Would it Even Be Possible to "Prioritize" When Republicans Create a Debt Ceiling Crisis?

by Neil H. Buchanan

The public discussion of the debt ceiling crisis is reverting to a slow burn, now that everyone (and I do mean apparently everyone) has offered their hot take on last week's report that the US has again hit the formal debt ceiling.  That was both big news and a non-event, the former because of course it is a very big deal that Republicans have made it clear that they truly are willing to shoot the hostages this time, the latter because the real drop-dead date is not when we hit the ceiling but when Treasury exhausts the (misleadingly named and revealingly absurd) "extraordinary measures" that a long-ago Congress made available for these situations.

With the heat temporarily turned down, it seems worth taking a moment to address an issue that I mentioned briefly in a column last week: What is wrong with the possibility of "prioritization"?  Specifically, if President Biden soon finds himself on the drop-dead date without a deal to increase/suspend/repeal the debt ceiling, would it be possible for him to ignore the Buchanan-Dorf advice of issuing new debt in order to pay all of the bills (which we have long called "the least unconstitutional option") and instead pick and choose which bills to refuse pay in full and on time?

I am not asking whether such a move would be constitutional or even a good idea, because it is obviously neither of those things.  I am asking whether it is possible.  It is not -- or, more accurately, even if it is logistically possible, it is legally untenable as well as politically suicidal.  Why?  Glad you asked.

To emphasize some constitutional points, Professor Dorf and I have made it clear that the separation of powers requires a president to obey the appropriations laws precisely as enacted, and we have also shown that even if a president were to try to prioritize who gets stiffed, he could do so only by either (a) promising to pay the overdue bills as soon as possible, which would turn those unpaid bills into new future obligations of the government and thus still debt, meaning that we would still go over the debt ceiling, or (b) repudiating obligations entirely, which even the most skeptical commentators have conceded is the very essence of a Fourteenth Amendment (Section 4) violation.

No one has even attempted to address the vast majority of those analyses in the decade since we published our first article.  The closest anyone has come to responding to any part of our analysis is to say that Section 4 applies only to payments due on Treasury securities (but not to any other US obligation), and since there is always enough incoming tax revenue to pay the interest on the debt on any given day, interest payments can and must be prioritized over everything else.

Again, note that such an analysis, even if true, addresses only one part of our analysis -- and in my estimation, it is the least important part.  Even so, I want to take this opportunity today to ask: How would this even work?  Answer: Not well, and probably not at all.  Understanding why is unexpectedly enlightening.

It is not surprising that Republicans have been bandying about the assertion that prioritization is possible, because if true, that would allow them to say that Biden has the responsibility to do what they claim to want: have the government spend less money -- even though they would have him do so by retroactively canceling payments for things that people and businesses have already provided or have otherwise already qualified for.  Who cares about a few disappointed reliance interests?!

It is surprising, however, that some supposedly neutral observers and even some liberals have started to make noises about prioritization.  In my column last week that I mentioned above, I noted that two reporters for The New York Times blithely claimed that Biden "could prioritize payments" and that their editors included that claim in the caption of the photo that sits under the headline of that article.  I did note that those reporters immediately undermined their own confident claim, but the point here is that they and their editors presented the availability of prioritization as simply a known fact.  It is anything but.

Along with purportedly neutral reporters, some people on the left have begun to push for prioritization as well, though for different reasons.  No less a liberal eminence grise than Robert Kuttner, for example, decided earlier this week to weigh in on the side of prioritizing payments on Treasury securities:

As our colleague David Dayen pointed out yesterday, the Treasury can keep bondholders whole even if the debt ceiling is technically exceeded, by prioritizing paying interest on bonds. This avoids the bond market collapse that Republicans hope will force the Democrats to agree to massive cuts. But using limited money to protect the bond market will indeed compel a government shutdown because it will divert money from other outlays.

