Earn next to nothing on your bank account but get more than 1% cash-back on every dollar you spend using a rewards card? If consumer banking is strange these days, the institutional banking business is even more bizarre.

Banks are now charging their big institutional customers to keep money on deposit—and these so-called negative interest rates are forcing liquidity out of the banking system. The irony is that this is the result of postcrisis financial regulations that are supposed to ensure that banks remain liquid.

J.P. Morgan Chase recently announced it may charge institutional clients as much as 5.5% on certain deposits, in an effort to push as much as $100 billion of these deposits out the door. Other U.S. banks already charge institutions negative interest to hold euro deposits.

What accounts for this strange behavior? Once a U.S. bank accepts a deposit, it must pay insurance premiums to the Federal Deposit Insurance Corp. Dodd-Frank changed the way the premiums are calculated, and the upshot is that insurance costs rise on nearly every new dollar deposited, even though only $250,000 of each customer account is insured. Insurance premiums on average are about 20 basis points on each dollar deposit, although they can be as high as 45 basis points for a large bank.

Deposits have to earn enough to cover deposit insurance and other bank operating costs. The money can be held as reserves at the Federal Reserve but it will earn only 25 basis points, not enough to offset insurance and other operating costs. Banks can also invest the money in loans or higher-yielding securities, but there is a catch. New regulations will soon restrict how banks may invest large institutional deposits.

These regulations have two parts. First, banks must estimate how much of their funding will “run” in a 30-day stress period. Regulatory guidelines require banks to assume that virtually all large unsecured financial-institution deposits will run.

Second, banks must hold high-quality liquid assets (HQLA) equal in amount to the funds that run under stress. Since all large banks must meet this requirement, it generates strong demand for HQLA—such as U.S. Treasurys or the sovereign debt of countries with a zero risk weight of default—that further reduces the near-zero yields on these assets. Given deposit insurance premiums, other operating costs and extremely meager yields on HQLA, it makes no business sense for banks to accept large institutions’ deposits.

How did this happen? The popular narrative is that the financial crisis was caused by investor “runs”—i.e., when large volumes of uninsured deposits and investors in commercial-paper simultaneously moved money out of banks and other financial institutions. Bank regulators embraced this narrative and the Basel Committee on Banking Supervision crafted new international rules to limit banks’ ability to fund themselves with short-term liabilities. These new rules have been adopted by U.S. bank regulators and they are set to be phased-in over several years.

But there is a huge glitch. The Basel Committee apparently never considered the possibility that interest rates would remain at or near zero for many years and that, in a zero-rate environment, the new liquidity rule would make it uneconomic for banks to hold large institutional deposits unless they charged these customers negative interest rates. The Basel liquidity rule was supposed to ensure that banks have adequate liquidity, but instead it is encouraging banks to reject liquid deposits.

Minimum prudential standards for bank liquidity are reasonable, but limiting short-term lending to banks can create an illusion of financial stability. When short-term bank investors balk, they are signaling a problem regulators may not be aware of. For example, when investors became concerned about the true value of the subprime mortgage-backed securities held by Bear Stearns in March 2008, Bear Stearns’s short-term funding dried up, forcing the firm’s sale to J.P. Morgan Chase. The run forced the institution to recognize a deeper problem that sooner or later would have surfaced.

The possibility of a short-term funding run is perhaps the strongest form of market discipline for a financial institution, and runs provided an accurate warning in the past financial crisis.

Only a meltdown of the asset-backed commercial-paper market in summer 2007 told the Federal Reserve there might be a significant subprime-mortgage problem. It took more than a year of additional runs before the Fed appreciated its magnitude.

Share prices of actively traded financial institutions may also signal trouble, but short-term funding runs require immediate remedial actions to replace the funding, or the institution must sell assets to raise funds to continue operating. Limiting a bank’s ability to fund its operations with short-term debt may buy time for regulators to fix institutional problems they identify, but regulations that drive liquidity from the banking system present their own problems. What are banks for if not to take deposits and lend them to foster economic growth?


Mr. Kupiec, a resident scholar at the American Enterprise Institute, is the former director of the Center for Financial Research at the Federal Deposit Insurance Corp. and former chairman of the Research Task Force of the Basel Committee on Banking Supervision.