once the penalty is construed as a tax, a new constitutional problem has to be addressed – the constitution’s requirement in Art. I, § 9, cl. 4, that “direct taxes” be (as the dissenters put it) “apportioned among the States according to their population.” (25) This requirement blocked the institution of the federal income tax until the 16th amendment was passed to remove the barrier, but it remains part of the Constitution, and while it wouldn’t apply to income taxes, it could conceivably apply to this tax. To figure out what the impact of the “direct tax” provision would be, the dissenters say, is no easy matter, and that’s another reason not to adopt an interpretation of the statute that requires resolving this additional constitutional problem. In fact, the government didn’t even discuss the question in its opening brief – a further sign, the dissenters suggest, of how far anyone was from really believing this statute could be read as a tax law. (25-26) (The majority, however, is undaunted and concludes that this tax isn’t a direct tax. (Roberts at 40-41.))  

Are the dissenters right? Does the Supreme Court’s upholding of the Individual Mandate rest on a judicial rewrite of the statute? My impression is that the answer is – just barely – no. It’s clear that Congress meant to present the country with something called a “requirement” (to buy health insurance), supported by something called a “penalty” for failure to do so. Congress, in other words, “framed” the law as a mandate backed up by punishment. But the punishment is too mild – it’s cheaper to disobey this law than to obey it. And it is at least unclear that Congress meant to turn the millions of anticipated non-purchasers of health insurance into lawbreakers – rather than tax-owers. Perhaps the weakness of the penalty and the ambiguity surrounding the status of non-purchasers reflect how controversial the law was; the legislators barely passed the statute, and when they did so they passed a soft version. But that ambivalence marks the space within which more than one interpretation of what Congress had done became “fairly possible.”

Even so, Roberts himself seems only just persuaded. The gossip about his having switched sides at the last moment would confirm this, but I mean to emphasize more the hesitation in the words he wrote on behalf of himself and the four liberals. They find “no insurmountable obstacle” to their decision. Roberts, for himself, says he reaches the decision “[g]ranting the Act the full measure of deference owed to federal statutes.” The canon of constitutional avoidance is a quite venerable technique of statutory interpretation, and is sometimes employed quite dramatically, but I suspect that this is one of the more aggressive acts of interpretive avoidance on record.

And what justifies that? Roberts’ answer, which I’ve quoted already a couple of times, is that he is “[g]ranting the Act the full measure of deference owed to federal statutes.” One translation of that phrase is that he is deftly avoiding a confrontation with the President and potential injury to the Supreme Court’s standing in the country – a legitimate concern for the justices, as I’ve already argued in my first post on this case. But as a matter of legal argument, Roberts is harking back to an old presumption, perhaps not so much acknowledged in recent decades, that federal statutes are constitutional. He writes, quoting a case from the 1880s, that “‘Proper respect for a co-ordinate branch of the government’ requires that we strike down an Act of Congress only if ‘the lack of constitutional authority to pass [the] Act in question is clearly demonstrated.’” (Roberts at 6, writing only for himself). This presumption is certainly rebuttable, but it rests on an important premise – namely that members of Congress and the President, who together make our statutes, are acting with fidelity to the Constitution. Maybe not, of course, but as a starting principle I think this idea has a lot to recommend it. If that is one’s starting point, then it makes sense to look, and even to look hard, for ways to square what Congress and the President have done with the Constitution’s commands. If the price of upholding their work is to reread it, even dramatically, that is a lesser price than would be overturning it.

It’s certainly possible to argue that the canon of constitutional avoidance, employed this way, is actually an abuse of judicial authority. That’s roughly what the joint dissenters would say in this very case. Sometimes that may be so. But I think its use here, to preserve what Congress and the President did while still laying out new constitutional ground rules for the future, was justified, and wise.

The Affordable Care Act, Part II: what's broccoli got to do with it?

How strong was the commerce clause argument in the Affordable Care Act case (National Federation of Independent Business v. Sebelius)?

I have to say that I sympathize with the conservative justices’ concern that the commerce clause shouldn’t become the basis for congressional power over everything. Of course, it wouldn’t be – that is, no matter what Congress has power to regulate, it can’t regulate anything in violation of the Bill of Rights. Moreover, just at the moment Congress can hardly pass legislation at all, on any subject, and so it’s a bit difficult to say we’re all in peril of congressional overreaching right now. In fact, it’s difficult to see any Congress ever legislating in true and utter disregard of the states, from which every member of Congress is elected. Nevertheless, I agree with the basic idea that liberty is safer if no one unit of government is too powerful. Moreover, perhaps in part as a result of having studied South African law from the days before that country’s Parliament was subject to meaningful constitutional limits, I’m not entirely comfortable with trusting that Congress just won’t choose to exercise power once we’ve decided that it could if it wanted to.  

