Bank Reserves since the Start of Quantitative Tightening

One important aspect of U.S. monetary policy implementation is to control a short-term interest rate known as the federal funds rate (FFR), keeping it within the target range set by the Federal Open Market Committee (FOMC). The FFR is the equilibrium interest rate in the federal funds market. The federal funds market allows financial institutions, primarily banks, to borrow or lend funds, usually in the form of bank reserves. Bank reserves (or simply reserves) are funds that are deposited by banks with the Federal Reserve. Lenders in the federal funds market are banks that have excess reserves while borrowers are financial institutions that require extra funds to meet their short-term liquidity needs or, historically, reserve requirements imposed by the Federal Reserve.The reserve requirement ratios for depository institutions have been zero percent since March 26, 2020. See the webpage on the Federal Reserve Board’s reserve requirements for more details on the history and implementation of the reserve requirement.

First, this blog post describes the framework through which interest rate policy is implemented. We then discuss the implications of recent changes in the demand for bank reserves and what this might mean for the aforementioned framework.

The Implementation Framework of Interest Rate Policy

How does the Federal Reserve implement interest rate policy such that the FFR remains in its target range? The Fed currently operates in an implementation framework known as a floor system. By paying interest on reserve balances (known as the IORB rate), the Fed can then set an interest rate floor on the federal funds market. Think of it this way: If every possible lender in the federal funds market has a reserve account with the Fed paying the IORB rate, then the IORB rate effectively imposes an interest rate floor since no lender will be willing to accept any market interest rate below that offered by the Fed. In other words, if the market interest rate is lower than the IORB rate, lenders would opt to deposit their money into the Fed and earn the IORB rate.In actuality, the floor system relies on not one but on two policy tools. The primary tool is the IOBR. However, not all federal fund lenders are banks; for that reason, these nonbank lenders do not hold reserves with the Federal Reserve. Thus, such lenders will be willing to accept a market interest rate lower than the IORB rate. This is known as leakage. To patch this leakage, the floor system also utilizes overnight reverse repurchase agreements (ON RRP) between the Fed and qualified institutions. The ON RRP rate is usually set slightly below the IORB rate and provides a “subfloor.”

Notably, the FFR could still be higher than the IORB rate given the IORB rate only imposes a lower bound on the FFR. More specifically, when the demand for reserves is relatively strong such that borrowers are willing to pay an equilibrium interest rate that is higher than the IORB rate offered by the Fed, this will likely drive the FFR above the IORB, consequently causing the FFR to deviate from the target range. Hence, in order for the floor system to operate smoothly—i.e., keep the market interest rate sufficiently close to its floor—the Fed must always supply a sufficient amount of liquidity relative to the demand in the market.

The Current QT

In June 2022, the Fed started a new round of quantitative tightening (QT), which has reduced securities holdings on the asset side of the Fed’s balance sheet. As the balance sheet shrank, the amount of market liquidity provided by the Fed declined accordingly. The goal of the current QT is to drain the excess liquidity available in the market from “abundant” to “ample.”

In a March 20, 2024, press conference, Federal Reserve Chair Jerome Powell noted that the point at which the Fed eventually reaches ample reserves will depend on a number of factors.For more, see the press conference’s transcript (PDF). In this post, we present two factors that have arisen since the start of the current QT, which may influence the demand for bank reserves and thus alter the level of ample reserves.

Additional Factors Impacting Reserve Demand

Since the COVID-19 pandemic, there have been two noticeable changes that could increase the demand for bank reserves relative to the historical average: changes in the composition of bank deposits and portfolio adjustments in the form of decreased holdings of long-term bonds resulting from the inverted yield curve.

First, the composition of bank deposits has changed significantly; the table below provides supporting evidence. In the fourth quarter of 2023, checkable deposits were 23.6% of gross domestic product (GDP)—nearly double the amount, 12.1%, at the end of 2019, or right before the COVID-19 pandemic. Given that the bank deposit-to-GDP ratio is close to the pre-pandemic level (i.e., total deposits are comparable in relative terms) and checkable deposits have grown, there must have been a reduction in savings or time deposits. Data from the Federal Deposit Insurance Corp. (FDIC) shown in the final row of the table indicate that there has been no significant change in time deposits. Thus, the reduction can be most attributed to a decrease in savings deposits.

Furthermore, the Fed implemented a change in the regulation of savings accounts during the pandemic by removing the six-per-month transfer limit on savings accounts. As a result, savings accounts have become more liquid.

In short, the combination of increased checking deposits and fewer saving deposits, paired with newly relaxed restrictions on the number of transactions for savings accounts, means that banks’ liabilities have become more liquid. Therefore, despite a similarity in the level of bank deposits, banks now may require more reserves compared with pre-pandemic periods in order to hedge increased liquidity risks.

Second, the yield curve has been inverted since the end of 2022. What this means, for example, is that short-term assets, such as bank reserves, offer a higher yield than long-term assets, such as the 10-year Treasury security. At the start of the current QT in the second quarter of 2022, the IORB rate was more than 200 basis points below the yield on 10-year Treasuries, as shown in the table. In the end of 2023, the IORB rate was 95 basis points above the yield on 10-year Treasuries.

The deep inverted yield curve suggests that bank reserves could currently be the preferable asset in a bank’s portfolio. Not only do bank reserves provide liquidity, but they also could offer a higher yield compared with those of other bank assets.

The next table reports the evolution of major asset holdings on banks’ balance sheets since the start of the current QT in the second quarter of 2022. Total assets held by banks did decline relative to the economy, although not due to the reduction of bank reserves. Their aggregate asset position was reduced from 100% of GDP at the beginning of the current QT to less than 93% of GDP at the end of 2023. During the same period, the reserves held by private banks changed little, from 11.6% to 11.1%. Instead, the reduction occurred because of the decline in debt securities, from 26.6% to 21.5%. This reinforces the idea that the inverted yield curve could be an additional factor driving the increased demand for bank reserves since the start of the current QT given that bank reserves might be more desirable relative to long-term debt securities.

Changing Demand for Bank Reserves

As Federal Reserve Chair Jerome Powell noted in his March 20 press conference, reaching a state of ample reserves involves more than aiming for a specific dollar amount or a percentage of GDP. The process of reaching ample reserves will depend on a variety of factors that are constantly evolving. One example is demand for bank reserves, which could be volatile.

In this post, we have presented two recently developed factors worth considering. The first is the change in deposit composition paired with modifications in savings account regulation. The second is the yield curve, which is currently inverted, suggesting that there may be portfolio adjustments happening in which banks increase their holdings of bank reserves relative to long-term bonds. Both suggest that banks may desire a higher reserve balance than they did prior to the pandemic.

Notes

1. The reserve requirement ratios for depository institutions have been zero percent since March 26, 2020. See the webpage on the Federal Reserve Board’s reserve requirements for more details on the history and implementation of the reserve requirement.
2. In actuality, the floor system relies on not one but on two policy tools. The primary tool is the IORB. However, not all federal fund lenders are banks; for that reason, these nonbank lenders do not hold reserves with the Federal Reserve. Thus, such lenders will be willing to accept a market interest rate lower than the IORB rate. This is known as leakage. To patch this leakage, the floor system also utilizes overnight reverse repurchase agreements (ON RRP) between the Fed and qualified institutions. The ON RRP rate is usually set slightly below the IORB rate and provides a “subfloor.”
3. For more, see the press conference’s transcript (PDF).