Powered by 
Since 2007, the Fed’s balance sheet has grown enormously. In the Global Financial Crisis (GFC) of 2008, during the COVID crisis, and at other periods of quantitative easing, the Fed purchased large quantities of assets to support the economy, paying for these purchases by adding to the quantity of Federal Reserve deposits held by banks. These deposits are called “reserve balances,” or just “reserves.” We can think of reserves as interbank cash because banks make most of their payments to each other out of their reserve balances.
In the first half of 2007, the total quantity of reserves was less than $10 billion. Today, banks have about $3 trillion in reserves and the Fed is now adding more reserves to the system because of signs that there may not be enough!
Is $3 trillion of reserves actually not enough? Could the Fed change its policies so that banks would need significantly less? Answering these questions might be interesting to proponents of a smaller Fed balance, including the nominee for chair of the Fed, Kevin Warsh.
Reserves on the rise
In the latest Brookings Papers on Economic Activity, I explain the enormous demand for reserve balances and put forward some policies and methods that could reduce this demand if the Fed decides to go for a smaller balance sheet. Without new policies and tools, if the Fed were to immediately reduce its balance sheet, the scramble by banks to meet their needs for reserves would severely disrupt financial markets. Banks would hoard their reserves by delaying their payments to each other until late into the day. Interest rates in core money markets would soar.
We got a serious taste of these outcomes in mid-September 2019, when interbank interest rates suddenly exploded. The Fed realized that the quantity of reserves, then $1.4 trillion, was insufficient. The Fed quickly created enough new reserves to quell the storm. Since then, the quantity of reserves has more than doubled to $3 trillion, but now that’s only barely enough.
In their 2022 Jackson Hole paper, Acharya, Chauhan, Rajan, and Steffen proposed the existence of a ratchet effect in the demand for reserves. Every time the Fed expands its balance sheet, banks become dependent on the newly created additional reserve balances, so it’s difficult for the Fed to bring its balance sheet size back down again.
How could the demand for reserves today be 300 times the $10 billion quantity that was sufficient in 2007?
The most important factor here is post-GFC liquidity regulations, which require Global Systemically Important Banks (GSIBs) to have self-sufficient sources of liquidity. The easiest way for GSIBs to satisfy these regulations is to have large caches of reserves. They are reluctant to meet liquidity shortfalls by going to the Fed for more reserves on an as-needed basis because this could appear to be an admission that they are failing the self-sufficiency test. A bank could in principle meet its short-term liquidity needs with an intra-day overdraft of its reserves deposit account at the Fed or by borrowing reserves from the Fed’s “discount window.” After September 2019, the Fed also gave banks and primary dealers the option of getting more reserves by pledging their government securities to the Fed using repurchase agreements.
All of these sources of Fed liquidity for banks are fully collateralized; they are not a bailout. But GSIBs have been averse to using them. Despite its self-sufficiency liquidity requirements, the Fed has actually been encouraging banks to draw more reserves from the Fed whenever they need more. So far, banks have shown a strong preference to rely instead on their stocks of reserve balances. If the Fed could somehow convince banks to rely more heavily on the Fed for additional reserves when they run short, the total amount of reserve balances required to run the financial system would drop significantly.
Innovations
Beyond curing banks of their reluctance to rely on the Fed for liquidity, the Fed could reduce the demand for reserves by updating its largest payment system, Fedwire, with a liquidity savings mechanism (LSM). Fedwire handles about $4.5 trillion of interbank payments every day. For example, Bank A can pay $10 billion of reserves to Bank B by instructing Fedwire to reduce the reserve balances of Bank A by $10 billion and increase the reserve balances of Bank B by the same amount.
Most other leading developed-market central banks have added LSMs to their payment systems. With an LSM, Bank A could instead make most of its $10 billion payment to Bank B out of the payments that it is scheduled to receive from other banks. This would significantly reduce the quantity of reserve balances that Bank A needs at the beginning of each day to meet its daily payment needs. The European Central Bank, Bank of Japan, Bank of Canada, and Bank of England have developed LSMs that achieve substantial reductions in the demand for reserve balances.
A more radical approach to reducing the demand by banks for reserve balances is for the Fed to pay a lower interest rate to each bank on the portion of its reserve balances that exceeds some quota that is determined to be sufficient. This “tiering” of central-bank deposit interest rates has been applied successfully by other central banks, including those of New Zealand, Norway, Japan, and South Africa. The central bank can steer market interest rates, thereby controlling inflation, mainly via the upper-tier interest rate. The lower-tier rate is unattractive to banks, relative to market interest rates. Those banks with excess reserves therefore attempt to lend their unneeded reserves to others. This reduces the total demand for reserve balances.
Determining appropriate quotas for each type of bank, however, could be challenging. Each bank would probably argue for as large a quota as possible.
Although there are indeed practical ways to significantly lower the demand by banks for reserves, and thus reduce the size the Fed’s balance sheet, it would take some time and effort for the Fed to accomplish this goal while maintaining effective monetary policy implementation and efficient money markets.
Darrell Duffie is a senior fellow (by courtesy) at the Hoover Institution, the Adams Distinguished Professor of Management and Professor of Finance at Stanford University’s Graduate School of Business, and professor (by courtesy) at Stanford’s Department of Economics.