The Debt Limit in Baseline Hell

As both of my regular readers know (hi mom!), I am a connoisseur of what I call “baseline hell.” “Baseline hell” is that infernal mode of debating about whether some state of affairs is “neutral” or “normal” rather than “good” or “bad.” It flares up everywhere — in takings doctrine, First Amendment doctrine, Free Exercise doctrine, etc. People land in baseline hell because they harbor the forlorn hope that they can sidestep contentious normative questions in favor of some elusive consensus about what constitutes the “normal” state of affairs — the “neutral” baseline from which departures are presumptively disfavored. Alas, disagreements about what constitutes a Neutral Baselines are just as intense as disagreements about what constitutes a Good World. The former debates unfortunately also have the disadvantage of being disputes about completely arbitrary lines.  The result is diabolically tortuous and eternal squabbling about “neutrality” — aka something that does not really exist.

@Pwnallthings (aka Matt Tait) has a substack nicely illustrating how the current crisis over raising the Debt Limit can land us in fiscal Baseline Hell. Tait argues that the debt limit is a bad thing, because it “provides an artificial negotiating advantage to those who want to extract concessions from the governing party.” What’s so “artificial” about this advantage? Tait asserts that it is artificial to require “passing new legislation” to “maintain[] the status-quo,” because such a requirement “is an inversion of the traditional ‘schoolbook’ model of US governance” under which “major changes to the status-quo would normally require legislation passed by the Congress and signed by the President to take effect.” According to Tait, “[t]hese timebomb-style games-of-chicken have always been structurally bad,” because “[p]assing legislation in the United States is constitutionally hard on purpose” given that “it is normatively good for big changes in the governance status-quo to require debate and agreement in the political branches.”

Tait’s use of the adjective “artificial” is a sure sign that he is venturing into the first circle of Baseline Hell. The idea that some proposals are “artificial” departures from some “natural” state of affairs is the classic method of shifting burdens of proof without really defending the merits of a position. Like all such Baseline arguments, the definition of what’s “natural” and “artificial” is buried in murky adjectives — here, “new,” “major,” and “big.” As I shall argue after the jump, these terms beg all of the important normative questions. That argument, however, requires a bit of a deep dive into the non-delegation problem lying at the heart of the debt limit. After the jump, I’ll describe (1) how Wilson’s and other Southerners’ ambivalence about executive power likely limited the switch to full executive management of debt and (2) why this semi-executive system of marketing federal securities invites people like Tait to get lost in baseline hell rather than debate the merits of executive power versus legislative control of the bond market.

The Incomplete Managerial Revolution of 1917 in Debt Management

As Conor Clarke notes in a recent paper that people should now be frantically downloading, the debt “limit” is really part of a necessary debt authorization, because the executive has no inherent Article II power to pledge the credit of the the United States without a delegation from Congress. According to Kenneth Garbade, Birth of a Market: The U.S. Treasury Securities Market from the Great War to the Great Depression, Congress tended prior to World War I to micro-manage the executive’s management of federal bonds, specifying quantities, interest, maturity, and convertibility in some debt-authorizing statute. The massive spending required to fight the Great War, however, required bond sales so large that the terms under which the bonds would be marketed could not be so meticulously spelled out without running the risk that a bond auction might fail. The Liberty Bond statutes, therefore, conferred much greater discretion on the Secretary of the Treasury to set the terms under which federal securities would be marketed.

In particular, The Second Liberty Bond Act of 1917 conferred on the Secretary of the Treasury the power generally to issue debt within an overall ceiling on the entire aggregated debt of the United States. That ceiling, now codified at 31 U.S.C. 3101 and adjusted upwards periodically by Congress, is what is causing all of the trouble today. The Liberty Bond Revolution was part of a more general early twentieth century managerial revolution that shifted power from Congress to the executive. This revolution was driven in part by wartime necessity but also, as described by Jesse Tarbert’s outstanding new book, When Good Government Meant Big Government, by ideology. A group of mostly Republican politicians and writers including William Howard Taft, Henry Stimson, and Herbert Hoover, pushed for presidential rather than congressional oversight and control of the federal government in the name of national efficiency. Calling for the government to be run like a business, these reformers noted that corporations were managed by a chief executive, not the board of directors. The centerpiece of these reformers’ program was a presidentially controlled budget to replace each federal agency’s independent budget requests to particular congressional committees. The Liberty Bonds, however, were in the same vein of enlarging presidential capacity to manage the executive branch free from rigid statutory standards.

This managerial revolution in debt management was incomplete because of the ceiling on aggregate debt. Clarke notes that Congress in 1941 greatly enlarged Treasury’s managerial discretion by combining all debt, regardless of date of maturity, into a single aggregate ceiling, thereby giving Treasury power to choose the mix of instruments that would minimize borrowing costs and maximize price. But the ceiling still required the executive to go to Congress periodically to increase the total amount of permitted debt whenever the indebtedness necessary to cover appropriations exceeded the last statutory ceiling. Total managerial discretion, by contrast, would allow Treasury to treat congressional appropriations as the only limit on the quantity of bonds that the federal government can sell.

