This Week in the Mayonnaise Clinic: Debt Limit Scenarios
There are (better) options besides not paying interest on government bonds
Welcome to the Mayonnaise Clinic!
We’re just a couple of weeks away from a potential default on US government bonds, but nobody seems to be treating it like a grave crisis — not even the Washington policymakers intermittently lecturing Wall Street investors for refusing to treat it as a grave crisis. There are positive signs of progress toward a legislative deal to raise the debt limit, but the public statements from lawmakers in both parties look a lot more like the usual posturing for credit and blame over a legislative fight than the utterances of people who are truly scared we’re about to face a very optional financial crisis. These people just don’t seem that spooked.
I think I understand why they’re so placid, but to understand it, we need to talk about what might actually happen if Congress and the president fail to enact a debt limit increase before the end of the month.
Fortunately, reader Sean sent in a relevant question:
The options presented in the media seem to be very simplified and binary: raise the limit, or default. Are there no other options Biden can take to kick the can down the road, and pass blame onto his political opponents? For example, could the president (whoever is in charge at the time) just unilaterally decide to cut Social Security payouts by 5%, for example, and try (successfully or not) to cast the blame for this on whatever party in Congress is refusing to raise the debt limit — urge Americans "if you want full Social Security payouts again, call your congressman"? Would such a decision count as a "default" in the eyes of the credit agencies? I know that this is no way to run a railroad, and the inertia of bureaucracy is probably too large to do such a thing this close to the cliff, but was just wondering why more options weren't on the table.
A lot of public discussions have been sloppy about what “default” or “breach” of the debt limit really means. Two weeks ago, when I moderated a panel discussion on the topic at the Milken Global Conference, we talked about four possible scenarios for what would happen when “extraordinary measures” are exhausted and the federal government runs out of room to borrow, as is currently expected to happen around June 1.
Roughly, the options are:
The Treasury could actually fail to make interest payments due on government bonds.
The Treasury could pay interest on bonds but withhold other payments, like Sean suggests.
The Treasury could use one or more new accounting gimmicks, beyond the existing set of “extraordinary measures,” to continue borrowing and spending without technically issuing bonds whose principal amount would exceed the debt limit. Examples of such gimmicks include premium bonds, consols, and minting the platinum coin. Today, Brad DeLong floats another suggestion: The Federal Reserve could allow the US Treasury to overdraw its spending account.
The Treasury could issue debt in contravention of the debt limit, asserting legal authority due to the 14th Amendment or due to other statutes that conflict with the debt limit.
All these options would have different consequences and pitfalls.
The problems created by option (1) are fairly obvious and would cascade through the financial system. I think a major driver of the blasé attitude about the debt limit on Wall Street and in Washington is driven by an increasing sense, on both sides of the aisle, that (1) won’t happen — that if push comes to shove, we’ll do something else that won’t have as severe a negative effect on the economy or the financial markets.
A lot of conservatives are banking on (2). But there are problems. An obvious one is that many people and entities — from military servicemembers to Social Security beneficiaries to Medicare providers to defense contractors — are relying on payments that could suddenly stop, and that would have negative economic effects. Because of the size of the deficit, you’d need way more than the 5% haircut to Social Security that Sean proposes; there would be lots of missing payments, and that would be a problem. And if the government stopped sending out bills, it’s quite likely we’d get downgraded anyway, because when rating agencies rate a credit, they pay attention to its commitment to all of its obligations to pay, not just the specific bonds being rated. A rating downgrade could have significant knock-on economic consequences because of the prevalence of contracts requiring parties to hold AAA-rated securities. Oh, and also this approach would be illegal and quite possibly technologically impossible, for reasons I discuss in a footnote.
Option (4) is the most popular among liberal commentators, and five senators on the left flank of the Democratic Senate caucus have issued a letter urging Biden to use it instead of agreeing to an austerity deal with Republicans. The problem with this option is that it would require selling a bunch of Treasury bonds into a market where participants would be unsure, at least initially, about whether those bonds were legally valid. Investors might demand a substantial discount to buy them, which would cause turmoil in the financial markets and increase the government’s cost to borrow. On the Milken panel I moderated, Alan Schwartz, the executive chairman of Guggenheim Partners, laid out a strategy to mitigate this problem: the Federal Reserve could buy all the newly issued debt at full price, and then sell previously issued Treasury bonds it holds on its balance sheet into the public markets, so the Fed would bear all the risk that the new bonds could later be declared invalid.
Still, the effect this would have on Treasury markets is unknown, as is how either the rating agencies or the courts would react. It would also draw the Federal Reserve into the middle of a political dispute in a way that could compromise its independence in the future. And probably it’s also illegal? Treasury Secretary Janet Yellen has been out there calling the move “legally questionable.” Personally, I find the legal argument that the 14th Amendment vitiates the debt limit very unconvincing — advocates love to cite the amendment as “The validity of the public debt of the United States… shall not be questioned” without noting that there’s an “authorized by law” within that ellipsis.
