What Exactly Is a "Default"?

The term gets thrown around loosely to refer to different concepts, but with very different implications.

Negotiations over the debt limit are still ongoing, and we are starting to hear a lot about the imminent prospect of a "default." The term gets thrown around a lot in the press, usually attached to words like "unprecedented" and "catastrophic." But the term creates confusion wherever it lands. Some of this reflects garden-variety misunderstanding. But I worry that some of it reflects a kind of intentional slipperiness by motivated actors who want to frame debt-limit issues a certain way.

There are at least two senses in which the term "default" gets used. The first, and probably the more intuitive usage, refers to a failure to make payments on the public debt securities of the United States, like bonds and Treasury bills. On this point I am not Pollyanna: If that kind of default happens it is likely to be a really bad thing. But as I'll explain more below, I don't think that outcome is particularly likely. (Though it's not impossible.)

The second sense in which the word "default" gets used is something like: "Any failure by the United States to make any payment in full and on time." I confess I find this usage of the term spectacularly unintuitive. But, putting intuitions aside for the moment, this second usage also covers a huge potential breadth of potential outcomes. What payments, exactly? And for how long? There is a big difference between a subset of federal salaries getting paid a day late (perhaps against the backdrop of a deal that has been struck but is still grinding its way through bicameralism and presentment), and large swathes of the federal budget going unfunded for many weeks as Congress and the President continue to flounder.

These distinctions are important. But the difference between usage one and usage two—and the important differences of degree within usage two—are sometimes obscured. Consider how Treasury Secretary Janet Yellen discussed the concept of default a couple of weeks ago: "Whether it's defaulting on interest payments that are due on the debt or payments due for Social Security recipients or to Medicare providers, we would simply not have enough cash to meet all of our obligations." Yellen is generally referring to the broader and less intuitive concept of "default" as any delayed payment—but also hinting that hey, maybe we'll default on securities in that more technical sense too. I find this conflation exasperating.

A "default" in the first and more intuitive sense of the term is terrible because the market for Treasuries is such an important part of the global financial system—and its reliability (including its reliability as a short-term investment) is one of the reasons the U.S. government can borrow relatively cheaply. But I think Treasury securities are still reliable even as X Date approaches. First, there's the possible constitutional backstop of the 14th Amendment: Even if you are skeptical (as I am) that the Public Debt Clause gives clear instructions, a failure to service the national debt when it was possible to do otherwise would at a minimum raise a profound and avoidable constitutional question. (Although, for reasons I've offered earlier and elsewhere, I don't think 14th Amendment provides a basis for ignoring the debt limit.) Second, as Kristin Shapiro and I argued in the Journal, Treasury can roll over the principal on the debt—as securities mature and create headroom, new ones can be auctioned. If Treasury plans ahead—and if you buy the 14th Amendment argument it has a constitutional obligation to plan ahead—it can ensure it has enough tax revenue to pay for interest. (In 2020, for instance, the government's net interest outlays were $345 billion, and it received about $3.4 trillion in tax revenue.) For a default on Treasury securities to take place, then, something unforeseen would have to take place—like Treasury auctions failing or tax revenue collapsing.

I don't know the odds of those disasters happening. I don't think anyone does. (Unlikely in the short run, unknown in the long run?) But I suspect those outcomes—especially auction outcomes—are partly a function of what the Administration says about its intentions. It strikes me that the obvious way to both honor the Public Debt Clause and reassure markets would be to say that Treasury has every intention of honoring the securities that compose the "public debt" (as that term appears in both the 14th Amendment at the debt limit statute, 31 U.S.C. § 3101), and that Treasury believes it's possible to do so. But there may be some unfortunate tradeoff between playing hardball in political negotiations and providing maximum reassurance to the market. Again, exasperating.

What about "default" in the second sense of the term—that is, any delay in any government payment? That is not good either. Even though I wish the government spent less at the margin, I don't like the standard debt-limit practice of Congress appearing to make various forms of spending legally obligatory and then providing questionable revenue to achieve those spending goals. Moreover, the potential scale of spending delays in the weeks that come is vast. But it's hardly "unprecedented" for government payments to be delayed—I mean, the whole reason we have a law called the Prompt Payment Act to enforce the government's prompt payment is precisely because government payments aren't always prompt! I certainly thought the government owed me a salary for working during the 35-day shutdown during the Trump Administration (this was before the Government Employee Fair Treatment Act became law)—but got paid only after a delay.

More generally, the degree to which spending delays will be damaging will depend on how far past X Date we go. "X Date" refers technically to the first day on which the federal government can't make a payment, and Treasury makes many thousands of payments every day. The mismatch between Treasury's available incoming cash and each day's thousands of required payments will continue to grow each day past X Date that we go. Short delays in payments will eventually become long delays in payments—and the case will look less and less like a short government shutdown and more and more like an unprecedented disaster.

But I worry that it's become too easy to simply hand-wave on the details and treat "X Date" as the bright-line day on which Treasury securities collapse and no longer become the stable investment of decades past. I still don't see a good basis for that line of thinking.