Why did not Congress go all the way, either in 1917 or 1941, by making appropriations the only limit on outstanding debt? I have only an educated guess: Southerners in both the Wilson and Roosevelt Administrations, backed by Southerners in Congress, were reluctant to remove that congressional restraint on the executive. Treasury was run by Southerners (William Gibbs McAdoo and Carter Glass) who, in other contexts, were resistant to centralization of power, resistance that was completely in line with the “New Freedom” ideology of Woodrow Wilson, their fellow Southerner. The double security of both appropriations and the debt ceiling must have been appealing to such skeptics of presidential power. As Ira Katznelson and Joanna Grisinger both explain, Southerners in Congress in the 1930s and 1940s were deep skeptics of the enlargement of presidential power, especially after Robert Wagner (NY) and Edward Costigan (CO) introduced an anti-lynching bill in 1934. (Tarbert notes that the Dyer anti-lynching bill also doomed the managerial revolution in the 1920s). Whatever the reasons, the choice to limit Treasury’s authority to issue debt was a deliberate choice. Prior to World War I, the norm throughout American history, after all, was that the executive had to go to Congress for permission to pledge the United States’ credit even after appropriations were made. The two World Wars enlarged congressional authorization for Treasury to sell bonds — but Congress never handed over the entire business of deciding how much bonded indebtedness the government would incur to the Secretary of the Treasury. For whatever reason, that was simply too broad a delegation of power for Congress to tolerate.

Welcome to Bond-Marketing Baseline Hell

With that long windup, here’s the pitch. It seems to me that Tait is using the ideal of a total managerial revolution in debt management to serve as the “neutral” baseline for assessing the “artificiality” of the debt ceiling. Without explicitly saying so, he treats the Secretary of the Treasury’s decisions to sell bonds as white noise — background conditions that define the normal baseline and do not change the status quo.

Here’s how Tait bakes bond-selling into his baselines. He argues that the debt ceiling “artificially” departs from “the traditional ‘schoolbook’ model of US governance” because the ceiling forces the federal government to default on some set of creditors to whom promises of payment were made without a vote of Congress. Such a default is a “major” (or “big” or “new”) change in policy that ought to occur only by a statute, not by mere federal inaction, because “it is normatively good for big changes in the governance status-quo to require debate and agreement in the political branches.”

This “no-stiffing-creditors-without-a-congressional-vote” argument, however, assumes that Treasury’s decision to sell bonds exceeding the debt ceiling somehow does not constitute a “big change[].” After all, getting rid of the debt ceiling empowers the Treasury to issue more debt without any further vote of Congress. Why is not this decision to increase the federal government’s bonded indebtedness also a “big change[] in the governance status-quo … requir[ing] debate and agreement in the political branches”?

Tait does not ask, let alone answer, this question. He assumes without any argument that Treasury’s marketing new tranches of federal bonds without any statutory authorization is not a change in the status quo but instead just part of the natural order of things. The assumption is natural enough, given specific changes in how Treasury marketed securities dating from 1982: As Kenneth Garbade explained in a 2007 article, Treasury instituted a set of reforms between 1975 and 1982 to make bond sales more “regular and predictable” after some ad hoc bond offerings in the 1970s disrupted the bond market. Nowadays the federal bond-selling machine hums along without people like Tait noticing that very complex market operations are taking place that affect future generations with “major” fiscal burdens. Whenever Treasury sells a tranches of bonds, however, it is changing the status quo in ways that many people, including some ornery Republican members of Congress, think is a pretty “major.”

Tait might reply that the fiscal burden on future generations was already incurred when Congress made an appropriation without sufficient revenue to pay for the expenditure. This response, however, proves too much: By that logic, the IRS ought to have the unilateral power to increase taxes to cover expenditures without any further legislation. After all, tax revenue avoids default on current obligations just as effectively as bond sales. Changes in taxation, however, seem “major” because they happen rarely, while Treasury’s bond-selling operations have been humming along for decades without anyone much noticing. Moreover, the existence of an appropriation does not dictate that revenue must be produced either through bond sales or taxation.

