Cochrane's speculates a bit about the Fed's motives near the end. He doesn't think they're outright evil, in cahoots with the incumbent commercial banks, etc. He thinks it's a misguided attempt to cross-subsidize the lending activities at the current commercial banks. If the super safe narrow bank draws away a lot of the common depositors, the commercial bank will need to get more other (more expensive) funding sources (bonds and other loans) which can make mortgages, business loans etc more expensive.
You're in good company if you find that outrageous. But the (speculative) reasoning is not so much to keep them competitive, but to keep borrowing cheap. In the real world, lots of people want to borrow money for 3, 5 or 10 years for their business or 10, 20, 30 years for their mortgage (and preferably at fixed rates), while very few people want to lend out money for such long terms at fixed rates. So the way banks handle this is pool together lots of small short terms loans like deposits and count on the observed regularity that they usually don't withdraw their money simultaneously (SVB notwithstanding). If the relatively stable and cheap small deposits are all going to narrow banks, how are the lending banks going to fund the longer term loans?
My understanding so far is that these deposits are often overburdened. In part due to speculative investments.
From a perspective of someone who understands very little of these matters, it seems like responsibilities are shuffled around and the whole structure is unclear.
There are other ways to solve the problem you describe right? For example credit unions come to mind.
The narrow bank restricts the Fed's ability to hawkishly raise the interest rate. It won't be able to get away with gross market Vs policy mismatches anymore. I mean think about it, the Fed makes an unintended policy error and raises the interest rate far above what banks can pay, everyone goes to the narrow banks. If the interest rate is too low nobody goes to the narrow bank. So the Fed essentially would have to perfectly choose the optimal interest rate which it probably can't do. It always overshoots or undershoots.
The narrow bank would be safer than US Treasury Bonds. In a financial crisis similar to 2008 this would amplify chaos as money drained from all other investment classes into the narrow bank at a time when the government probably needs low interest rates on their debt to solve things.
I don’t know how long this will last, but I started doing this recently with Vanguard’s VUSXX (short-term Treasury) fund when I realized it had significantly higher after-tax yield than high-interest savings accounts without the hassle of manually rolling over T-bills, and it looks like many other people have been moving in this direction.
I’m curious about this as well, because the top comment (TNB takes over an existing bank that already has this account) seems like a perfectly viable alternative. Given that they had the resources to get to this point in the first place, was there a reason why this wasn’t an option?
edit: It was Scott Sumner, commenting on Cochrane's blog, who speculated that the motive might be cross-subsidizing the normal bank lending activities: https://www.econlib.org/why-does-the-fed-oppose-narrow-banki...