US Banks: The Good, The Bad, And The Ugly

Emergency Fed support for US banks is reminiscent of levels seen during the Great Financial Crisis on some measures. Deposit flight from smaller lenders remains a concern to boot. It should not be, but is. That said, there is more evidence pointing to no substantial increase in angst at a regional level in the past week. It’s early days, and things can still take a negative turn. But a silver lining it is.

It has been two weeks since Silicon Valley Bank collapsed. The question at this point is where are we? And more importantly, where are we headed?

We are at the initial stage of some banking sector naval gazing at a minimum

When Lehman Brothers collapsed on 15 September 2008 we had already reached a deep stage of the subprime crisis. The troubled Bear Stearns had been acquired by JP Morgan six months before that, and the Federal Reserve was already in rate-cutting mode.

Currently, we are at the initial stage of some banking sector naval gazing at a minimum, and grappling with concerns about the stickiness of deposits in the smaller banks.

If we add all those small banks up they quickly become systemic, which is why they have our undivided attention. As a result, their deposits are now implicitly insured. However, if there is a deposit flight regardless, then such banks will (have to) close. Not our expected scenario, but that would be a painful process, effectively a managed crisis.

Silicon Valley Bank’s collapse, while not small, was not nearly as widespread as Lehman’s. Its balance sheet was also much less complex, and counterparty linkages were less detrimental.

The same can be said of First Republic Bank, which has received a dig-out deposit boost from the bigger US banks, but still faces significant pressure. Deposit flight is a risk for most banks, particularly for the smallest ones right now; substantial flight has already been seen.

While we can speculate on what’s to come (and we will), let’s first examine the evidence coming from the take-up of the Fed’s exceptional liquidity measures. Here, there is good news and bad news. Let’s start with the bad first.

Data published on March 22 shows that loans extended by the Fed to banks increased to US$354B. That’s a rise from US$15bn before Silicon Valley Bank collapsed. It’s also historically high, in comparison to a peak of US$440bn seen during the Great Financial Crisis.

There are also some differences compared to 2008. Back then, the majority of the loan support was through the Primary Dealer Credit Facility (which provided short-term liquidity to primary dealers) and the Money Market Mutual Fund Facility (assisted money market funds meet some redemptions and increased liquidity in the asset-backed commercial paper market).

Here the focus was on the functionality of the repo and money market funds. The very heartbeat of the system was in threat. A very unique dynamic today.

Fast forward to today, and the allocation of the US$354bn borrowed is very different. Of this amount, some US$110bn has been provided through the discount window. This is the traditional means by which banks can obtain emergency liquidity. This had usually been granted at emergency terms too, but the Fed eased these requirements and extended the available tenor.

The volumes here are reminiscent of the peaks seen during the Great Financial Crisis (peaked at US$117bn in end-October 2008). The good news is the most recent figure is in fact lower than the US$153bn drawn in the previous week, which points to a calming (even if that was a calming from the highest Discount window drawdown on record).

The silver lining is that in the previous week there was no noticeable surge in the use of emergency funding facilities

At the same time, the use of the new Bank Term Funding Program increased to US$54bn, a rise from US$12bn in the past week (the first week of its existence). This program is an alternative to the Discount window, the key differences being a 12-month tenor and better pricing terms.

It serves as a way to liquidate hold-to-maturity bond portfolios, and even sub-par priced bonds receive liquidity back priced at par (the bond redemption value). Basically, the decline in the use of the Discount window was compensated by an increase in the use of the Bank Term Funding facility.

Furthermore, we know from official commentary that a huge chunk of the use of these facilities was for First Republic Bank, and if that’s the case, it’s possible that the rest of the banking system were not big takers. Additionally, if we examine the provision of liquidity and other interactions at Regional Fed Banks, there was no material evidence of heightened concern at a regional level.

Most of the increase was at the New York Fed and the San Francisco Fed in the preceding week, which most likely was an immediate result of Signature Bank in New York and Silicon Valley Bank and First Republic Bank on the West Coast.

A final crucial factor to this is the loans given as a corollary of the Federal Deposit Insurance Corporation provision of support, where all deposits at Silicon Valley Bank and Signature Bank were made whole. That now amounts to US$179bn, which includes the US$36bn rise for the latest week.

On the asset side of bank balance sheets, the sub-par pricing of hold-to-maturity bond portfolios is no longer a concern, as the Fed can liquify them at 100%. Mortgages can also be liquefied through exchanges with Freddie Mac and Fannie Mae, but valuations at the point of exchange are more subject to interpretation.

Liquidity is king right now, which poses a challenge for the banks, as their role is to convert liquidity into ‘assets’ that are prone to illiquidity and/or price uncertainty. At the moment, many small banks require that liquidity back, which complicates matters.

It does not have to go wrong though, and examining the hard evidence away from financial market vaguaries, evidence over the past week indicates that things are at least not taking a drastic turn for the worse here in the US. Although this could change, the latest week has in fact seen some stabilization.