Debt Ceiling Headroom: The Fed Has an Extra $14 Billion

A new "extraordinary measure" for the Treasury

The whole debt ceiling thing is back again. I’m not going to recapitulate the whole issue, as it’s been well covered elsewhere. But it’s the classic debt ceiling story: the U.S. government is about to run up against its statutory limit on debt issuance, forcing the Treasury to resort to “extraordinary measures” with respect to its cash/debt management in order for the government to keep meeting its obligations. After those measures run out and we hit the so-called “X date” of an empty cash register, well, let’s hope Treasury (and the Treasury market) has its affairs in order.

With just $78 billion of debt space from the statutory ceiling, Treasury Secretary Yellen announced that these extraordinary measures would start January 19. As per Bloomberg:

In her letter, Yellen said the Treasury’s extraordinary measures would begin by redeeming existing — and suspending new — investments of the Civil Service Retirement and Disability Fund and the Postal Service Retiree Health Benefits Fund. The department will also suspend reinvestment of the Government Securities Investment Fund of the Federal Employees Retirement System Thrift Savings Plan.

Not mentioned in Yellen’s letter is the other commonly included extraordinary measure of ceasing reinvestment of the dollar balances of the Exchange Stabilization Fund. This has been included in past summaries of extraordinary measures, and maybe will be if/when the Treasury releases a similar document (as opposed to the letter format from Yellen) this time around.

Here, for instance, is the bulletin from August 2021. Then, ceasing reinvestment of ESF dollar balances into nonmarketable, one-day Treasuries would free up $23 billion of headroom. Currently, it looks like it would free up about $17 billion.

Ok but, the Fed

In the event there is a breach of the X date, the Federal Reserve would certainly have a role to play on the operational/markets side. Would it put QT in reverse? Would it do full allotment renewable repos—effectively taking any and all Treasuries off the Street’s balance sheets? Would it invoke Section 13(13), which allows it to provide liquidity against Treasuries for anyone who might want to put them to the Fed?¹ Or the more familiar Section 13(3)?²

But, irrespective of that, the Treasury needs to add the Fed to its list of “extraordinary measures” as there’s approximately $14 billion of debt ceiling headroom parked uselessly on the Fed balance sheet — a relic of the pandemic-era interventions by the Fed and Treasury.

Quick Background

In 2020’s CARES Act, Congress allocated $454 billion to the Treasury to subordinated for Fed facilities to lend to a variety of businesses, nonprofits, and government entities. The Fed’s Section 13(3) emergency liquidity statute requires that it be reasonably confident in full repayment. Thus, having more capital structure seniority thanks to an equity investment from Treasury expands the Fed’s ability to intervene.

Of the $454 billion, the Treasury committed $195 billion to four Fed programs. The 13(3) programs that had equity funding from the Treasury under the CARES Act purchased corporate bonds and loans, municipal bonds, and asset-backed securities (the Corporate Credit Facilities (CCFs), Main Street Lending Program (MSLP), Municipal Liquidity Facility (MLF), and Term Asset-Backed Securities Loan Facility (TALF)). The programs were designed to leverage the Treasury’s equity up to as much as 14x, depending on the type of loan or asset purchased or used as collateral.

The Treasury disbursed a total of $102.5 billion to the Fed: the full $10 billion for TALF and the first half of the committed amounts for the other three facilities. Total purchases under these facilities came in even lower than that; none of these programs’ volumes ever even surpassed the total equity the Treasury had injected, let alone “leveraged” it in the traditional sense. (Yay, backstop effect!)

After some legislative drama in late 2020, the Consolidated Appropriations Act, 2021 mandated that the Fed facilities with CARES Act funding support be closed and that the Fed return unused funds. Specifically, it called for the rescission of CARES Act funds that were “not needed to meet the commitments, as of January 9, 2021, of the programs and facilities established under section 13(3) of the Federal Reserve Act in which the Secretary of the Treasury has made or committed to make a loan, loan guarantee, or other investment using funds appropriated under section 4027 of the CARES Act.”

Weirdly however, in effecting this statute, the Fed and Treasury reduced the Treasury’s position in these facilities to the “maximum risk” to the Fed. Because none of the CARES Act-supported Fed facilities did volumes that exceeded even the Treasury’s equity injection, the Fed had more than dollar-for-dollar equity coverage from the Treasury. Pursuantly, it only reduced the Treasury position to full dollar-for-dollar coverage — leaving the facilities still with no leverage, inconsistent with facility design.

