Recent Posts

Wednesday, November 21, 2018

The U.S. Debt Limit (Preliminary Primer)

The debt limit in the United States is currently not an object of worry, but it represents one possible avenue to default. From the perspective of a non-American, it is rather difficult to understand how such a strange custom could arise. This article outlines very briefly the history of the debt limit, and then moves to discuss the risks associated with it. This issue underlines the argument that default risk in floating currency sovereigns is political risk, not financial.


For my background comments I am relying heavily on the report "The Debt Limit: History and Recent Increases" by D. Andrew Austin, of the Congressional Research Service. (Thanks to Rohan Grey for sending me the reference.) I will refer to it at various points later. The real meat of the discussion of the debt limit is going to be legal: what happens when it is hit, and the Treasury runs out of work-arounds? From my perspective, all I can safely write is that readers should seek their own legal counsel on such questions.

Instead, I want to focus on the general principles, and how investors should judge the risk of something going horribly wrong as a result of the debt limit.

(This article is a very preliminary draft of a section that would appear on a potential book on default risks; it will be fleshed out with more references once my line of thinking has solidified.

What is the Debt Limit, and Why Does it Exist?

The debt limit is relatively straightforward: the U.S. Congress passes laws that sets the limit for debt that can be issued by the U.S. Treasury. (There are some categories of debts that are excluded from the limit; from the Austin article , 0.5% of debts were not covered by the limit in 2015.)

What makes this limit a source of trouble is that the debt limit is set in legislation that is distinct from the laws that set tax rates and spending budgets. If the Federal government runs a large enough deficit -- as determined by laws set by Congress -- the associated debt could surpass the limit, which is set by the very same Congress.

From the perspective of a Canadian, and presumably other countries that follow the British parliamentary system (I'm familiar with the system in the United Kingdom), this seems incredible. The difference for a country like Canada is straightforward: the Minister of Finance is a key cabinet minister of the government, and the Ministry of Finance (that executes the budget planning and almost entirely handles the financial operations for the government[1]) reports directly to the minister. The budget is a unified plan, and includes discussion of borrowing operations. Although the parliamentary system allows for some freedom for lawmaking by backbench members of parliament (which is no longer true in Canada), the budget is a "take it or leave it" proposition from the cabinet. If the budget is not passed, the government will dissolve parliament and call for new elections. Furthermore, the Canadian Senate (counterpart of the House of Lords) has almost no right to touch spending bills.[2] The end result is that the budget is always going to be a unified plan, prepared by responsible adults, and will be internally consistent.

Conversely, the United States has a mandated separation of powers. The Congress passes laws, and the Executive Branch (which reports to the President) (unsurprisingly) executes the laws. The Treasury -- which manages borrowing -- is in the executive branch. It has no choice but to follow the laws passed by Congress, which may or may not be internally consistent.

Furthermore, the Congress is split between the Senate and the House of Representatives, and both bodies are involved in lawmaking. To top it off, various committees are involved in the determination of fiscal policy.

To quote Austin,
While the budget process provides Congress with one means of controlling federal spending, the debt limit may provide a different sort of leverage that is not redundant. Congress ordinarily delegates work to its committees. The Committees on Appropriations have special responsibilities regarding discretionary spending, and authorizing committees are generally responsible for mandatory program spending decisions, while Committees on the Budget are tasked with drafting an overall budgetary framework that specifies aggregate levels for federal spending and taxation. While those committees often incorporate views of other committees and Members, measures involving the debt limit often provide individual Members not belonging to those committees with a separate instrument to influence federal fiscal policy.
In other words, it gives politicians who have not snagged a committee membership some say in fiscal policy.

As outlined in Austin's report, the debt limit represents an improvement in Congressional lawmaking. Previously, Congress authorised specific debt issuance; by creating the unified limit, the Treasury was given the flexibility to manage debt issuance more efficiently, while preserving Congress' constitutional responsibilities.

What Happens When the Limit is Hit?

The conventional framing of what happens when the United States is running a deficit and hits the debt limit is straightforward. (The report by D. Andrew Austin accepts conventional framing.) Since the Treasury has very limited ability to sell assets, it needs to issue debt to "finance" the deficit. (There are limited workarounds in practice, as discussed below.) When it can no longer issue debt because of the debt limit, it cannot meet all the legal spending obligations given to it by Congress. A standard argument is that the Treasury cannot prioritise spending on its own, so it would be forced to miss all of its contractual obligations. Principal and interest payments on Treasury securities are such obligations, and so a default on bond payments would be required.

(I have seen arguments that the Treasury could prioritise Treasury security payments, and so a default is not needed. I will turn to that point later.)

The strange part of the discussed scenario is that the default only arises because of inconsistent laws passed by the same branch of government. Since the debt limit itself exists solely because of law, any such default would have to be considered a political decision.