So I think the idea that there should be some limits on the commerce power has appeal. But what’s startling about this case is how implausible it is as the occasion for finding such limits. Health care is a huge part of the United States economy, the stuff of interstate commerce every minute of the day. Moreover, it seems quite clear that if the individual mandate had not been upheld, the elaborate scheme of the Act would have been greatly undercut. The various provisions of the Act designed to make health insurance available at reasonable prices to people with preexisting health problems – an essential feature of the law – would likely have been unsustainable without the premiums to be paid by healthy young people purchasing insurance because of the individual mandate. In fact, the four dissenters (Scalia, Kennedy, Thomas and Alito) were convinced that without the individual mandate and the expansion of Medicaid, the whole system would be so compromised that they would have thrown out the entire statute, every single section. So not only is health care clearly part of interstate commerce, but the individual mandate – the portion of the statute evaluated under the commerce clause – was necessary to the overall regulation of commerce achieved by the law, and so should have been seen as “necessary and proper” to Congress’ exercise of its power over interstate commerce.

Except for one thing. That was that, as the conservative justices (including Roberts) saw the matter, what Congress was regulating with the individual mandate was not activity but inactivity. There was no commerce to regulate, these justices believed, until the statute forced everyone into the insurance market by mandating that they get insurance. The conservative justices insisted that the power to regulate was not the power to create commerce but the power to manage what already existed, and that Congress had never previously been allowed to regulate the failure to engage in interstate commerce.

This argument strikes me as particularly weak. It may well be that Congress has never regulated commercial non-activity, and correspondingly that no precedent ever said that Congress could do that. It’s probably also true that no precedent ever said that Congress could not do it, and that the reason Congress didn’t regulate inactivity was that doing so wasn’t so deeply integral to a larger regulatory effort as it was in this case.

Yet one might respond that “inactivity” is simply, definitionally, beyond the range of both Congress’ commerce power and its adjunct, the necessary and proper clause. One might, but why? As Justice Ginsburg says in her separate opinion, dissenting on this issue, similar efforts were made in the 1930s, notably to limit Congress’ power to regulating activities with “direct” effects on interstate commerce, while barring Congress from dealing with activities whose effects, however large, were merely “indirect.” That idea has long been abandoned, for at least three good reasons, each applicable here too.

First, the words “direct” and “indirect” don’t appear in the commerce clause or the necessary and proper clause. However conservative the justices who employed these terms, they are judicial interjections rather than part of the constitutional text. “Activity” and “inactivity” are similarly absent from the constitutional text.

Second, and more or less by design, the concepts of “direct” and “indirect,” like those of “activity” and “inactivity,” have nothing to do with the actual impact of what people are doing (or not doing), either its impact on the economy or – the real point – its effects on their other citizens and residents of the United States. It isn’t a virtue to be deliberately disconnected from sensitivity to real impacts.

Third, these words – “direct” and “indirect,” “activity” and “inactivity” – are pretty deeply obscure. Justice Ginsburg argues at length that people who don’t buy health insurance are not inactive in the market for health care; they will, on the contrary, very likely consume health care within the foreseeable future, say 5 years. They may need that health care in a moment, since no one knows when injury or illness will strike without warning. So they are, as Ginsburg suggests, actually “active” in the market, via the route of “self-insurance.” For some, moreover, self-insurance will fail; they will wind up unable to pay for their health care when they actually need it, and these health care consumers are every day engaged in a process of free riding on the rest of us. Others – those whose premiums are so needed in order to pay for the costs of making insurance widely available to people who can’t currently purchase it – are not free riding but rather are resisting paying part of society’s bill; but they too may be seen as actively refusing to purchase, rather than simply being inactive. When is a “refusal to act” (or, to use another phrase, a “failure to act”) actually an “activity”? The constitution doesn’t say. Chief Justice Roberts says that someone who is sitting around and doing nothing is not in the “rest” market – but we’re a long ways from that case in talking about how people deal with the inevitability that they will need health care.

All of this makes me feel that it was simply a mistake to try to draw a line between what the commerce clause reaches and what it doesn’t that is based on the supposed distinction between “activity” and “inactivity.” But what about the broccoli argument? It is, after all, true that broccoli purchases, aggregated across all the consumers in the United States, have a substantial effect on interstate commerce, and so – as the conservatives said – the kind of logic I’m endorsing would suggest that the federal government could order us all to buy broccoli.

I don’t think the government should be able to make us buy broccoli. (Should it be able to prevent us from buying huge containers of sugared soft drinks? That’s actually an easier question under the commerce clause – buying soft drinks is an activity, by any lights, and so regulating that activity shouldn’t raise any of the questions that mandating insurance did.) But back to broccoli – I think that heading off this possible extension of federal power is a matter that deserves attention, and that finding a coherent rule that does this may not be easy. But broccoli is not this case, as lawyers say. We are a long ways from broccoli in thinking about how to finance health care for the American people, an issue of almost overwhelming commercial and economic import.

