Since March, we have experienced three of the largest bank failures in United States history. Silicon Valley Bank, Signature Bank, and New Republic Bank had combined assets worth over half a trillion dollars — roughly equivalent to the combined assets of the 165 banks that failed in 2008 and 2009. The Federal Reserve has responded to these failures by injecting emergency liquidity into the banking system, reversing months of quantitative tightening.
Commentators have argued that these failures have put the Fed in a challenging position. The Fed must flood the banking system with reserves to prevent a financial crisis. But doing so may undermine the Fed’s ability to bring down historically high inflation. While monetary policymakers are in an unenviable position, these concerns are overblown for two reasons. First, monetary stimulus does not generally have the same effect on the economy as emergency lending. Second, and more importantly, the Fed’s operating framework reduces the extent to which changes in reserves affect the broader money supply.
Suppose several banks become illiquid simultaneously, perhaps because of an unexpected outflow of deposits, and cannot borrow reserves from other banks as a result. The Fed can step in as the lender of last resort to provide emergency loans to these banks through the discount window, provided that the banks in need have good collateral. By providing emergency liquidity, the Fed can prevent these banks from having to liquidate their assets at fire-sale prices, which — if left unchecked — could trigger a financial crisis.
In this example, banks are not borrowing reserves from the Fed to make loans, which would increase the broader money supply and, by implication, inflation. Instead, they are borrowing to meet the deposit outflow. Thus, in this example, the Fed’s emergency lending is unlikely to result in higher inflation because the banks are using the borrowed reserves to meet their depositors’ withdrawals. Indeed, if the Fed failed to provide such liquidity, and a financial crisis ensued, the likely result would be deflation rather than inflation.
Of course, the Fed cannot directly control how banks use the funds acquired through the discount window. Banks could borrow from the Fed and use the funds to make loans, expanding the money supply and increasing inflation. Doing so would be especially attractive if the discount rate — the rate the Fed charges on such loans — were below what banks could earn by making loans. However, as long as the Fed sets the discount rate sufficiently high, it will not be profitable for banks to use the discount window as a source of loanable funds. Thus, there is no tradeoff between providing emergency liquidity via the discount window and reducing inflation.
Even if we grant that emergency lending may be stimulative, the Fed’s operating framework allows policymakers to expand liquidity substantially without having as much effect as such expansions would have had under its prior framework. The distinguishing feature of the current framework is that banks now receive interest on the reserve balances they keep on deposit with the Fed. Unsurprisingly, this feature affects the incentives banks have to lend, altering the relationship between reserves and the broader money supply in ways that make changes in reserves an unreliable indicator of the stance of monetary policy.
Before 2008, the Fed did not pay interest on bank reserves. As a result, banks had little incentive to hold reserves beyond those necessary to meet minimum reserve requirements. Thus, if the Fed increased the quantity of reserves in the banking system, banks would lend most of these reserves, and the money supply would increase by some multiple of the initial increase in reserves.
This multiple, which economists call the money multiplier, depends on several factors, including the public’s preference for cash relative to deposits, the minimum amount of reserves the law requires banks to hold relative to their deposits, and their willingness to hold reserves over that minimum. If banks hold more reserves relative to their deposits, either because they are required to do so by the Fed or because they want to increase their liquidity position, the money multiplier declines, and vice versa.
During the decade before the Fed paid interest on reserve balances, the money multiplier hovered between eight and nine, meaning that if the Fed increased the amount of reserves by $1, the broader money supply would increase by $8 to $9. Since the Fed began paying interest on reserve balances in 2008, the multiplier has hovered between three and four, and, aside from the final months of 2008, has never gone above five. In other words, since 2008, a $1 increase in reserves has tended to increase the broader money supply by $3 to $4.
The collapse of the money multiplier is easy to understand. Once the Fed began paying interest on reserves, banks became more willing to hold excess reserves, as the rate of return they received from the Fed was superior to what they could earn from lending (after accounting for default risk and loan origination costs). As the ratio of excess reserves to deposits increased, the money multiplier decreased because banks opted to hold reserves instead of lending them.
In principle, the Fed has the ability to provide liquidity to the banking system while simultaneously reducing the incentives banks have to make loans. Before 2008, the Fed did not have this option. The Fed faced a potentially painful tradeoff between providing liquidity to the banking system and price stability. Indeed, Ben Bernanke pointed to this tradeoff when he asked Congress for the authority to pay interest on reserves.
Under the current framework, the Fed retains the ability to control the quantity of reserves in the banking system but can now also influence the money multiplier by adjusting the rate it pays on reserve balances. Of course, as long as the money multiplier is greater than zero, increasing the reserve amount will also increase the money supply. Thus, the tradeoff still exists. Nonetheless, the extent to which it is binding has fallen substantially — a point that proponents of the current framework point to as one of its benefits.
The current framework has its downsides. For one, it has resulted in the Fed’s earning substantial losses over the past seven months in its efforts to fight inflation. The current framework has also permitted the Fed to expand its role in allocating credit, reducing the efficiency of the financial system. Nevertheless, if policymakers intend to stick with the framework, they can fight inflation without risking a financial crisis.