One concern with the conventional framing revolves around the underlying monetary operations. For simplicity, we will assume that the Federal government spends solely by issuing cheques (checks in American); no new-fangled electronic transfers. The Treasury has a "chequing account" at the Federal Reserve, with a balance. When a Treasury cheque is cashed, the Federal Reserve drops the Treasury's account balance by the amount of the cheque, and it increases the balance of the bank at the Federal Reserve. That is, it creates "bank reserves" by the amount of the cheque. The private bank has a new asset that it holds against the increase in the deposit of its customer (that cashed the cheque).

The conventional assumption is that the Federal Reserve will not honour any cheques issued by the Treasury that result in it having a negative balance in its account. That is, it will returned on the grounds of non-sufficient funds (or "bounced"). 

The question then arises: does the Federal Reserve have the legal authority to bounce Treasury cheques? If so, which ones? Let's take a simple example. Assume that the Treasury has a balance of $100, and it has two cheques for $75 cashed in the same day. Will it bounce one, or both? If the Federal Reserve does not have the authority to make that decision, the end result is that both cheques have to be honoured, and the Treasury will end up with a $50 overdraft. Such an overdraft is not unprecedented in central banking history. Even if such an overdraft is against some law or regulation, bouncing the cheques is probably against other laws or regulations. Congress will have to clean up the mess created by its own internally inconsistent laws.

An similar line of attack is the fourteenth amendment of the American Constitution. Austin's summary of this is:
Some have suggested that the Fourteenth Amendment (Section 4), which states that “(t)he validity of the public debt of the United States ... shall not be questioned,” could provide the President with authority to ignore the statutory debt limit. President Obama has rejected such claims, as have most legal analysts.[3]
I will not pursue this argument, although I would note that it would give most conventional economists the vapours. It implies that the Federal government can completely abolish issuing debt. For the rest of the article, I will assume that this line of argument is not being pursued, and that the Treasury needs to keep a positive balance at the Fed at all times.


There are a number of ways for the Treasury to juggle its accounts and keep debt below the limit. These extraordinary measures have been a feature of the various political standoffs around the debt limit historically, as documented in the Austin article. They are mainly accounting exercises that have almost no effect on the real economy. They stretch out the timeline before the limit is binding on policy, and allows for political drama and posturing.

The next method is for the Treasury to delay payments that do not have a fixed time limit. That is, lengthen payment times on accounts payable. This is costly, as this incurs an interest expense that the government is legally required to give as recompense. Such a step will have an effect on the real economy, as vendors will end up with less liquidity than planned, and one can imagine this being magnified into becoming a general flight to liquidity, which is a characteristic of financial crises.

(One technical point about such developments is the question of sovereign credit default swaps. Could such steps trigger a sovereign CDS? One would not think so, but the legal risks associated with such instruments is a reason to be very cautious when selling protection, particularly if you are not getting compensated properly for selling that insurance.)

Finally, there are novel legal theories, such as the "trillion dollar coin." The idea is straightforward: the Treasury issues coins, and there is no cap on denominations. The Treasury just needs to mint a coin with a face value of one trillion dollars.[4]  Since there is no relation between the face value of coins and the value of their metal content, there is no economic law against such a move. The Treasury then sells the coin to the Federal Reserve (like all its other coinage), and its balance will increase enough to allow all its cheques to clear easily

Historically, these workarounds allowed the Treasury to stick at its cap for a known period of time. This allowed everyone to posture and then cut a deal.

Is This a Real Risk?

It is not particularly novel to note the importance of big donors in modern politics. In general, the donor class will not benefit from a potential collapse of the financial system. Therefore, one would argue that the default scenario[5] should be that a deal is cut at the last minute.

I am somewhat more cautious. (As a disclaimer, I own Treasury securities via exchange-traded funds at the time of writing, so I am obviously not deeply concerned about default risk.) The problem I see lies in what I view as unhealthy trends in politics. These trends are there in other countries, but are accentuated in the United States.

Anyone allergic to political discussion might consider stopping reading after this sentence, but one must note that it is difficult to discuss political risk without in fact discussing politics. 

It is relatively common to argue that politics is more polarised in recent decades. That is a point that could be debated. If we do accept that politics is indeed more polarised, then there is a greater willingness to risk a crisis to force a change of direction. One can point to both parties using the debt limit as a stick to push their agenda.

However, I would argue that the risks are more concentrated within one political party: the Republicans. The demise of the "country club Republicans" has led to a feistier Republican base. (My own political views are best described as being in the free market wing of the Canadian prairie populism, which gives me some overlap in economic views with Eisenhower Republicans. I do not pay much attention to current politics, but closely observed the final demise of the country club Republicans in the early 1990s.) One result of the polarisation of the Republican base was a greater willingness to entertain a fairly radical strategy: use the debt limit as a means to create a de facto "balanced budget amendment." If the amount of debt is not allowed to rise (and an Treasury overdraft not permitted), the budget would have to be balanced. 

