Concern Turning from Liquidity to Credit Quality

TL;DR: The Federal Reserve data in the past week offers further reassurance that near term liquidity concerns have receded, but the flow of funds from bank deposits to money market funds remains elevated. Absent a new flare up of liquidity worries, we anticipate that attention will increasingly turn toward credit quality risks. While First Republic continues working towards some resolution, reports indicate some (hitherto modest) attrition in both Advisors and AUM in the wealth management segment.

Wealth management attrition at First Republic. Deposit outflows pose a more severe risk to banks than AUM outflows; the former sit on the balance sheet, and rapid outflows threaten a bank’s survival, while the latter, as recently seen at Credit Suisse, are segregated and generally threaten only (“only”) earnings potential and, by extension, enterprise value. Nonetheless, in a highly scalable business like wealth management that threat to earnings and enterprise value can be material, which makes activity at First Republic worthy of attention. Wealth management trade press report an early wave of departures of both clients and advisors. At least seven FAs are known to have left since the start of the banking crisis; that’s scarcely 3% of First Republic’s total, but in just three weeks. It is also a sharp reversal of the trend of recent years; First Republic added 13 experienced teams last year, managing a combined $12 billion in AUM, and 11 teams in 2021. With luck, a transaction of some sort, be it a restructuring or a sale, will cauterize this wound.  

March 29 Federal Reserve balances suggest stabilization. Figure 1 highlights the key data points from last week’s release of Federal Reserve balances (Release H.4.1). The data suggests a continued return of stability; combined bank usage of the Discount Window and the BTFP fell 7% to $153 billion, with a continued mix shift from the Discount Window to the BTFP.  That mix shift is partially explained by Figure 2; over its short life, the BTFP rate has averaged 20bps below that of the Discount Window. The other part of the explanation, of course, is that the BTFP values collateral at par, which on some MBS means actual LTVs exceeding 110%. Other credit extensions—loans to failed banks—was flat at $180 billion.

Figure 1: Federal Reserve Balances

Source: Federal Reserve

Figure 2: The BTFP Daily Rate Has Averaged 20bps Below The Discount Window Rate

Source: Federal Reserve

Money market assets see third straight jump. After two straight weeks of 2%+ increases, the surge of cash into money market funds decelerated last week, but still rose by $66 billion, or 1.3%. Since March 8, the aggregate inflows to money market funds total $304 billion (6.2%), about equal to banking deposit outflows for the first two weeks since March 8 (we only have data for the first two weeks as of this writing). 

Figure 3: Money Market Fund Assets ($Trillions)

Source: FactSet

Decline in total commercial bank deposits remains elevated, but decelerating. The flow of deposits out of banks decelerated in the week ended March 22, but remains very elevated; total deposits declined by $126 billion, or 72bps. Deposits at small banks fully stabilized, however, eking out a $6 billion (11bps) increase for the week. At large banks, the initial surge of deposits during the first week of the banking crisis reversed itself, with deposits dropping $90 billion (84bps) for the week, leaving that cohort down $23 billion (21bps) since the commencement of festivities on March 8.

Figure 4: Small Bank Deposits Stabilized While Large Bank Outflows Leapt

Source: FactSet

Figure 5: Total Bank Deposits by Week ($Trillions)

Source: FactSet

Total loans dipped slightly. While total loans are up $46 billion (38bps) since the start of the banking crisis, they dipped last week by $20 billion (17bps). This Friday’s release may offer some hint of whether this was an aberration or the first sign of banks pulling back on lending, but that will be the only small signal we get before earnings season begins. In terms of a bogey for earnings season, total banks loans grew at a sluggish 6% annualized rate in the first quarter (through March 22). Notably, commercial real estate loans—a focus of the recent wobble in Deutsche Bank as well as many subsequent headlines—have grown slightly faster, up an annualized 8% quarter-to-date, and even managed a nominal increase last week.

Figure 6: Total Loans Declined In The Most Recent Week

Source: FactSet

Figure 7: Total Bank Loans by Week ($Trillions)

Source: FactSet

What We’re Watching

Key data releases. As covered above, weekly FRB releases covering the Fed balance sheet, and aggregate US bank balance sheet, as well as the Investment Company Institute’s weekly Money Market Fund Assets release, will offer insight into the pace of normalization (or lack thereof). The coming week should also see quarterly data from Edmunds on auto finance, an area of consumer credit that seems to hold particular risk in this cycle. Looking out a bit further, the Federal Reserve’s quarterly Senior Loan Officer Opinion Survey (SLOOS) should be in the hands of said loan officers by now, with results due to be released in about a month; since this survey will have gone out after the onset of the banking crisis, the results may offer insight into the extent to which banks’ willingness to extend credit has been affected.

First quarter earnings. We are now two weeks away from the start of banks’ first quarter earnings releases. JP Morgan Chase, Wells Fargo, Citigroup and PNC all scheduled to release on April 14, though we conjecture that, in contrast with recent years, First Republic may not report results on the first day of the season. We would anticipate some common additions to financial disclosures in bank earnings releases, such as uninsured deposits, but more incremental disclosure may come from earnings calls, with investors likely to use Q&A to probe both for color on current trends, and insight into longer term strategic shifts reflecting known unknowns such as forthcoming regulatory changes, and known knowns, such as the planned July introduction of FedNow, a real-time payments system which, for all its potential benefits, would only have accelerated March’s bank runs.

