The debt ceiling
A liquidity view
It seems that a deal has been reached to give the US Treasury a reprieve, for now at least. The Republican minority in the US Senate has consented to an increase of $480 billion for the so-called debt ceiling, a deal which allows the Treasury to raise enough cash to get it to December, at which time, presumably, the exercise will be repeated. This graph shows the new debt ceiling:
The pressure is off for the moment, but it will be back. So maybe this is a good time to formulate a couple of brief opinions regarding the debt ceiling and the consequences of a possible US default. It seems that the thing I could hope to add to this complex and speculative debate would be to view the issue through the lens of liquidity—the ability to make payments.
The US government's survival constraint
We might helpfully consider a comparison with the situation of Evergrande, the gigantic and heavily indebted Chinese property developer. Evergrande, recall, faced a debt service payment on September 23, and whether they could make the deadline was doubtful. The day came and went without news, though, and it emerged that the company in fact has a 30-day grace period before it is technically in default. The survival constraint, in Minsky's terms, will likely still prompt action—a bailout or restructuring—but not quite in the way it seemed.
Does the US government similarly face a survival constraint? This question is central to monetary arguments about the debt ceiling. US Treasury Secretary Janet Yellen told Congress that the Treasury was coming close to not being able to keep all of its obligations. She was not talking about something theoretical, but rather about the practical inability to pay, within the rules of the system as they are. As with Evergrande, the Treasury has payments that it must make, and needs to get the money from somewhere:
The last few weeks show us what it looks like when the obligation to pay starts to bind for the US Treasury. Having exhausted its sources of cash, the Treasury and its attorneys concluded that the debt ceiling means it has to choose which obligations to pay, and which not to pay. Yes the debt ceiling is arbitrary, yes Congress has already chosen the path of spending and taxation that put the Treasury on this path, yes there are plenty of people who would buy the Treasury's debt securities; but nonetheless, the Secretary feels the constraint as binding.
Legislative and monetary loopholes
This might explain the weight of procedural and technical considerations in the discussions and commentary on the issue. The urgency of the debt ceiling has prompted calls for legislative quick-fixes, including modifying the rules of the Senate so that the vote can proceed with a simple majority, or using the budget reconciliation process to achieve the same result. It has also prompted calls for monetary quick-fixes, including the perennial suggestion of minting a trillion-dollar platinum coin. By request, I will devote a moment to this question.
I am no legal scholar, but the law seems clear enough in allowing the Treasury to mint a trillion-dollar platinum coin. (Note, there is no reason to think that the price of platinum as a precious metal, some $1,000/oz., has anything to do with the value of the coin.) Immediately upon minting the coin, the Treasury would face the problem of trying to spend it. The only place to make a deposit in that denomination is at the Federal Reserve. The central bank, in my opinion, would be unlikely to accept such an asset without an understanding that the transaction would eventually be reversed. The Fed would be convinced to accept the coin, that is, because the Treasury would implicitly be creating a liability:
The first step, minting the coin, is making something from nothing. But the coin doesn't really have any value until it gets spent in the second step. And what gives it value has nothing to do with the coin, obviously. It has to do with the promise to buy it back. As the T accounts show, this is just a procedural loophole to allow the Treasury to issue debt not subject to the limit.
I think this is unlikely ever to happen. For the Fed to accept a trillion-dollar platinum coin would require the central bank to cross a big threshold in the types of assets it can accept. But if the Fed were willing to do that, there would be other, easier options also available. As this explainer from the Financial Times notes, the Fed discussed in 2013 the possibility of expanding its collateral eligibility to allow it to buy defaulted Treasury securities, or to accept defaulted securities against loans of non-defaulted securities. True, Jay Powell called the idea "loathsome" at the time. But if we are looking at technical quick-fixes, this one seems more palatable than accepting an entirely new instrument. The rule broken by such a step is much more obscure than the more obviously transgressive trillion-dollar coin.
More generally, the real liquidity constraint is that the Treasury is up against the limits of its rules, neither more nor less than that. But other actors still have wiggle room. The Fed could accept defaulted securities. Perhaps the legal definition of default could be suspended. Maybe everyone could simply ignore a default, like the emperor's new clothes, until the political storm passes.
The most likely scenario
I think this is the kind of response we might expect if the US does eventually default: someone will create some space, just enough, without actually changing the balance of power. That describes what the US legislature did today: a stop-gap deal for the next two months. In all likelihood, another stop-gap deal will be reached in December. And so on.
Could this be the end of the dominance of the dollar? Some are making that argument. But there is not an actual alternative on offer, that I can see. Meanwhile trillions of US dollars churn through money markets every day. That reality creates a strong need for the continuity of the system. In the end, the need might be so strong that Congress doesn't have as much say in the matter as it thinks.