Are there free market ideologues willing to risk a financial crisis in order to cripple the welfare state? Almost certainly, although they would not be happy to use that phrasing; they will typically argue that crisis can be avoided by bouncing the cheques for welfare recipients, while making all the payments due to bondholders.

Any lender has to make judgements about borrowers. Two factors are paramount: capacity to pay, and willingness to pay. There is little doubt about the U.S. Federal Government's capacity to pay; the willingness to pay is where the problem is going to lie.[6] 

At present, the risks seem small (particularly with a Republican President). That said, political developments in recent years have been throwing up many negative surprises.

Concluding Remarks

The Congressional debt limit is a somewhat unique feature of American politics. The risks posed by its existence should be minimal, but that assessment is reliant on a willingness for compromise among the political class.


[1] The Bank of Canada manages the auctions of Government of Canada bonds, but as they always emphasise, they do so on the behalf of the Ministry of Finance. 

[2] That's an assertion based on my memory. Will need a reference, or cut the text.

[3] His citation is as follows. Adam Liptak, “The 14th Amendment, the Debt Ceiling and a Way Out,” New York Times, January 24, 2011; Remarks by the President at University of Maryland Town Hall, available at For a legal analysis, see CRS congressional distribution memorandum, Whether the Public Debt Clause Authorizes the President to Borrow Money in Excess of the Debt Ceiling, December 21, 2012, by Kenneth R. Thomas.

[4] Best read with a Dr. Evil voice.

[5] Ha ha.

[6] I believe that was Warren Mosler's line during the last Debt Ceiling Crisis.

(c) Brian Romanchuk 2018


  1. Brian,
    Loved your post as it attempts to focus discussion on the realities of the process and how things might play out. It made me do some thinking and question asking here at work to get a better idea how FED settlement works when the payments are coming from the Treasury.

    My gut says that the FED would honor all of the payments; that it has to. The caveat I’d add is that it may not debit the Treasuries account, but it would have to debit something and would have to make good on the payment.

    Here’s why I say that… Let’s take the example of a person (A) who deposits/cashes a check from another person (B) who does not actually have the funds. What happens… the check gets credited to person A’s account. The bank sends the check for clearing and settling. The FED takes all the checks each day from A’s bank and B’s bank out and nets everything. That includes netting the accounts at the FED of both banks. Bank B will at some point will determine that person B does not have funds to cover that check and will send the check back through the process in reverse.

    The FED, in this situation, is just a middle man. In the case of the US Treasury the FED is also one of the institutions directly involved. Also, there are orders of operation that would likely leave the FED paying the bill regardless. As I understand it, the FED performs account settlement at night. So in the case of say Social Security payments, the receiving bank obtains files during the day which are used to credit customer accounts. The actual settlement of payments between the US Treasury and each bank (the netting of payments to banks and to the Treasury) occurs at night - afterwards. This is why the FED can’t be the arbitrator of individual payments, because it only first knows what payments are being made; it’s not until the end of the day before the FED knows actually how much the Treasury is short.

    If a normal bank overdraws its FED account, the FED automatically brings it to zero and all payments continue. This is operationally how the Discount Window works. If the US Treasury has zero in its account, the FED is not supposed to provide it overdraft by law. However, the FED also, “shall maintain long run growth of the monetary and credit aggregates commensurate with the economy's long run potential to increase production, so as to promote effectively the goals of maximum employment, stable prices, and moderate long-term interest rates.” Which means it needs to ensure those payments get completed/settled because if it doesn’t then the financial system is likely short BILLIONS, probably $10’s of billions, of dollars every day. If it doesn’t just clear the payments that shortage would show up in rising Federal Fund Rates or banks piling into the discount window and or banks not meeting their capital ratios or acting differently because they are going to be taking a capital hit until the situation is resolved. The FED can’t maintain its mandate and not clear the payments one way or another.

    1. Thanks. I deliberately simplified things, and I also wanted to explain the worry. If I argued that default is impossible, why does everyone care about the debt ceiling?
      (This is Brian. Google comments have been giving me grief for logging in on my tablet.)

    2. LOL!!!

      “why does everyone care about the debt ceiling?”

      That’s like asking how do you convince a flat Earther that the Earth is round.

      While I’d still agree policy error or political choice could still yield default, for most people they just don’t understand the mechanisms and just assume it’s a red line that with hard coded features – it’s not and again that’s what I loved about your article it makes you think beyond the political babble.

      Oh.. And for some, it’s politically expedient to keep caring about the debt ceiling.

  2. This comment has been removed by a blog administrator.


Note: Posts are manually moderated, with a varying delay. Some disappear.

The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.

Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.