Could the King v. Burwell fallout be even worse than anyone thought? (Probably not)

NOTE I posted a partial version of this entry last night, then yanked it overnight in order to think about whether it was a valid concern or simply the same sort of click-bait headline hysteria that I've criticized other sites for posting.

After thinking it through, I"ve decided to go ahead and post a modified version with a bit of additional context and a less panicky headline. Think of it as more of a legal thinking exercise than anything else...but at this point, anything is conceivable.

For months now, I and many others have been sounding the warning bells re. the absurd King v. Burwell ACA case, currently awaiting a decision by the Supreme Court.

Until now, many people have pointed out that in a worst-case scenario situation, anywhere from 6 - 7 million people across 34-37 states enrolled in healthcare policies via would lose their federal tax credits, and thus, in nearly all cases, their policy as well...since they'd no longer be able to afford it. The actual number who are currently enrolled as of July (the earliest point that the credits could be cut off) will be roughly 6.5 million people (assuming that Oregon, Nevada & New Mexico are in the clear) or 6.7 million (assuming those 3 states aren't).

In addition to these 6.5 million or so folks, I've also estimated that an additional 1-2 million more would also lose their policies indirectly...because the massive price spikes (up to 50% in some cases) resulting from the first 6-7 million dropping their policies would likely price them out of the market as well.

Finally, I've also pointed out that another 5 million or so people (mostly enrolled in off-exchange individual plans) would likely keep their policies...but pay through the nose to continue to do so.

So...around 8.5 million or so losing their coverage completely and another 5 million facing crushing premiums. Pretty bad, right?

But wait! There may be even more!

A few months back, I made a bad joke by suggesting that if they do decide to effectively destroy the lives of up to 14 million people, the SCOTUS "might as well go ahead and make the ruling retroactive while they're at it" by forcing the IRS to collect back the billions of dollars in tax credits which have already been given out over the past year and a half to millions of folks (some of whom aren't even enrolled via the ACA exchanges anymore).

A couple of weeks ago, the subject came up again in a Twitter exchange (I can't bring myself to call it a "conversation") with Michael Cannon, one of the lead architects of the King v. Burwell case itself:

So the possibility that 50+% of the subsidies & penalties are illegal is *not* a significant caveat? @charles_gaba @RANDhealth #KingvBurwell

— Michael F. Cannon (@mfcannon) May 8, 2015

@mfcannon @randhealth However, unless you're suggesting SCOTUS would make the past 2 years RETROACTIVE (i.e., make everyone pay 'em back)…

— Charles Gaba (@charles_gaba) May 8, 2015

SCOTUS wouldn’t have to. The #ACA itself requires illegal-subsidy recipients to pay most/all of it back. @charles_gaba @randhealth

— Michael F. Cannon (@mfcannon) May 8, 2015

@mfcannon @randhealth You're saying if you win your case those folks will be retroactively req. to pay back up to 1.5Y worth of credits?

— Charles Gaba (@charles_gaba) May 8, 2015

I was reassured on this point by Ken Kelly, who pointed me towards the relevant U.S. Code:

(8) Application to rulings
The Secretary may prescribe the extent, if any, to which any ruling (including any judicial decision or any administrative determination other than by regulation) relating to the internal revenue laws shall be applied without retroactive effect.

...before getting into a bit of a pissing match with Mr. Cannon:

@mfcannon @irs @charles_gaba Nobody but you (appears) to think there is any such threat.

— Ken Kelly (@_KJKelly) May 10, 2015

So everything I said is accurate but you can’t say so directly because you’d rather disparage me. Stay classy. @_KJKelly @irs @charles_gaba

— Michael F. Cannon (@mfcannon) May 10, 2015

(etc, etc...they go back & forth like that for a bit).

Last night, I was further reassured by Nicholas Bagley, an associate law professor at the University of Michigan and an Obamacare legal expert, that this shouldn't be an issue because, according to a paper he co-wrote at the Yale Law Journal:

Could the administration decline to claw back tax credits?

If the Court invalidates the IRS rule at issue in King, those who purchased coverage through the federally facilitated exchanges will be in a tough spot. Under the ACA, their insurers will already have received “advance payment tax credits” from the IRS on their behalf. At tax time, the IRS is supposed to reconcile those advance payments with the tax credits that the purchaser is entitled to under the ACA.9 Since purchasers would not have received any tax credits but for the unlawful IRS rule, they would, in the normal course, have to pay back the amount that was improperly disbursed to their insurers.

Fortunately, Bagley goes on to bring up the same section of code that Kelly referred to:

The administration has the legal flexibility to avoid that harsh result. Under 26 U.S.C. § 7805(b)(8), the IRS can choose whether to give “retroactive effect” to a court decision or instead to apply that decision only in the future.10 Pursuant to this authority, the IRS could issue a rule specifying that individuals could properly claim tax credits, based on their annual income, for the period until the Supreme Court’s decision takes effect. (For the 2015 tax year, tax credits would have to be pro-rated.) If the IRS were to issue such a rule, it would not need to claw back tax credits that were disbursed to taxpayers prior to the Court’s decision in King.

OK, that sounds pretty better, although you never know; IRS commissioner John Koskinen and/or Treasury Secretary Jack Lew could, hypothetically, lose their friggin' minds and not issue such a rule.

