Last week, I suggested the Fed is attempting a difficult tightrope act with regard to deposit outflows and the desirability of engineering tighter credit conditions in the service of fighting inflation.

In “Deposit Flight And More Fed Hubris,” I sketched the contours of what some might describe as an absurd situation. A Fed facility is contributing to instability in the banking sector, and that instability last month prompted the Fed to create a new facility to prevent the problem from becoming systemic.

That isn’t quite as counterproductive and absurd as it sounds when put in the context of inflation realities (and when you consider that, as designed, the “problem” facility is just an optimization tool for managing rates), but it almost is.

By now, everyone has seen some version of the chart depicting falling deposits and rising money market AUM.

Simply put: Depositors are fleeing banks for the safety and yield of government money funds. That deposit flight went into overdrive last month when SVB’s dramatic, overnight collapse put a spotlight on uninsured deposit balances.

Government money market funds, like their name suggests, invest in government paper and repos, which means there’s no risk. No credit risk, and no duration risk either. Yields on those funds are the highest in ~15 years.

The figure below is likewise familiar to most serious market observers, but perhaps not to everyone else. It’s just usage of the Fed’s RRP facility.

That facility currently pays almost 5% (annualized, obviously). The Fed is “borrowing” money, and paying triple (at least) what depositors can get from most FDIC-insured accounts and bank products.

If you’re wondering why anyone with access to that facility would eschew the opportunity to “lend” the Fed dollars (collateralized, by the way, just in case you don’t trust the Fed to give you your money back) at 4.8%, the answer is they wouldn’t. Which is why there’s been $2 trillion parked in there pretty much every day since last June. (There’s quite a bit more to this discussion, but delving too far into it risks getting mired in the SLR debate, which is a black hole.)

On Wednesday, Wall Street Journal “Fed whisperer” Nick Timiraos highlighted this dynamic in a fairly prominent article, underscoring the extent to which this is probably a topic of growing concern for policymakers. As Nick wrote, “more than 40% of money market fund assets are invested in the reverse repo facility.”

The RRP tsunami began in 2021 amid a confluence of factors, including, but not limited to, a rate tweak and the expiration of the SLR exemptions. The amount of cash sloshing around the system as a result of the pandemic stimulus is the backdrop for this entire phenomenon, and yes, that’d be the same stimulus which eventually drove up inflation, forcing the Fed to hike rates at what, by the standards of the post-Volcker era anyway, looked like a cartoonish pace.

Those rate hikes made the RRP more attractive, and ultimately contributed to the second-largest bank failure in American history. That bank failure put the spotlight on uninsured deposits, whose owners suddenly realized that government money funds are safer and higher-yielding than uninsured bank accounts. Money fund assets subsequently surged even more (to $5.2 trillion as of last Wednesday, prompting some to use the word “bubble,” which I think is a misnomer), raising the specter of even higher RRP usage.

Hopefully, you can understand why this looks somewhat farcical to critics, even as a more level-headed assessment says that by encouraging competition for deposits, the Fed should be able to compress bank margins and thereby curtail credit creation to Main Street with the ultimate goal of slowing inflation.

That brings us quickly full circle. This is a tightrope walk. Obviously, deposits are still elevated, so it’s not as if the system can’t afford to “lose” some, but at the same time, money market funds are no longer at a loss figuratively (you don’t have to scour the globe for yield anymore) or literally (ZIRP is over, probably for good, so the industry isn’t facing the same operational challenges it had two years ago).

If you can take some of the stress off banks by lowering the rate you’re paying money funds to park cash with the Fed, that might not be the worst idea, particularly if the attractiveness of the facility is contributing to the dynamics that are compelling banks to utilize the newly-created term-funding facility.

The arguments against favoring bank deposits seem to be couched almost entirely in normative terms. Tweaking the reverse repo rate isn’t going to impair the Fed’s capacity to control overnight rates, and I doubt officials would be comfortable with the current dynamic (i.e., ongoing deposit drain to money funds, with the effect of forestalling RRP decline) in perpetuity.

When considered in the context of ongoing QT, this is a bit labyrinthine, to put it politely. According to some observers, it’s all working like it’s supposed to, but you’d be forgiven for casting a wary eye at the plumbing blueprints, which seem to get more complex by the year.