The Big Idea
Ten days that shook US banking
Steven Abrahams | March 17, 2023
This document is intended for institutional investors and is not subject to all of the independence and disclosure standards applicable to debt research reports prepared for retail investors. This material does not constitute research.
In the 10 days since SVB collapsed, banks and their regulators have been caught in an extraordinarily difficult balancing act to create stability in the US bank deposit base. Washington guaranteed the deposits of SVB and Signature Bank but to all others offered only Fed liquidity backstops—facilities to let solvent banks borrow easily to cover possible withdrawals. Political fallout from bank support during the Global Financial Crisis has almost certainly left Washington wary of guaranteeing all bank deposits. And borrowing from the Fed and the FHLBanks since SVB suggests ongoing concern about reallocation of deposits within and possibly outside the banking system. This almost certainly will have material impact in the short run on bank behavior and markets, and likely in the long run as well.
Although Fed or other government action could easily bounce events in new directions, a first, rough draft of conclusions from the last 10 days suggests important ways that bank balance sheets and markets seem likely to change:
- Banks will cling more closely to liquidity, tightening credit and investing
- Banks will take material mark-to-market interest rate risk at their peril
- SVB and its aftermath will complicate Fed policy
A few thoughts on each and on a grab bag of other possibilities.
Banks will cling more closely to liquidity, limiting lending and investing
Liquidity has arguably become the single highest priority at most banks for now. Government guarantees for all deposits at SVB and Signature Bank along with the Fed liquidity programs should have let most depositors just go about their business, at least according to the central banking playbook before SVB. But that apparently is not the case. In the week ending March 15, the Fed provided $292 billion in cash to banks through various programs and the FHLBanks issued $356 billion in debt, the vast majority of proceeds likely passed through to member banks. The same week, bank reserve balances at the Fed went up $435 billion. All of that suggests intense actual or anticipated movement in deposits.
Beyond the Fed and FHLBanks, banks in the last week have knocked on new doors to ask about repo and other forms of funding. Interest in repo at brokers anecdotally has picked up significantly in part to diversify sources of funding. And banks thinking strategically about funding have started exploring securitizations that would issue only highly rated senior debt to fund loan portfolios while keeping the assets on the balance sheet—attracted by debt that cannot run and that matches the cash flows of the asset. Some banks have continued to fund in the market for brokered certificates of deposit despite extraordinary wide spreads, likely a signal of new value attached to funding with defined maturities.
With received wisdom about deposit behavior now up for grabs, lending and investing becomes at least temporarily more complex. At the margin, keeping cash in reserves at the Fed becomes much easier. And bank regulators anecdotally have long viewed reserves as the gold standard for liquidity. The latest Fed Senior Loan Officer Opinion Survey showed lending standards tightening and reported lenders’ plans to tighten further this year. SVB likely turns that up to 11. Bank loan and investment portfolios at least temporarily look likely to decline with cash assets rising.
If deposits are getting reallocated within the banking system, it could have variable effect on different sectors of credit. My colleague Tom O’Hara, following the lead of Apollo’s Torsten Slok, highlights the lead role that smaller banks have taken in the last few years in lending to consumers and against residential and commercial real estate. Bigger banks have led on commercial and industrial lending. It deposits flow to bigger banks that choose not to lend the same way in the markets of smaller banks, consumer and real estate credit could tighten first.
If deposits are getting reallocated outside the banking system into money market funds or other near-cash assets, then more lending will likely flow to nonbank investors such as private debt and direct lenders. Insurers and pension funds have steadily increased their holdings in private debt in the last decade and could become a bigger source of funds.
It is likely that SVB and its aftermath will leave a long and lingering preference for liquidity on bank balance sheets, although probably not as intense as it is now. But bank management, boards, regulators and debt and equity investors have long memories.
Key bank balance sheet implications in summary:
- With deposit behavior in disarray, stable funding with defined maturities becomes more valuable, tilting bank funding toward retail or wholesale certificates of deposit, FHLBank advances or securities repo
- Funding with debt matched to fixed-rate assets becomes more valuable, such as securitizations of senior classes of MBS, CMBS or ABS that remain consolidated on balance sheet
Key market implications in summary:
- Bank preference for cash assets should remain high for the next few months and elevated in the long run, constraining loan and investment portfolios
- More lending flows on the margin to nonbank lenders able to match fund or bear net interest rate risk
- Over time, funding securitizations of bank assets show up more often
Banks will take material mark-to-market interest rate risk at their own peril
Bank management, boards, regulators, and debt and equity investors seem sure to draw the lesson from SVB that mark-to-market interest rate risk can kill. There is nuance to that, of course. But nuance tends to get lost in the long run. And stories get simplified. The simple story of SVB is that mark-to-market interest rate risk sank the bank. When higher rates drove down value of its securities, the bank optically looked insolvent—optically for reasons I’ll get to. And when SVB tried to raise equity to rebuild capital, deposits ran. Boom. Down goes Frazier.
