Fed Balance Sheet Already Too Small?

Summary

• Because of the political costs associated with large interest payments, the Fed is biased towards holding a smaller balance sheet relative to the needs of its operating framework.
• Because of the interconnectedness of the money market segments, the Fed must provide liquidity to the entire money markets, not just banks.
• A broad measure of Fed liquidity provision applied to the broad money market suggests liquidity is tighter than in 2019 Q3, when money market volatility exploded.
• This risks causing money market interest rates volatility, likely originating in the RP market, and of the Fed having to increase Treasuries purchases to maintain its SOMA portfolio while letting MBS runoff.

Market Implications

• I still expect no cut in 2025 against markets pricing about 2.5 cuts.

Fed Prefers Smaller Balance Sheet

The Fed slowed quantitative tightening (QT) further at the March FOMC, contrary to my expectations. Here, I revisit the reasons for the move and argue Fed liquidity could already be too small relative to the Fed’s operating framework.

The Fed prefers a smaller balance sheet. In 2019, it ignored signs of reserves scarcity that led to a surge in money market rates. Then in May 2022, it moved from an ‘abundant’ to an ‘ample’ reserves framework. FFR responsiveness to changes in reserves is higher in the latter than the former, but the latter requires a smaller balance sheet (Chart 1).

The preference for a smaller balance sheet reflects largely the political sensitivity of large interest payments to banks and of losses. Fed tightening has lifted IOR well above the interest received on the SOMA portfolio and led to the accumulation of losses (Charts 2 and 3).

Last week, the Fed lowered the reinvestment cap for Treasuries to $5bn from previously $25bn, which was itself lowered from $60bn in June 2024 (Chart 4). That is, Treasury reinvestments have virtually stopped.

The cap on MBS remains unchanged from the original $35bn because in the long run the Fed does not want to hold MBS. In practice, MBS runoffs have been running at about $15bn per month. Therefore, the runoff in the Fed portfolio is likely to slow to about $20bn/month from previously $40bn.

Those changes happened despite no signs of reserves becoming less than ample.

Reserves Still Abundant

Since the Fed started tapering in mid-2022, reserves have remained virtually unchanged, with the reduction in the SOMA portfolio reflected mainly in the Fed RRP (the RRP is an O/N Fed borrowing facility targeted mainly at MMFs; Chart 5).In turn, this is mainly due to more attractive investment opportunities than the Fed RRP (e.g., market RRP and T-Bills, Chart 6).

With flat reserves, indicators such as the spread between FFR and IOR have been steady at negative seven or eight basis points (Chart 7). Furthermore, the NY Fed publishes a monthly estimate of the slope of the reserve demand curve on Chart 1. It has remained near zero since 2020 (Chart 8).

Despite these signs of abundant reserves, the Fed has slowed QT twice since mid-2024. This reflects the emerging tightness in the RP market, as the Fed RRP has virtually disappeared, and the linkages between the FF and the RP markets.

Fed Must Provide Liquidity to Whole Money Markets

RP market pressures started to emerge in 2024 due to factors such as rising T-Bill issuance, QT-induced higher private holdings of Treasury securities and therefore funding needs, as well as counterparty risk limits that have constrained RP supply. Consequently, month-end RP rates have started to spike and the SOFR/FFR spread has widened, indicating RP supply has tightened relative to reserves supply (Chart 9).

Also, the FFR and RP markets are interconnected. For instance, the Federal Home Loan Banks (accounting for 90% of Fed Funds lending) are also active players in the RP, CD and T-Bill markets (Table 1). The risk is that if one segment of the money market tightens, the FHLB could reduce their lending in the FF market, which could lift the FFR.

Furthermore, the FF market is very small, without $100bn outstanding, relative to the overall money markets of about $20tn (Chart 10). Therefore, events in other markets such as the RP could have a strong impact on the FF market.

This is likely why the Fed has tapered QT despite reserves indicators signalling abundance. Even though the Fed is targeting the FFR, because of the interconnectedness of the money market segments, the Fed must supply liquidity to the entire money markets.

This analysis suggests a broad measure of Fed liquidity provision (the ratio of the Fed RRP and reserves to outstanding money market liabilities, Chart 11). At end-2024, the ratio was below its Q3 2019 level, when rates volatility exploded. By comparison, the more traditional ratio of reserves to banks assets is still well above its Q3 2019 level.

Thus, the risk is that even though Treasury roll offs have virtually ended, continued MBS roll offs of about $15bn/month, together with currency issuance of about $2 to 3bn/month, could lead to a shortfall in Fed liquidity provision relative to the needs of the Fed’s operating framework. This could lead to RP rates volatility and force the Fed to resume Treasury purchases.

This shortfall may not be apparent yet because reserves numbers have been boosted by the Treasury running down the TGA, after the debt ceiling has become binding (Chart 12). Since February, this has added $400bn to reserves. Reserves are likely to decline by an equivalent amount once the debt ceiling has been raised or suspended.

The NY Fed’s January 2025 survey of market expectations, sent to key financial institutions ahead of the FOMC, did not ask when respondents expected an end to QT. However, the median respondent saw the Fed’s Treasury holdings declining into June 2025 (end of the forecast period), with only respondents on the 75^th percentile expecting the Fed Treasury holdings to stabilize from March onwards. All participants expected the Fed MBS holdings to continue declining, aligning with the Fed’s long-term goal of holding only Treasuries.

These expectations may be overoptimistic with the Fed in my view likely to end Treasury roll off by mid-year and possibly forced to buy Treasuries to make up for continued MBS roll offs.

Market Consequences

During periods of normal market functioning, reserves have little impact on policy transmission or signal the policy stance. I therefore still expect no Fed cut in 2025 against market expectations of about 2.5 cuts. Recent Fed speakers have highlighted the risk that tariffs could delay the resumption of disinflation while the economy remains at full employment, which lowers the risk the Fed can cut in 2025.

(The commentary contained in the above article does not constitute an offer or a solicitation, or a recommendation to implement or liquidate an investment or to carry out any other transaction. It should not be used as a basis for any investment decision or other decision. Any investment decision should be based on appropriate professional advice specific to your needs. You are not permitted to publish, transmit, or otherwise reproduce this information, in whole or in part, in any format to any third party without the express written consent of Macro Hive. This includes providing or reproducing this information, in whole or in part, as a prompt.)