The debt ceiling began in 1917 to make borrowing easier, not harder

The debt ceiling is the only wall in Washington that was built to make the house easier to enter. Everyone talks about it as a fence around the Treasury. It began as a doorway.

That is not the version you hear in a budget fight, where the ceiling is cast either as the last line of fiscal defence or as a hostage-taking device that threatens the credit of the United States. Both stories skip the origin. Let me tell it straight, in order, and then weigh the conservative argument for keeping the limit against the argument that it only fights over bills already incurred.

Define the limit

The debt limit, or debt ceiling, is a figure written in federal statute that caps the total amount the Treasury may owe at any one time. It is not a spending limit and it does not authorize any spending. It is a limit on the borrowing needed to pay for spending that Congress has already enacted through appropriations and through laws that set benefits. Nor does it limit the deficit; it limits the stock of debt that accumulates when deficits persist.

That distinction is the crux of the whole argument, so keep it close. The decisions that create debt are the ones about taxes, programs and appropriations. The debt limit is a separate vote on whether the Treasury may borrow to meet the bills those decisions created.

Before 1917: Congress approved each loan

Under the Constitution's Article I, Section 8, Clause 2, Congress holds the power "To borrow Money on the credit of the United States." Before 1917, Congress often exercised that power directly and specifically. It authorized borrowing for particular purposes, such as building the Panama Canal, and the Treasury could not issue anything beyond what the specific act allowed.

That arrangement was safe, and it was slow. In ordinary times the pace did not matter. In a war it did.

1917: the Second Liberty Bond Act

When the United States entered the First World War in April 1917, the government suddenly needed to borrow on a scale without precedent. The first Liberty Loan Act of April 1917 authorized a bond issue for the war. By the autumn the Treasury needed more and needed flexibility. Congress passed the Second Liberty Bond Act in September 1917, and it did something new. Instead of relying only on one authorization after another, it included an aggregate limit on federal debt as well as limits on specific kinds of debt issues, which left the Treasury room to manage its borrowing inside limits Congress had set in advance.

Read that carefully. The effect of the new structure was to give the Treasury more freedom. Before 1917, Treasury often waited for Congress to approve each borrowing. After 1917, Treasury could act within a ceiling that Congress set in advance. The ceiling was not a brake on a driver who wanted to go fast. It was a loosening of a leash that had been shorter. Congress set the outer boundary, and the Treasury ran the day-to-day operation. It is the difference between approving every purchase and giving a manager a budget.

1939: one limit for the whole debt

Through the 1920s and 1930s the structure became complicated, with different limits for different types of securities. In 1939, Congress passed the Public Debt Act, which established a single aggregate limit on the total debt outstanding, set at $45 billion. This is the ancestor of the modern debt limit: one number that applies to almost everything the Treasury owes.

Notice the motive again. The change in 1939 was about administrative simplicity and Treasury flexibility. The country was nearing another war, the government was borrowing heavily, and an aggregate limit let the Treasury manage its borrowing without returning to Congress for each kind of security. The limit rose repeatedly during and after the Second World War as the debt grew, and it has been raised many times since. It has been raised, suspended or redefined dozens of times, a point that fits with a pattern I describe in From Gramm-Rudman to PAYGO, a timeline of fiscal rules that bent.

1979: the Gephardt rule

Over the decades the vote to raise the limit became a recurring ritual, and a politically painful one. Voting to raise the debt ceiling means voting for a higher number on a thing that sounds terrible. In 1979, the House adopted a rule named after Representative Richard Gephardt of Missouri that was meant to take some of the pain out. Under the Gephardt rule, when the House adopted the annual budget resolution, it was deemed to have passed a bill raising the debt limit to the level the resolution assumed. The members voted on the budget, and the debt limit followed automatically, without a separate recorded vote.

Defenders called it a way to avoid brinkmanship. Critics said it did the opposite of what a limit is meant to do: it removed the moment when members had to look at the number and say it aloud. The rule was dropped and brought back several times over the following decades, and it was discarded again when Congress's majority changed hands. The history of that rule is a small comedy of its own, since each party tended to favour it when it held the White House and oppose it when it did not. I will not give you dates for each swing that I cannot source, and I would suggest the reader check the House Rules for the current Congress.

Extraordinary measures: how the Treasury keeps paying

What happens when the debt reaches the limit and Congress has not yet raised it? The Treasury cannot issue new debt that would push the total over the ceiling. It can, however, take certain accounting steps that create room under the limit for a time. These are called extraordinary measures.

The best-known examples involve government accounts that hold Treasury securities. The Treasury can suspend new investments of certain federal employee retirement funds, such as the Civil Service Retirement and Disability Fund and the G Fund of the Thrift Savings Plan, and it can use other accounting devices, such as redeeming existing investments. Each action reduces the amount of debt subject to the limit without reducing what the government owes to its employees and retirees, and federal law provides that those funds are made whole when the limit is raised. The measures buy weeks or months, depending on the pace of receipts and payments. They do not create money. They shuffle which bucket the debt sits in.

That is the mechanism behind the phrase "X-date": the date by which, even with extraordinary measures, the Treasury would run out of cash and cannot pay all of its bills on time. Treasury Secretaries announce estimated X-dates, and they are estimates, because daily tax receipts are uncertain.

