The archive · Business Models · Financial decision · 2018
The Narrow Bank: a bank with no loans that the Fed refused to let open
A proposed bank would take deposits and park 100% at the Fed, paying near-central-bank rates — no loans, no risk; the Fed stalled it.
The Narrow Bank
What it had to solve
A would-be bank noticed that the Fed pays banks interest on reserves while most depositors earn far less. Its model: take deposits from institutions that hold big cash reserves but lack banking licenses — money-market funds, foreign central banks, pension funds — hold the money at the Fed, and pay out most of that rate, lending nothing and taking no credit risk.
How it works
In September 2018 The Economist described a bank whose creation the Federal Reserve had stalled, and Hacker News turned it into a 193-comment debate. The idea, known as a narrow bank, was disarmingly simple: take deposits, do not lend them out, hold them at the Federal Reserve instead, and pay depositors most of the interest the Fed pays on reserves.
The target customers were not ordinary savers but institutions with large cash positions and no banking license — money-market funds, foreign central banks, pension funds and the like. With the Fed paying 1.95% on reserves against a fed-funds rate near 1.92%, the bank could offer its depositors a rate ordinary banks would struggle to match, while carrying none of the credit risk that lending creates.
The objections were systemic rather than technical. Commenters quoted The Economist's concern that narrow banks, which lend to neither companies nor individuals, could hamper the effectiveness of monetary policy and unsettle incumbent banks; others worried that if the model caught on, trillions of deposits could leave the fractional-reserve system, shrinking what banks had available for lending. Some dismissed the model as freeloading on a central bank that pays interest only to banks as part of its policy machinery.
The Fed's stall meant the bank could not launch, and the discussion moved from banking mechanics to regulation: whether the central bank should be allowed to refuse a bank whose entire business was holding deposits at the central bank. Commenters pointed to George Selgin's Alt-M posts on interest on reserves and the narrow bank as the place the economics was worked out properly.
Why it lands
- The model stripped banking to its safest core — deposits backed one-for-one by central-bank reserves — so its only real risk was regulatory, not credit.
- Interest on reserves created the opening: the Fed's 1.95% rate was available to banks, and a pass-through bank could extend something close to it to institutions that had no other way to earn it.
- The Fed's concern was that the idea could disintermediate the banking system, so the objection was about consequences for monetary policy, not about the bank's safety.
- The episode showed that when the regulator is also the counterparty the model depends on, the idea lives or dies on policy grounds, not on customer demand.
What it did
As of the September 2018 coverage the bank had not been allowed to operate, so the model became a case study rather than a business. The stall produced a 193-comment Hacker News debate and follow-up analyses by monetary economists, and it put a live question on the table: should the interest the Fed pays on reserves be passed through to non-bank depositors?
What you can take
A business model whose edge is access to a privileged counterparty can be vetoed by it: when the regulator is also the gatekeeper to the model's key input, novelty needs permission, not logic.
Since then
The narrow bank's creation remained stalled through the 2018 coverage, and the argument shifted from banking mechanics to whether the Fed could sensibly refuse a bank whose whole business was holding deposits at the central bank. The 193-comment Hacker News thread and George Selgin's follow-up posts on Alt-M treated the episode as a live policy dispute over interest on reserves, fractional-reserve banking and who should earn central-bank rates. The bank did not launch; the model survived as the reference case for what happens when a novel banking idea depends on the central bank's permission.
Sources
- The Fed stalls the creation of a bank with a novel business model
- The Fed stalls the creation of a bank with a novel business model
spotted an error? The archive wants to know.
Your turn
You just read one. Describe the brief you are staring at, and see who has been given the same problem.
Free account · 3 free questions · no card