So it seems to me, in the end, that the conservatives picked the wrong case to draw a commerce clause limit in. Limits may be needed, but they should have been drawn so that this statute fell within them rather than beyond their bounds. And it would have been good to find limits that avoided the incoherence of the “activity”/”inactivity” line that is now apparently part of our constitutional law.

Why did Chief Justice Roberts vote to affirm (most of) the Affordable Care Act?

A first reaction to the Supreme Court’s Affordable Care Act decision (National Federation of Independent Business v. Sebelius, decided June 28, 2012, and available at the Supreme Court's website):

Few people thought Chief Justice Roberts would supply the fifth vote to keep President Obama’s health care law on the books. What should we make of the fact that he did?

One possibility is, of course, that he simply voted his convictions about the important issues of constitutional law the case presented. He is committed, then, to finding limits on Congress’ power under the constitution's commerce clause and spending clause (and on the spending clause issue he does cut back on what the statute can do), but still he accepts that Congress’ authority under the tax power is very broad. He also  honors in full the rule of statutory interpretation that says statutes should be interpreted, if fairly possible, so as to be constitutional rather than unconstitutional -- which in this case meant to discern that the law’s “penalty” for those who don’t purchase insurance was not a regulation, which would have been beyond Congress’ power under the commerce clause, but a tax, which was within Congress’ power to tax.

I have no ground for doubting that these are in fact Chief Justice Roberts’ beliefs. But suppose for a moment that they weren’t. Suppose that what actually happened was not that he voted his beliefs but that he came to the conclusion that for the Supreme Court to overturn the Affordable Care Act would deal a damaging blow to the Court itself by calling its reputation for impartial, nonpolitical judgment even more into question than is already the case. So, while carefully laying out his views on the Commerce and Spending Clauses, he found a way to extricate the Court from the logic of those conservative principles – through the convenient medium of a flexible use of statutory interpretation to turn the law into something it might not have been (and something which could be held constitutional after all).

Did he do this? I don’t know, and I don’t know that the interesting speculation about the possibility that Roberts changed his mind late in the game helps us to decide why he did so. If he changed course late, he could have done so either because of legal argument or political calculation.

But let’s assume that it was the latter. Was that bad? That is, was Roberts wrong to consider the institutional position of the Court when he decided how to vote? I would say not. I think that there are issues of fundamental human rights on which judges must say, “Fiat justitia et ruat caelum.” (I’ve been waiting since my high school Latin classes for a chance to use those words, which roughly mean “Let justice be done and the heavens fall” – although even judges today probably would put the point in English!) But the job of the Supreme Court is not only to do justice in individual cases, but also to build the law of the land over time, and actually for both of those tasks the Court must retain the trust of the people at least to a substantial extent. The Court has no army, as has been pointed out many, many times. Its ability to enforce its judgments depends on the cooperation of the other branches of government and on the people. So if Roberts felt that he should lay out constitutional principles but find a way not to apply them so as to overturn the central legislation of the Obama presidency, he was acting in a tradition that goes back at least to Chief Justice Marshall in Marbury v. Madison in 1803 – a decision that established the power of the courts to hold federal statutes unconstitutional and by doing so actually avoided a much graver confrontation with the Jefferson Administration.

But isn’t this just a form of stealth jurisprudence, a device to further Roberts’ long-term goal of shifting the law to the right while escaping sharp public scrutiny for what is subtly underway? Maybe. That’s a good reason to criticize Roberts’ views, if they deserve criticism (and I think many of them do). But I don’t think that it’s wrong in principle for judges to seek to change the interpretation of the constitution – they may be right or wrong in their interpretations, but the fact that they’re changing past interpretations isn’t what makes them either right or wrong. It’s just not possible to say that our constitutional law is or should be fixed and changeless (even if some of the current conservatives assert its supposed unchanging, original meaning as their basis for overturning what they see as the mistakes of recent decades). The law will change.

What stops legal change from being political change pure and simple isn’t easy to define (and some people may believe there really is no distinction). But I think part of what makes law something other than politics is that it proceeds, usually (there are important and valuable exceptions), in a deliberate and incremental way. If Roberts’ views continue to command support, there will for sure be more laws overturned in the future – but it is important that we come to that point along a path that gives weight to contrary convictions, that shifts slowly rather than avulsively from past decisions, and that gives us all more time to take stock. All of that will give us more reason to believe that we are in fact observing (to use a distinction Alexander Hamilton affirmed in The Federalist Papers) the application of judges’ “judgment” rather than merely their “will.” 

So: if Roberts was acting on the basis of institutional calculation as well as legal principle, was he engaged in manipulation or statesmanship? I'd pick the latter. But in a way the question isn't a good one, because in this context there is no absolute line between these two. Here, as perhaps in many other situations as well, some measure of calculation is an integral part of wisdom.