The modern fights

The debt limit became a leverage point in the budget wars of the last thirty years. In 1995 and 1996, a Republican Congress and President Clinton clashed over the budget, and Treasury used extraordinary measures to postpone default. In 2011, after a long standoff, Congress and President Obama agreed on the Budget Control Act, which raised the limit and imposed caps on discretionary spending; the credit rating agency Standard & Poor's downgraded the United States from AAA to AA+ that August, the first time it had ever done so, citing the political brinkmanship. In 2013, another standoff came near the limit. And in June 2023, the Fiscal Responsibility Act suspended the limit through January 1, 2025 and capped some spending for two years.

I would note two facts about that list. First, the standoffs produced actual spending restraints twice, in 2011 and in 2023. Second, they also produced a downgrade, brinkmanship and uncertainty in the markets, and in neither case did they end the pattern of rising debt.

The steelman: "the limit is a brake"

Here is the conservative argument for the limit, in the strongest form I can give it.

One, almost nothing else forces Congress to look at the total. Appropriations bills deal with spending in pieces, and no one ever votes on the sum. The debt limit is the one moment when the aggregate comes to a vote. Two, it is a point of leverage: a minority, or a party with a narrow majority in one chamber, can use a must-pass bill to extract spending restraint that could not pass on its own. Three, the two times it worked, in 2011 and 2023, produced caps that did slow some spending. Four, Congress has a duty to ask whether the country can afford a debt, and a standing limit forces the question to be asked in public.

That argument has real force, and I do not dismiss it. A republic needs moments when its representatives must go on the record about the total.

The answer: what the limit does not do

The best case against the limit as a brake, in numbered points.

One: it comes after the spending. The ceiling does not stop the spending decision. By the time it binds, Congress has already passed the appropriations and the entitlement laws that created the bills. Refusing to raise the limit does not cancel those obligations. It risks failing to pay for what was already promised, which is the wrong end of the process. A household that has run up a credit card bill does not repair its finances by refusing to pay it.

Two: it puts the full faith and credit of the United States at risk. The Fourteenth Amendment, Section 4, says: "The validity of the public debt of the United States, authorized by law, including debts incurred for payment of pensions and bounties for services in suppressing insurrection or rebellion, shall not be questioned." The text was written to guarantee the Union's war debts after the Civil War. Some scholars argue that it means the government may not default on lawfully authorized debt, and others argue about whether it empowers a president to borrow past the limit. That is a question for the courts, and I would rather Congress never force it. The point is that the Constitution treats the public debt as something to be honoured.

Three: the limit has not restrained the debt. Since the limit was created, the debt has grown almost without interruption. If a brake that has been released dozens of times is a brake at all, it is a weak one.

Four: it can distract from the real lever. The real tools for restraining spending are appropriations, which Congress controls, and the structure of programs such as Social Security and Medicare, which need reform. I discuss the shape of the problem in What the Social Security and Medicare trust funds hold, and depletion day. The debt limit sits downstream of all of them. Using it as the main tool is trying to steer the car by grabbing the back bumper.

Five: there are better rules. A balanced budget requirement, spending caps tied to the economy, or a rule that ties automatic debt-limit increases to a binding budget plan would put the discipline where it belongs, at the moment of spending. I weigh one such proposal in a balanced budget amendment, the case for it and the conservative doubts.

A verdict

So is the limit a brake or a bluff? Both, depending on whether Congress has a plan.

The history points to a conclusion that surprises both sides. The limit was born as an administrative convenience and has survived as a ritual. Its real value is political: it forces a public vote on the total, once in a while, and gives a determined minority a place to stand. Its real danger is that it can threaten the nation's credit to win a point about spending already enacted.

A conservative should therefore hold two thoughts. The first is that the country should never default, because the debt is an obligation and keeping one's word is a moral duty, and Article I gives Congress the duty to pay. The second is that Congress ought to make the real decisions earlier, at the level of appropriations, with rules that bind. A limit that is raised every time, with nothing attached, is a ceiling in name only. A limit that is raised with a real constraint attached, as in 2011 and 2023, at least shows what a deal can do. And a country that wants fiscal discipline should not depend on a once-a-year crisis to find it.

The text of the clauses is on the Constitution page, and you can watch the total yourself on the national debt clock.

The case in brief

1. Before 1917 Congress approved each bond issue by a separate act; the Second Liberty Bond Act of 1917 set an overall cap so Treasury could borrow without asking for each issue, which made borrowing easier.
2. The 1939 Public Debt Act created one aggregate limit of $45 billion, again for administrative flexibility, and the limit has been raised or suspended many times since.
3. The Gephardt rule of 1979 let the House deem the limit raised when it adopted a budget, and extraordinary measures let Treasury keep paying for a time when the limit binds.
4. The limit is a useful public vote on the total and a point of leverage, but it comes after the spending decision and risks the nation's credit.
5. Real restraint belongs at the moment of appropriation, under rules that bind, and a ceiling raised every time with nothing attached restrains nothing.

What you can do: look up the last vote on the debt limit on Congress.gov, see what, if anything, was attached to it, and ask your representative what spending rule he would attach to the next one.