That onus should be on the Republicans, as vital services are suspended and Social Security checks don’t go out. Eventually, the backlash will force the Republicans’ hand and enough Republicans will vote with Democrats to increase the debt ceiling and reopen the government. But hanging tough on this strategy will be hard for many at the White House because of the sheer messiness of shutting down the government and worry about who will ultimately take the blame.

Among other problems (having to do with some apparent confusion about what causes a government shutdown), I should say that I find Kuttner's overall strategy here puzzling and even a bit cruel, to say nothing about its politically toxic nature.  (Seriously, a liberal is telling Biden to pay bondholders -- which includes the Chinese government, hedge funds, and foreign corporations -- while telling Social Security recipients that they have to wait?)  But that is an argument for another day.  What matters here is that Kuttner simply accepts as a starting point that prioritization can be done, relying on a claim by one of the writers at his magazine.

In turn, that writer's argument (in a blog post that was mostly about the platinum-coin option) had this to say about prioritization: "[In 2011] the Fed set up a process for making prompt payments to service debt and delaying other obligations," with the embedded link leading to a Wall Street Journal news article reporting on the transcript of an emergency conference call at the Federal Reserve on August 1, 2011:

Transcripts from the call show the Fed—acting as the Treasury’s fiscal agent and at its direction—was prepared to keep making principal and interest payments on time, and delay other payments as necessary, if the deal fell through and the debt limit wasn’t raised in time.

Louise Roseman, the director of the Fed’s division of reserve bank operations and payment systems at the time, told Fed officials that the central bank, its regional reserve banks and the Treasury had jointly developed and codified the procedures into a “special operating circular,” which it planned to issue to banks if Congress failed to strike a timely agreement. The plan was “approved by the Treasury as reflecting what it would like the Reserve Banks to do as its fiscal agents if it ever came to that,” she said.

That sounds fairly definitive, at least in the sense that the relevant officials appear to have convinced themselves at the time that this was technically feasible.  That is in contrast to the views of Jacob Lew, who at the time was Director of the Office of Management and Budget and who later served as Treasury Secretary.  Lew has argued (as reported in that NYT piece that I mentioned above) that "[t]he systems used to send out payments are not finely calibrated enough for the government to quickly and surgically adjust who receives checks."

So which is it?  Fed and Treasury officials held an emergency meeting shortly before the first debt ceiling drop-dead date (the one in which Barack Obama caved and rewarded the hostage-takers) and said that they "had jointly developed and codified the procedures" to pay principal and interest.  Lew says, by contrast, that maybe they were fooling themselves.  Rather than try to resolve that question, which is in part a question of setting up computer algorithms (and which further assumes that Congress will be willing to pay for Treasury officials to do something off-book, to say the least), I want to ask exactly what the human thinking would have to involve before it could be reduced to operating code.

As described above, this sounds simple: When the time comes, pay the obligations on Treasury Securities before everything else.  But as the WSJ piece goes on to report, Roseman (the Fed official who claimed that they had worked out a way to put a prioritization plan into operation) offered a further comment that was potentially explosive: "Ms. Roseman said the Treasury planned to ensure it would be able to make the interest payments 'by holding back other government payments and accumulating sufficient cash balances in its Fed account to pay upcoming coupon payments.'"

Why do I say that this is possibly controversial, to put it mildly?  The combination of Roseman's line "if it ever came to that" in the block quote with the quoted lines just above should focus our attention on a problem of timing and the precision needed to prioritize payments: When, specifically, will it have come to that?  When, in other words, will Treasury and the Fed be in the position of knowing that they have to start stiffing people?

The day-to-day operations of Treasury are utterly banal, back-room accounting activities.  Bills have to be paid, so Treasury writes checks (thousands of them) every day.  At the same time, they have to make sure that the checks will not bounce, so they track how much tax revenue is arriving on any given day.  And in normal times, they float new bond issues when they need to make up any difference.  On some days, there is a surge of tax revenue, but not on others, which means that on some days, they do not need to borrow money.  There is no reason why the timing of the two flows (checks going out and revenue coming in) should match up at all, much less on a daily basis.