The third alternative is simply to bilk, or at least delay paying, some of the federal government’s creditors. The significance of such defaults would vary with the creditor being deprived of prompt payment: As Clarke has pointed out in a sharp blog post, “defaulting” on “debt” covers a range of outcomes, from delaying payment of a federal employee’s salary for a few days to failing to make timely payments on the United States’ public debt securities. None of these outcomes is great, but it is not obvious that every single one of them is a catastrophe. Tait assumes that being a deadbeat is a “big change” while selling more bonds is not. Maybe he is right — but he never bothers even noticing that this is the sort of distinction that he is implicitly drawing. That is indeed how baseline arguments generally run: Proponents of baselines treat some aspects of the status quo as “natural” and some as “artificial,” leveraging the former (e.g., selling billions in bonds annually) to condemn the latter (e.g., defaulting on debts). Opponents of those baselines argue that the “natural” parts of the status quo are not really natural after all. All the while, some abstruse exercise in characterization about what is “artificial” and what is not gets in the way of a debate about the actual normative merits of the status quo. That impediment to the important normative issue is the essence of Baseline Hell.

 “Eliminate the Ceiling” versus “Mint the Coin”: Alternative Ways of Side-Stepping Congress?

Suppose we decide to climb out of the infernal pit of Baseline Arguments and instead focus on the normative merits without focusing on what is natural, artificial, normal, or abnormal. One might compare such baseline-free discussion of these merits to Purgatory rather than Hell: The discussion might be difficult or unpleasant, but least there is hope of progress to some sort of meaningful agreement. What exactly are the normative questions at stake with federal debt?

I suggest that at the heart of the matter is congressional oversight of fiscal burdens. The idea behind the debt ceiling is that the Congress should periodically assess the total quantity of indebtedness incurred by the federal government’s spending. Anita Krishnakumar laid out the basic argument in favor of such periodic congressional oversight back in 2005: The ceiling, in her words, “act[s] as a catalyst for budget-reform and budget-balance measures aimed at reducing national borrowing.” The weakness of this argument, of course, is its implicit assumption that Congress is not paralyzed by partisan grandstanding. To focus on the question of congressional capacity, however, is to ask the right question: Can Congress Govern?

If we focus on this question rather than baseline distractions, then we can also see why Tait’s “Mint the Coin” solution evades the problem that a debt ceiling tries to solve. The problem is that Congress is inclined to shirk its constitutional responsibility to define how revenue should be raised. Since George “Read My Lips” Bush’s defeat in 1992, it has been conventional wisdom in both parties to steer clear of proposing revenue measures, because raising revenue is unpopular. By confronting Congress with the threat of the federal government’s defaulting on its debts, the debt ceiling ideally forces Congress to take responsibility for raising the revenue required by its appropriations. “Minting the Coin” plainly does not serve this congress-forcing function.

I will not rehearse Tait’s admirably clear exposition of what it means for the Secretary of the Treasury to mint a trillion dollar coin and deposit that coin in the Federal Reserve to be used as an asset on which the Treasury can draw to pay the debts of the United States. The important point is that, because it relies on an existing statutory provision, Tait’s argument fails to address the purpose of the debt ceiling: Force Congress to confront publicly the fiscal or monetary consequences of its spending decisions. What’s worse, Tait’s argument violates his own principle that “it is normatively good for big changes in the governance status-quo to require debate and agreement in the political branches.” Conferring unlimited power on Treasury to mint money on which the federal government alone can draw to pay the federal debts might be a great idea, but it is plainly not the idea behind the statutory provision invoked by Tait. That provision, 31 U.S.C. 5112(k), simply provides that

The Secretary may mint and issue platinum bullion coins and proof platinum coins in accordance with such specifications, designs, varieties, quantities, denominations, and inscriptions as the Secretary, in the Secretary’s discretion, may prescribe from time to time.

Buried in a longer statutory provision dedicated to specifying in excruciating detail the characteristics of commemorative or collectible coins, the provision was obviously enacted without any thought that it would allow the Secretary of the Treasury to mint money to pay federal debts. Assuming that anyone would ever have standing to challenge such a use of this statutory provision, it is almost inevitable that SCOTUS would strike down the “Mint the Coin” strategy faster than you can say “elephants don’t fit in mouseholes.”

Indeed, Tait’s argument, which he admits is “gimmicky and undignified,” actually makes the Major Questions Doctrine look good. After all, as Tait says, “it is normatively good for big changes in the governance status-quo to require debate and agreement in the political branches.” Conferring massive power on Treasury to deal with revenue questions that Congress is too divided to confront looks like a Big Change. Unless you believe that someone in Congress had the idea that the Platinum Coin provision could be used in the way suggested by Tait, that institutional change never received any such debate and agreement in the political branches.

Tait, in short, contradicted his own “normatively good” principle in defending an interpretation of a statute that lets Congress off the hook. I myself am not so sure that this principle is a decisive objection to the “Mint the Coin” plan. If Congress has become completely dysfunctional through affective polarization, then maybe we need to re-delegate revenue decisions to the executive, giving to Congress the option of vetoing such decisions ex post. I am more certain, however, that, by starting with baseline distractions about what is artificial and natural, Tait made it more likely that he would miss the critical nondelegation question at stake: In Baseline Hell, the smoke gets in your eyes.