| When you make a deposit at a bank, the bank is required by the Fed to hold 10 percent of that deposit as cash. Confusingly, the Fed does not allow the bank to count that cash as available to help meet a bank run in an emergency. In fact, as explained below, the more cash the Fed requires a bank to hold, the less prepared to meet such a run the Fed considers the bank to be. As a result, the Fed has not accomplished its primary objective for paying interest on required reserves for over 10 years. The Fed can easily fix this expensive mistake by simply judging a bank that holds more cash at the Fed’s own behest to be better prepared to meet its liquidity needs in a financial panic. |
In 2006, Congress granted the Federal Reserve the authority to pay interest on deposits of depository institutions (commercial banks, thrifts, and credit unions, henceforth “banks”) at Reserve Banks. Such deposits are also referred to as “reserves.” Originally, Congress authorized interest payments to begin on October 1, 2011, but in the fall of 2008, Congress permitted the Fed to begin paying interest immediately so that the Fed could conduct monetary policy with a swollen balance sheet during the financial crisis. (See the BPI research note “Understanding the Fed’s implementation framework debate,” for a more complete explanation.)
Banks keep deposits at Reserve Banks in part because they are required to do so. Such “reserve requirements” equal 10 percent of transaction accounts (such as checking accounts). Prior to 2008, because the Fed did not pay interest on the deposits, banks sought to keep required reserves as low as possible. One mechanism they used was “sweeping” funds at the end of each day out of some transaction accounts, which have a reserve requirement, into savings accounts, which do not. Such accounts are referred to as “sweep accounts.” Customers, who chose to have their funds swept, earn higher interest on funds in sweep accounts, and banks economize on reserve requirements.
Currently, more attention is paid to the interest rate the Fed pays on excess reserves (the IOER rate) than the interest on required reserves because of the importance of the IOER rate for conducting monetary policy with a large balance sheet. However, the Fed originally sought the authority to pay interest on reserves in order to reduce banks’ incentives to engage in inefficient behavior to minimize reserve requirements such as establishing sweep accounts. The Fed argued that if it could pay a market interest rate in required reserves, the opportunity cost to the banks of the reserves would be much lower, reducing or eliminating their incentive to sweep.
For example, when Donald Kohn, then a Governor of the Federal Reserve Board, testified before Congress on June 22, 2004, in support of granting the Fed the authority to pay interest on reserves, he stated:
Because no interest is paid on the balances held at Reserve Banks to meet reserve requirements, depositories have an incentive to reduce their reserve requirements to a minimum. To do so, they engage in a variety of reserve avoidance activities, including sweep arrangements that move funds from deposits that are subject to reserve requirements to those that are not and to money market investments.
It is surprising and disappointing, therefore, that despite paying a market interest rate on required reserves for over 10 years, banks continue to sweep funds out of transaction accounts and into savings accounts (including MMDAs). Since the authority was granted, the Federal Reserve has paid approximately $6 billion in interest on required reserves.
The Fed stopped collecting data on sweep accounts in 2012. A rough proxy for their prevalence, however, can be calculated using data from the Call Report. Each quarter, banks report the interest they pay and the average amount deposited in savings- and interest-bearing-transaction-accounts, respectively. Those four statistics make it possible to estimate the interest rates banks are paying on transactions and savings accounts. Surprisingly, the estimated interest rate on transaction accounts is much higher than on savings accounts, even though transactions accounts provide customers additional functionality and are less stable sources of funds for banks.
The paradoxical difference is the result of sweep accounts. Customers earn interest on the balance before the sweep, but the Call report items report the balance after the sweep. The sweeping reduces the reported levels in transaction accounts and raises the levels in savings accounts, pushing up the implied interest rate on transaction accounts and down the rate on savings accounts.
If we assume that the interest rate on the two types of accounts are equal (in reality, the rates on transaction accounts are lower than on savings accounts, so this is a conservative assumption), we can infer the amount swept. Using this technique, we estimate that about half of the amount held in transaction accounts is swept into savings accounts.
There are many problems with the estimate. Banks sweep funds out of non-interest-bearing transaction accounts, not just interest-bearing-transaction accounts. And the funds can be swept into repo, Eurodollars, or money market funds, not just savings accounts. Even so, the crude estimate indicates strongly that sweep accounts are still prevalent.
An informal survey of bank treasurers indicates that banks continue to sweep funds out of reservable accounts largely because required reserves do not count as high-quality liquid assets (HQLA) for satisfying the liquidity coverage ratio (LCR). Thus, between two banks with the same balance sheet, the one that sweeps funds out of accounts with reserve requirements into accounts without reserve requirements has a higher LCR.
In a recent blog post (“Realizing the liquidity benefit of required deposits at the Fed”), we argue that the Fed should change its rules so that it can count required reserves as HQLA. Doing so would be consistent with the original intent of making banks hold a fraction of their demand deposits as cash or deposits at the Federal Reserve banks – meeting liquidity demands if the deposits were withdrawn. Moreover, an overlooked provision of the Federal Reserve Act requires the Fed to allow banks to use funds held to meet reserve requirements to satisfy liquidity requirements. To those two reasons for counting required reserves as HQLA we now add a third – not counting required reserves as HQLA is preventing the Fed from accomplishing the main objective for which it sought the authority to pay interest on reserves in the first place.
Disclaimer: The views expressed in this post are those of the author(s) and do not necessarily reflect the position of the Bank Policy Institute or its membership, and are not intended to be, and should not be construed as, legal advice of any kind.