HOWEVER, I was also given a different take by a commenter over at Daily Kos who referred me to a different paper by an associate law professor at the University of Iowa, Andy Grewal, in the Yale Journal on Regulation which says that:

 However, Section 7805(b)(8) may provide some relief to consumers.¹⁴ Under that statute, the Treasury can deny retroactive effect to judicial rulings, even ones made by the Supreme Court. But any action by the Treasury will fully protect only those who purchased federal policies during the 2013-2014 enrollment season. Purchasers of federal policies during the current enrollment season will not definitively establish their right to tax credits until after December 31, 2015, approximately six months after a potentially adverse decision in King v. Burwell.¹⁵ These taxpayers would need the Treasury to deny prospective effect to the Court’s ruling, a power not contemplated by Section 7805(b)(8).

    Arguably, Section 36B reflects a departure from the annual accounting concept, and the Treasury can use Section 7805(b)(8) to protect any advance payments processed before King v. Burwell takes effect. Under Section 36B, the eligibility for a premium tax credit turns on a month-by-month analysis even though a consumer’s actual tax credit or liability depends on annual household income and other factors established at the end of the year.¹⁶ Consequently, the Treasury might treat consumers as having established their right to tax credits at the close of each month and might establish some type of pro-ration regime for computing allowable credits.¹⁷

    But even under this scenario, consumers face potential problems. In the months after King v. Burwell takes effect, no credit related to a federal policy would be allowable, and taxpayers would have to repay any advance payments made for those months. Alternatively, the government might stop making advance payments on federal policies in July 2015, such that taxpayers would effectively see an unaffordable spike in their monthly premium payments. Either way, trouble awaits.

    Also, although Section 7805(b)(8) may provide relief for pre-King v. Burwell months, there’s no guarantee that the Treasury will exercise its authority under that statute, given the potential blowback it might face.¹⁸ If the Treasury flatly rules that King v. Burwell does not apply for the months preceding the Court’s decision, penalties on individuals and employers would follow. That is, the individual penalty for failure to obtain coverage and the employer penalty for a failure to provide coverage depend, in part, on whether Section 36B extends to consumers who purchase federal policies.¹⁹ If the Treasury broadly denies retroactive effect to King v. Burwell, then some individuals and employers will find themselves paying penalties even though they prevailed in the Supreme Court. Although it is doubtful that Section 7805(b)(8) was intended to let the Treasury rob taxpayers of judicial victories, the statute’s plain text does not force the Treasury to exercise its authority in a purely taxpayer-favorable manner.²⁰

    Arguably, the Treasury can turn off King v. Burwell only for consumers who purchase federal policies and allow it to take full effect for other individuals and for employers. Section 7805(b)(8) allows the Treasury to “prescribe the extent, if any, to which any ruling” operates without retroactive effect. The Treasury might thus deny retroactive effect to King v. Burwell only to the “extent” that it protects a consumer’s tax credits for federal policies, but no further.

    However, it is not obvious that Section 7805(b)(8) allows the Treasury to slice and dice a judicial decision that way. Instead, Section 7805(b)(8) might refer solely to temporal elements, not substantive ones. That is, the Treasury can choose only the “extent” of King v. Burwell’s retroactivity period and may prescribe, for example, that it takes effect as of June 1, 2015, or as of May 1, 2015, or as of some other date. Under this reading, the Treasury could not chop up the Court’s holding; it would have to accept the decision in toto, subject to whatever period of retroactivity it chooses.

In other words, as explained by "Throw the Bums Out" over at Daily Kos: theory under Section 7805(b)(8) Treasury can deny retroactive effect to judicial rulings.  However, there are several problems with that.

1.  The treasury might not do so given the backlash it may face from employers.

2.  The authority under Section 7805(b)(8) with regard to court rulings is new and has not been tested in court.  It is entirely possible that it is unconstitutional as applied to court rulings.

3.  The authority to do so may be limited.  Specifically, whether or not the Treasury can deny retroactive effect to only part of King v. Burwell.  Specifically the part regarding a consumer’s tax credits for federal policies but not the part that fines employers. (and again, this feeds into the first issue)


In 2014, nearly $15.5 billion was disbursed in Advance Premium Tax Credits by the Treasury Dept. to ACA exchange enrollees (the report gives the amount as $11 billion, but that doesn't incluee the 4th quarter since the federal "Fiscal Year" runs from September - September).

68% (5.45 million) of the original 8.02 million enrollees in 2014 did so via, and about 85% of all enrollees received tax credits last year. By my calculations, the total average monthly ACA exchange enrollment was around 5.5 million people. That means that every month, about 4.67 million people nationally received around $280 in federal tax credits apiece.

In other words, if I'm understanding the worst-case scenario described above properly, something like 3.2 million people might have to pay back an average of $3,300 apiece.

And that's just for 2014.

For 2015, assuming 6 months at an average monthly tax credit of $263, that means that many of those same 3.2 million people, plus millions more (6.5 million total) would theoretically have to pay back as much as $1,600 apiece.

Now, it's not an exact overlap. Many of those 3.2 million from 2014 didn't renew their policies, so there will be some who would only have to pay for 2014, some for 2015 and some for both.

However, the fact remains that if I'm correct about this potential nightmare, a strict, draconian ruling in favor of the plaintiffs in King v. Burwell could potentially result in some low-income Americans having to pay an extra $4,900 in taxes back to the IRS, dating back to 1 1/2 years ago.

Prof. Bagley's resonse to this was pretty cut and dry:

@charles_gaba @sangerkatz @afrakt Doesn't worry me. There's lots of helpful case law for the IRS.

— Nicholas Bagley (@nicholas_bagley) May 27, 2015

@charles_gaba @sangerkatz @afrakt I'm not worried about the practicality or legality of giving the decision prospective effect.

— Nicholas Bagley (@nicholas_bagley) May 27, 2015

I certainly hope he's correct, or a King plaintiff win could get far uglier than it's already looking.