Credit ratings and metrics. We continue to watch for further action by ratings agencies, of course, as well as spreads on bank debt and credit default swaps.

Washington. We continue to await clarity on specific changes to regulatory standards, which will presumably entail increased scrutiny of areas like deposit concentrations (perhaps bringing this more in line with corresponding scrutiny of loan concentrations), and also extend greater overall scrutiny to smaller banks than is currently the case.

Where Do We Go From Here?

In the Spring of 2022, we wrote the following:

“A soft landing remains possible, but probably not the way to bet. Soft landings are rare beasts; the last was in 1994-95, though to be fair, there are some rough parallels. From 1989 to 1991 the U.S. experienced what was, at the time, the worst banking crisis since the great depression, accompanied by a (comparatively shallow) recession. The Fed’s response was lower, longer; QE had not yet been imagined, but the Fed Funds rate stood at 9.75% in mid-1989, and the Fed kept loosening until it hit 3% in late 1992, where it would sit for another year and a half. The tightening cycle that began in early 1994 wasn’t anticipated by the markets, and catalyzed some significant dislocations – a brutal year for fixed income and some high profile derivatives-related blow-ups – but there was no recession. Yet 2007-08 was a much larger financial crisis, the lower, longer policy response has been much greater in both scale and duration, and inflation is running well ahead of 1994. As a practical matter, there has been a much larger and longer window for the misallocation of capital, and policy responses to inflation will tend to bring those errors to the surface.

"First-order effects are easily foreseeable. In this cycle, banks get their dessert before their spinach; rising rates are manna from Heaven (for now). Yields are rising, deposit betas are lagging, and credit remains quiescent; quarterly hits to AOCI are a small price to pay for that combination. As rates rise further, yields will also rise, but deposit betas can be expected to rise, and more importantly, credit will deteriorate as over a decade’s accumulated misallocation of capital is exposed. In M&A, financing is still available, if a bit more costly, and target prices are now appreciably lower, but as the old saw goes, companies are not bought; they are sold – and seller expectations adjust downward much more slowly than valuations. There will be some sales under duress but broadly the widening bid-ask spread will curtail activity. Capital raising in both public and private markets dries up, with certain exceptions. This is all bog-standard, and where guidance in the financial sector is very different, well, just give it a bit of time. Some of that is motivated reasoning, but it is also a function of the very robust starting point of the decline. The housing market has been extremely strong, so the early sign of weakness there is merely reduced foot traffic, but the pace and degree of deterioration there will be informative.

"Discontinuities are difficult to anticipate. Readers who have joined any of your humble correspondent's "office hours" calls in the past couple months have endured his droning on about the risk of not just another recession, but the potential end of a 40-year phase of financial history characterized by declining rates and disinflationary growth drivers. That doesn’t mean it’s at hand; the future is not yet written. Perhaps, as noted above, we get no recession at all. But tail risks appear elevated, which suggests an environment in which many assumptions that have proved durable over the years are liable to be revealed as stale.”

In the Spring of 2023: A soft landing is still possible, but still doesn’t seem like the high-probability bet (certainly not if the inverted yield curve in Figure 8 is anything to go by). Rising rates provided the expected early margin relief to banks but it seems safe to say that we are past the rising-rates-are-good-actually phase of the cycle, even for banks that didn’t wildly mismanage duration risk. We would be inclined to categorize the re-emergence of bank runs as a discontinuity, and while some market participants may have been sufficiently prescient to anticipate it, we make no such claim. Taken as a whole, however, we would suggest that the trajectory of the economy and financial sector during the past year of rising rates has pretty closely tracked what a knowledge of financial history would have led one to expect.

Awaiting the credit cycle. The rate hike phase of this cycle may or may not be over—as Figure 9 shows, the May 3 FOMC meeting looks like a coin flip at the moment—but the credit cycle is in its early stages. Commercial real estate is currently attracting investor attention, but causes for concern continue to emerge across loan types. We would highlight auto finance is one area worth watching specially closely due to several COVID-related dynamics layered on top of the garden-variety cyclical ones. That doesn’t mean a full-blown banking crisis is inevitable; banks are certainly better capitalized now than they were heading into 2008. On the other hand, money was cheaper, longer than in the expansion preceding 2008. As Warren Buffett has observed, “You don’t find out who’s been swimming naked until the tide goes out.” So we wait and watch.

Figure 8: The Yield Curve Remains Inverted

Source: FactSet

Figure 9: The May 3 FOMC Meeting IS A Coin Flip According To Fed Funds Futures

Source: FactSet

This blog post is for informational purposes only. The information contained in this blog post is not legal, tax, or investment advice. FactSet does not endorse or recommend any investments and assumes no liability for any consequence relating directly or indirectly to any action or inaction taken based on the information contained in this article.