So what?  The government is required by law to pay the bills when they come due.  The people who disagree with Buchanan-Dorf say that when there is not enough money to cover a check, the would-be recipient is out of luck (either temporarily or permanently).  But no one says (at least explicitly) that it is acceptable for the government not to pay a bill that is due when there is enough money to cover it.  It is one thing to say, "Sorry, there's no money here to give you, and Congress won't let us borrow to cover our obligation to you," and quite another to say, "We have the money, but you'll still have to wait."

Why would such a person be told to wait?  Because, per Roseman, the government would be "holding back other government payments and accumulating sufficient cash balances in its Fed account to pay upcoming coupon payments."  Think about what that means.  The government anticipates that it might not have enough money when bond payments come due tomorrow, next week, or next month; so it tells veterans' hospitals, Social Security recipients, and SNAP (food stamps) recipients who are legally entitled to receive money today that they all have to wait.  Just in case.

The point is not that that is impossible to do this logistically, although it does open up a boundless set of questions about how much the government is supposed/allowed to do to anticipate possibly having to pay bondholders.  How far in advance?  How big a cushion can/must it amass?  But as a matter of law, what would be happening?  The government would be lying to the people who hold legally enforceable obligations, saying to them: "Sorry, we don't have enough money to pay you," when in fact it does have enough money to pay the people with legitimate and immediate claims on those funds.

I have no doubt that the emergency conference call in August 2011 went smoothly in one regard, with everyone agreeing that they had come up with a contingency plan "if it ever came to that."  But I strongly doubt that they confronted this timing problem.

Think about it this way.  Given that the stiffed would-have-been payees would have particularized injuries, what would the government's defense be upon being sued for nonpayment?  "We decided that your payments were not important enough to pay, even when we had enough money, because there were more important people who we didn't want to take the chance of disappointing later."

Again, it is one thing to claim to have been between a rock and a hard place -- "I only had one dollar and had to choose between two people to whom I owed a dollar each at the same time" -- but quite another to say that refusing to pay someone was based on predicted daily cash flows on later dates.

And when I say "quite another," I mean that in two very different ways.  First, as a matter of law, prioritization means violating the law and not paying some bills -- which, again, is not the least unconstitutional option, but I digress -- even when there is no defense of impossibility.  Second, as a matter of politics, are you kidding me?  Now, we are looking at someone saying this: "I was told that I might not be able to buy groceries because the government decided that it would pay interest on the bonds that Deutsche Bank is holding if there wasn't enough money to pay us both.  Now, I'm told that I'm not going to eat today because the government might or might not need the money to pay some mutual fund at some point in the future."

That hypothetical person's lament captures both the legal problem and the political problem.  When you stop paying people because of a maybe and not because of a definitely, you create both of those problems.

Note also that the existence of a binding debt ceiling and the (mostly) predictable paths of both collecting taxes and paying bills over time does not mean that these timing problems would not arise.  "I was going to stiff someone so that I could make good for the bondholders" does not answer the question of how or why the government chooses the unlucky few among the potential pool of non-prioritized obligees.  My hypothetical person could say, "You know, it would be a shame if someone else can't eat tomorrow, but my problem is that I can't eat today.  And maybe something good will happen tomorrow -- like, maybe, congressional Republicans will get their heads out of their collective rectum and end this insanity."  Guaranteeing injury today when it might not be necessary tomorrow, or at all, is not a viable answer.

This is madness, and it is truly amazing that so many people are lulled into complacency by -- or, even worse, see a political advantage in -- an anodyne word like prioritization.  Here is an accurate definition of that word: "Having the President of the United States decide to harm some people and not others, in violation of the law, because Congress gave him an impossible set of commands."  That the most talked-about form of prioritization would have the President taking from the middle class and the poor to keep the global financial community happy only makes it that much more shocking.