The US debt ceiling: A slow burn, not a big bang

OVER the past decade, writing about the debt ceiling has become a rite of passage, for both sell-side analysts and the financial press. Endless ink has been spilled speculating whether this is finally the one time when the debt ceiling is not raised. And sure enough, it’s that time of year again.

Here we go again

Treasury Secretary Janet Yellen has told Congress that Jan 19 was the date on which the US officially reached the debt limit. The Treasury is now using extraordinary measures to pay the country’s bills and service its debt, as it has done several times in the past decade. 

“But wait!” we can hear some clients say. Surely this time the risks are higher? After all, for the first time in over 150 years, the House needed 15 ballots to decide the Speaker election. Republican House members, especially the ones who refused to vote for Speaker Kevin McCarthy for 14 long rounds, have demanded that any increase in the debt limit be paired with spending cuts – a condition that the Biden administration has so far refused point-blank. Doesn’t this kind of political dysfunction make it more difficult to raise the debt limit this time around?

Perhaps. We are not political forecasters and will not assess the chances of the debt limit being breached. Tempting though it is to pontificate, we also will not waste readers’ time discussing the legality of the trillion-dollar coin, the chances of the 14th Amendment being invoked, et cetera. Such issues have been endlessly debated over the past decade in the financial press; we have nothing new to add.

Importantly, our economic forecasts do not include a drag from debt ceiling-related issues. Our implicit assumption is that this will not turn out to be a material economic event for the country. But in this piece, we consider the “what-if” scenario. If the debt limit is not raised in time, how will the economy and financial markets react? The answer is not intuitive. A debt ceiling crisis will not immediately be a big bang, in our view; it would be a slow burn that builds to a bigger problem.

The ‘x-date’ is many months away

But first, a reminder that any potential problem is many months away. Yellen has stated that extraordinary measures (as well as the cash in the Treasury General Account) will carry the country until at least early June. We think there is more time. We use 2019 as a baseline for estimating 2023’s cash inflows and outflows for Treasury, since 2020 to 2022 were heavily affected first by the pandemic, and then by fiscal stimulus. Our estimate is that extraordinary measures will likely run out in August, not June. Of course, this is subject to change – for example, a sharp recession in the coming months would lower tax revenue and pull the “x-date” forward. But as it stands, we think the x-date is an issue for the third quarter, not the second.

Eventually, though, that day will be upon us. If the debt ceiling is not raised by then, the Treasury will have to rein in outlays. But it would not stop making payments on Treasury debt. In 2011, Treasury and Fed officials devised a payment schedule if the debt limit was not raised on time, and reiterated these principles in 2013. They agreed that Treasury would prioritise first paying principal on interest on Treasury bonds and make them on time. Settlement would not be affected, and new issuance would fund maturing securities (which can happen without breaching the debt limit).

A temporary fiscal hit, not a bond default 

While these principles have never been put to the test, they strongly imply that Treasury debt obligations will not be affected. But other forms of government expenditure will be, and suddenly. For example, if Treasury cash flows mimic the path of 2019, the US government’s cash outlays in August this year will be US$200 billion over inflows. If the debt limit is not raised to cover this extra US$200 billion, the government will need to cut spending by that amount. In fact, the cut might be even greater; the Treasury would have to delay or hold back spending even if it has sufficient balances on a given day, because it would want to accumulate sufficient funds to pay a large, future coupon payment.

We do not mean to sound too sanguine; debt prioritisation assumes that Treasury can predict upcoming cash flows accurately and has enough time to start withholding cash to cover interest payments. In addition, paying bondholders but not fulfilling other existing commitments – including to US businesses and households – is a tricky proposition, and one likely to spark political backlash. Still, in our view, the real effect of the debt ceiling not being raised is a sharp but temporary fiscal contraction, not a bond default.

Any fiscal drag should be temporary, unless there is a sequestration deal

We do not foresee any fiscal drag having a lasting effect. After all, all of the forced spending cuts will have to be made up whenever the debt ceiling is raised; these are all spending commitments that the government has already made. Even in past government shutdowns, federal workers were retroactively paid for the period when their departments were closed, even when that was not a binding commitment. But if several weeks pass after the x-date without an increase in the debt limit, the economic pain will start to make itself felt.

