EASING LIQUIDITY RULES WON'T WORK IF THE FED KEEPS PAYING BANKS TO HOLD RESERVES

The following information was released by the Cato Institute:

Jai Kedia

Improving bank liquidity rules is a focus for some of Washington's top policymakers. In prepared remarks for a March roundtable, Treasury Secretary Scott Bessent called bank liquidity "the next big-ticket item" in the administration's regulatory reset. Fed Vice Chair for Supervision Michelle Bowman has warned that the current framework leaves banks "hoarding high-quality liquid assets rather than lending."

They are correct: liquidity coverage ratios (LCRs) are overly restrictive, and the entire banking regulatory infrastructure is too archaic, complex, and burdensome. However, simply easing the LCR won't significantly reduce banks' demand for reserves, because the Fed still pays banks interest to hold them through the interest on reserves (IOR) program. Liquidity reform that leaves IOR in place is likely to fall short of its goals, which is one more reason Congress should revoke the Fed's authority to pay IOR.

LCR Requirements Are Too Restrictive

The LCR requires large banks to hold enough high-quality liquid assetssuch as Treasury securities and reserve balances at the Fedto cover projected net cash outflows over a 30-day stress period. Bessent's remarks note that safe assets now account for about 25 percent of large banks' balance sheets, a big increase from roughly 10 percent before the 2008 crisis. Bowman made a parallel case about capital rules at Cato days later, arguing that excessive requirements "impair the banking system's fundamental function of providing credit to the real economy."

In both cases, the concern is that rules are calibrated too conservatively. In other words, the rules require banks to hold too many liquid assets at the expense of making loans. As a result of the rules, Americans can't borrow what they need to open businesses or buy cars and homes.

IOR Also Increases Demand for Reserves

The LCR is only part of what drives banks' demand for reserves. Since 2008, the Fed has paid banks interest on their reserve balances. Before IOR, holding reserves carried an opportunity cost for the bank: every dollar parked at the Fed was a dollar not earning interest elsewhere. As a result, banks kept as few reserves as possible. IOR removes most of that cost.

The IOR rate, currently almost 4 percent, is close to other short-term market rates, meaning banks don't lose much by holding reserves to earn IOR rather than other short-term assets. Moreover, IOR comes with the backing of the Fedthe institution that is literally in charge of base money circulationso it is as risk-free a source of interest as exists in the US.

Reserves now earn roughly what Treasury bills do, carry almost no risk of losing their value, and can be used for interbank payments immediately without being sold. (The last of these facts is the main reason why the two assets are not interchangeable.) Banks currently hold about $3 trillion in reserves. If regulators loosen the LCR requirements, banks may trim their Treasury holdings, but they would still have little reason to shed reserves that pay such a competitive, risk-free return. As I have argued before, removing the IOR incentive would free banks to devote more of their balance sheets to lending.

Congress Should End the IOR Program

There are many reasons to end IOR beyond its strain on liquidity. For instance, Fed Chairman Kevin Warsh has publicly committed to shrinking the Fed's balance sheet, but that can only happen if banks want to hold fewer reserves at the Fed. (A large balance sheet funded by interest-bearing reserves also carries costs of its own.) Moreover, the Fed lost nearly $200 billion in 2023 and 2024 paying IOR to large banks, money that would otherwise have been remitted to the Treasury. Every increase in the Fed's rate target means larger payments to banks and further operating losses, creating an incentive for the Fed to keep its rate target too low. The IOR framework also allows the Fed to expand its balance sheet without the usual inflationary consequences, creating a backdoor channel to finance government spending without proper appropriations. And the framework has almost eliminated the private interbank lending market, removing numerous clear price signals about the health of the financial sector.

Bessent and Bowman are right that post-crisis liquidity rules are too restrictive, and loosening them is a good idea. But policymakers should not expect such changes to unlock significant new lending while IOR remains in place. The best economic outcome requires policy coordination. Regulators should ease the liquidity rules, the Fed should shrink its balance sheet, and Congress should set a firm statutory sunset for the Fed's authority to pay interest on reserves. Otherwise, banks will keep holding reserves for the simplest of reasons: the Fed pays them to.