For instance, the MLF closed with an outstanding balance $6.3 billion of purchased munis, but had had $17.5 billion injected by the Treasury. It transferred $11.2 billion back to Treasury following the facility’s legally mandated closure, and it (as is the case with the MSLP and TALF) now currently does only incremental biannual distributions as the assets mature—maintaining dollar-for-dollar coverage.

Uhh, why?

This is prudence in no one’s interest.³ The original intent with the MLF (for instance) was to leverage $35 billion of Treasury equity in to $500 billion of muni purchases. Thus, the Fed had determined that it could meet Section 13(3)’s legal requirement to confidently expect full repayment if it had ex ante loss protection from Treasury on the most junior 7% of the facility’s capital structure. Yet, when the Fed downsized the facility following its mandated closure, it left Treasury with 100% of the capital structure of all the CARES Act-supported facilities.

This also contrasted with the (singular) historical example of a Treasury-supported Fed facility becoming overcapitalized after its closure: the 2008 TALF. The 2008 TALF (as with the 2020 iteration) was designed as a 10x leverage facility to support the purchase of securitized assets. The Treasury had allocated $20B of TARP money as subordination for the facility to provide $200B of liquidity into structured finance markets. When the facility closed on June 30, 2010, it had $43B of assets outstanding. Twenty days later, the Fed announced it was returning excess funds to the Treasury—reducing its subordinated stake to $4.3 billion and the facility to 10x leverage:

There’s Always Money in the Banana Stand

The corporate bond facilities have totally wound down, but the MLF, MSLP, and TALF still have outstanding balances. If the Fed and Treasury returned them to their designed leverage ratios, the Fed could create $14 billion more headroom for the Treasury. (The below numbers are rounded/approximate as there are some oddities with the likes of accrued interest.⁴)

Some back-of-the-envelope:

The MLF has $2.9B of assets and $2.9B of Treasury support. Returning the facility to its designed ~14x leverage would release about $2.7 billion to Treasury.

The TALF has $1.0B of assets and $1B of Treasury support. Returning the facility to its designed 10x leverage would release about $0.9 billion to Treasury.

The MSLP has $10.3B of assets (net of loan loss allowances) and $11.6B of Treasury support. Returning the facility to its designed 8x leverage would release about $10.4 billion to Treasury.

All told, that’s $14 billion. It’s no platinum coin. But given these debt ceiling debates risk the deadline lapsing even for just minutes or hours, it’s also not nothing. This would not represent any material risk to the Fed and would certainly leave the facilities within the Fed's legal mandate. That, combined with the historical precedent, seems as though it would leave the Fed as a happy accomplice to this release of funds.

Comments are turned off here but more than welcome via email (steven.kelly@yale.edu) and Twitter (@StevenKelly49). View Without Warning in browser.

1. Section 13(13):

2. For instance, during covid, the Fed ran the Paycheck Protection Program Liquidity Facility (PPPLF) under Section 13(3), despite that it could’ve been run under Section 13(13) as per the statute in the previous footnote. The Fed invoked 13(13) in 2008 to establish credit lines with Fannie and Freddie before they were rescued (any debt they issued would fall neatly in the above statute), but no credit was ultimately extended under the line.

3. The counterargument is here is that this guarantees that the Fed itself will not take a loss, which may be beneficial for political/sentiment/posterity reasons. But, that’s some non-GAAP, WeWork buffoonery. Moreover, we shouldn’t be doing backflips to reinforce the notion that all emergency lending is profitable; it’s not designed—or, at least shouldn’t be designed—to avoid all losses. Yellen knows this. (Mnuchin maybe didn’t.) Regardless, this all goes away on a combined government balance sheet; i.e., a Fed loss just results in lower remittances (“all money is green”). So there’s no need to go beyond the legal requirement for the Fed here.

4. Technical distinction without a difference: Treasury’s equity injections in these facilities are a mix of cash and nonmarketable Treasurys. To the extent the equity disbursement is done via a return of the nonmarketable Treasurys, the Treasury can retire them — creating space under the debt ceiling for the Treasury to issue more. To the extent its done in cash, the Treasury simply has more cash without issuing debt.

The data are from the Fed’s latest report to congressional committees on its emergency liquidity facilities — January 10, 2023 — and the Fed’s latest weekly H.4.1 balance sheet report — January 12, 2023.