As you might now suspect, while my first preference is that the debt limit should be increased legislatively within the next two weeks, my second preference is option (3): Treasury should use one or more gimmicks to keep paying the bills without issuing debt that explicitly violates the statutory limit.
One big advantage of option (3) is that we’ve already done it many times — in fact, we’re doing it right now. Treasury reached its statutory limit for outstanding bonds in February, and we are currently in a “debt issuance suspension period,” in which Treasury uses accounting gimmicks like raiding a federal employee retirement fund to keep paying bills. This is what “extraordinary measures” are, and if we layer on more gimmicks, that doesn’t need to be an earth-shattering event. Just think of it as making the measures even more extraordinary.
I think it’s instructive to look at what happened in 1985. As a congressional impasse over deficit reduction pushed us up against the debt limit, Treasury Secretary James Baker faced a choice: miss an interest payment, or raid the Social Security Trust Fund in order to make the payment. He chose the latter. After the impasse was resolved, the Government Accountability Office issued a report saying his actions were probably illegal but understandable under the circumstances. I’d say so — that workaround worked well enough that people barely even remember it happened. Following the 1985 crisis, Congress passed a law designed to make the “extraordinary measures” process more orderly and clearly legal — also stipulating the Treasury could raid certain trust funds but not the Social Security fund — but at the time Baker did it, it was a wacky idea, and not that different from the untested gimmicks on the table today.
There are some downsides to the gimmicks in option (3) — selling bonds with new and unusual structures designed to avoid the debt limit could have unpredictable effects on Treasury markets; the platinum coin would be litigated and would have uncertain effects on inflation expectations. But while these are problems, they’re smaller than the problems that would ensue from missing an interest payment. Plus, as with the 14th Amendment option, the Fed could smooth matters through the use of its own balance sheet, including by buying up the weird new bonds (such as consols) if necessary.
And this brings me back around to why nobody seems to really have their hair on fire. I think people — lawmakers and market participants — mostly agree with me that the US will not miss any bond payments even if Congress fails to raise the debt limit.
I don’t think everyone agrees with me about why bond payments won’t be missed — on that Milken panel, former White House Chief of Staff and OMB head Mick Mulvaney said he thought Treasury would (despite its technological protestations) prioritize bond interest payments and stiff some other payees. Many liberals believe the Biden administration will flip its position and assert a 14th Amendment authority to continue bond issuance if necessary. Even if the administration and the Fed are prepared to use one or more workarounds, they’d be unlikely to telegraph that in advance — the administration doesn’t want to undermine its ongoing effort to get a legislative deal, and the Fed surely doesn’t want to wade into this political fight if it doesn’t have to.
I want to be clear: For reasons I’ve laid out above, these scenarios in which bond payments continue are not necessarily without economic cost. But it’s plausible to think their economic costs would be modest enough that it’s rational for lawmakers in both political parties to be willing to bear them if they think it will help them get to a better policy outcome in the eventual budget negotiations. The prospect of failing to raise the debt limit isn’t necessarily dire enough for conservative Republicans to agree to continued spending at current levels or for liberal Democrats to agree to sharply cut spending or further restrict access to welfare programs. And that’s how it can be that we’re in this crazy position — the position isn’t actually as crazy as it looks.
All that said, there has to be some sort of budget deal this year anyway, and I think it is good that lawmakers are trying to leverage that necessity to get a debt limit increase passed. If it gets done promptly, we’ll never need to find out what any of these creative solutions will cost the economy.
Do you have more questions you’d like me to answer? Email them to email@example.com — all topics are welcome, from cooking and dating advice to travel tips to voting strategically to why everything seems like it sucks.
Under both Republican and Democratic administrations, the Treasury Department has insisted that it lacks the technological capability to pick and choose which payments to make in order to manage a debt limit crisis. There is also no legal authority for Treasury to pick and choose — Congress has passed laws instructing the executive to spend certain funds. Sending out Social Security checks is not optional. Some congressional Republicans have periodically proposed laws to authorize payment prioritization in the event of a debt limit breach, but they haven’t been enacted. This illegality is one of the key arguments for the other strategies — even when there are questions about the legality of other options, failure to pay is also illegal.
I don't have strong feelings about the 14th amendment solution, but aren't all debts of the US "authorized by law"? Doesn't the federal government, at least in theory, need legal authority to incur any debts?
So there is no Chesterton's Fence-type argument whatsoever for the debt ceiling to exist? Has it ever led to any reflection about government size/spending, or is it merely an occasion for the exercise of cleverness such as minting a big coin* or further mining the word salad of the 14th amendment? If the latter, conservatives should indeed be quite prepared to give up on it. There is no sense in which fakery serves the nation's interests; better to forget whatever the principle was, than to preserve it only to mock it, if it had any validity.
*Rather poignant that Google chooses to illustrate the dimensions of this shiny totem by invoking seven Saturn V rockets.