For example, if the government is forced to cut US$200 billion in purchases of non-essential goods and services in August, US gross domestic product (GDP) growth for Q3 will still show little effect, assuming the debt ceiling is raised in a few weeks and spending resumes. The temporary cut is merely delayed spending within the quarter, leaving little imprint on GDP. But take a harsher scenario. Spending resumes in September, but the August cuts are reimbursed only in Q4. There could then be a sharp drop in Q3 GDP (perhaps as much as 15 per cent quarter over quarter seasonally adjusted annualised rate). But once again, growth likely would bounce back strongly in Q4, leaving average GDP unchanged for the year. In fact, any delayed government spending would likely not even translate one-for-one into production cuts. Businesses presumably would continue to produce and accumulate inventories in expectation of an eventual resumption of government purchases.

If the Treasury instead cuts US$200 billion in transfers to individuals in August, borrowing-constrained households might cut spending. But once again, there should be a bounce back in demand when those payments are finally made. Any effect likely would be non-linear; workers and businesses that are owed money by the government should be able to manage a few weeks’ delay, in our view. But as time passes, more things may “break” in the real economy. The economic effect would start slow but worsen progressively, in our opinion, especially if concerns build (though we think this is very unlikely) that the spending cut is permanent.

A material hit to economic growth would be likely only if debt ceiling negotiations result in lasting spending cuts. For now, the Biden administration is insisting on a “clean” debt ceiling bill. Say a sequestration deal is ultimately agreed upon, along the lines of the Budget Control Act of 2011, which followed 2011’s debt ceiling crisis. In late July that year, just before the x-date, the debt ceiling was raised in exchange for spending cuts of over US$900 billion over the next decade. Since Congress failed to produce a large enough deficit reduction bill, sequestration started in 2013 – spending cuts of US$85 billion in fiscal year 2013, followed by US$109 billion per year from 2014 to 2021, split equally between defence and non-defence programmes. These caps on spending were, however, then raised several times. In all, fiscal policy reduced GDP growth by about 1 percentage point per year from 2012 to 2014, according to the Hutchins Center Fiscal Impact Measure, with a smaller effect in later years.

Market effect: Focus on money markets and munis

Given this backdrop, we believe that the day after the debt limit is “truly” hit, market reaction should be measured, even if a generalised risk-off is still likely. Equities and other risk assets could fall as investors express concerns about US political dysfunction. Some might move out of bills and into longer-term Treasuries, while others might turn to bank deposits, as happened in 2011. But ultimately, markets and analysts (ourselves included) feel that it is a matter of time before political pressure would force Congress to raise the limit. As time passes and this assumption is tested, market reaction likely would intensify, but similar to any other time when investors are resetting US growth expectations lower.

What is very unlikely, even if the debt limit is breached, is a buyer’s strike on US Treasuries, breathless media articles to the contrary notwithstanding. First, as noted above, a debt ceiling breach is a temporary forced spending cut, not a bond default. Second, on the off chance that Treasury fails to pay a T-bill on time, investors would likely treat this as a payment delay, not a default. Admittedly, money market funds (which tend to eschew all risks) would look sceptically at bonds that pay on dates near the x-date. Such bills could cheapen 30 basis points or more relative to issues maturing just before or after. But we think any such cheapening is an excellent opportunity for non-money fund investors. Indeed, historically, such issues have attracted foreign official institution interest. Beyond the front end, a debt limit breach is likely to be a catalyst for a Treasury bond rally.

We are also not very worried about a rating agency downgrade (though it might affect some municipal bonds). All the rating agencies have warned that extended political gridlock puts the US rating at risk. In 2011, S&P did downgrade the US due to an extended debt ceiling fight. Markets clearly disagreed with the S&P argument that US sovereign debt was more risky; Treasuries rallied during that debt ceiling back-and-forth. This time, in our view, will be similar. This is ultimately a fight that will affect how much the government can spend, not whether it can service its sovereign debt obligations. 

Away from the front end, the municipal bond market could be specifically affected. In past debt ceiling episodes, the Treasury suspended issuance of State and Local Government Securities (SLGS) as one of its extraordinary measures; the SLGS window has been open thus far, but could be closed in the future. Separately, payment delays and fears of downgrades of the US credit rating might affect municipalities that rely on federal funds, including bonds issued by military housing, state housing financing authorities, bonds backed by transportation anticipation revenues, et cetera. Additionally, delays in Medicaid and Medicare reimbursements could pressure non-profit hospitals, which are already hurting. If there are cuts to discretionary spending due to sequestration, municipalities will receive less funding. Depending on how any sequestration is implemented, extraordinary redemption provisions might get triggered, and more Build America Bonds from 2009 to 2010 could be called below market prices. 

This is a Thinking Macro report by Barclays’ fixed income, currencies and commodities research department.