The cautionary tale of SVB will likely encourage banks to keep mark-to-market exposure short, whether it is in available-for-sale (AFS) or held-to-maturity (HTM) portfolios. As current intense liquidity concerns ease and banks start to think about investing, US banks will likely look more like European banks and lean toward short duration investments or hedge long duration into short.
The irony is that SVB seems to have followed all the rules. The bank followed the rules and reported the risk in its AFS portfolio through GAAP financials. Regulators followed the rules and did not count losses in AFS against the bank’s regulatory capital. The bank also followed the rules and kept the risk in its HTM portfolio out of its GAAP financials. And regulators kept losses in HTM out of regulatory capital. The GAAP financials consequently showed capital on par with peers. Move along folks, nothing to look at here.
But one important party looked behind the GAAP curtain: Moody’s. Moody’s saw that mark-to-market losses in AFS and in HTM together had almost eliminated SVB’s tangible capital, and it prepared to downgrade the bank. When the bank tried to raise capital, the lights went out.
Many banks will argue in their frustration that SVB and others might have looked solvent if you just marked to market the loans along with deposits and other liabilities. The deposit side is especially important. After all, market interest rates have moved up much faster than deposit rates, making those deposits steadily more valuable. To slightly oversimplify, a bank is long assets and short deposits. Rising rates lowers the value of many assets but raises the value of many deposits. The two just might offset. But neither GAAP financials nor regulators mark deposits to market. The old debate about marking the full balance sheet will resurface but will almost certainly not get resolved.
All of this, too, has implications for bank balance sheets and for markets.
Key balance sheet implications in summary:
- Tangible capital will matter more than regulatory capital to everyone because that’s the stuff of solvency
- Managing bank duration of equity with instruments that get marked to market becomes harder, encouraging banks to manage more of the duration on the liability side of the balance sheet
Key market implications in summary:
- Bank demand for short assets should get much stronger, all else equal, and steepen the yield curve and tighten spreads in agency floating-rate debt, CMOs, Ginnie Mae HECMs and other assets
- Demand for assets easily hedged under GAAP should tighten spreads in assets with stable cash flows such as agency debt, agency CMBS and certain classes of CMOs
- Assets such as agency MBS and callable muni debt with large bank holdings but variable cash flows—cash flows complex to hedge under GAAP—should widen, or, through a different lens, floating-rate or short MBS cash flows should tighten while long fixed-rate cash flows should widen
- Bank activity in the swaps markets should pick up significantly
SVB will complicate Fed policy
The impact of SVB on bank lending has already complicated Fed policy. My colleague Stephen Stanley does an excellent job of outlining the Fed’s challenges in balancing clearly persistent inflation against the prospects of tighter credit. Tighter bank lending had already started to do some of the Fed’s work for it in slowing down the economy, and the Fed will have to figure out how deflationary SVB and its aftermath has been. Some senior commercial bankers think the Fed should skip a March hike to see more of the SVB fallout.
Heightened bank sensitivity to liquidity could also complicate QT. The last round of QT ended in September 2019 after repo rates shot up and showed excess reserves had run too low. SVB may move bank preference for liquidity well above levels equivalent to September 2019. Since the only practical way to tell when bank reserves get too low is to see stress in funding markets, the Fed may decide to pull up on QT earlier than before.
Complications in Fed policy may not have direct implications for bank balance sheets, but they do have key market implications in summary:
- Fed funds may follow a lower path than the market priced in early March if the Fed concludes that SVB and its aftermath are deflationary
- QT may end earlier than expected, taking supply pressure off Treasury and MBS markets
A grab bag of other likely implications
- The FHLBank system will become much more active—eventually. Bank demand for liquidity and heavy FHLBank debt issuance sets off a spin cycle of activity in agency MBS, CMBS and swaps and options markets. But before that starts to speed up, the FHLBanks will want to see how much of the bank demand for liquidity persists—the amount of overnight, 1-, 3-, 6-month and longer debt that continues to get rolled. After all, the FHLBank system can just hold cash, too.
- The legacy of the GFC could complicate disposition of SVB and others. One unexpected outcome of the acquisition of Bear Stearns by JPMorgan and of Countrywide and Merrill Lynch by Bank of America and other transactions was a long tail of legal liability. Admittedly, most of those revolved around origination and securitization of mortgage loans. But the memory of liability could slow disposition or lower the price or both. The liability here seems more likely to come from shareholders or creditors. But memories of GFC linger.
- The implications of a SOFR index during a crisis have become clearer. The spread between 3-month LIBOR and 3-month SOFR has widened in the aftermath of SVB, likely the last opportunity for the market to see LIBOR widen before it disappears before the end of June. The use of the riskless SOFR benchmark means that widening of credit spreads in future crises will not happen in part through the widening of the index. It will have to happen more through the drop in price of the indexed security. The price volatility of SOFR-indexed assets has shown itself.
* * *
The view in rates
A week before SVB went into receivership, OIS forward rates projected fed funds would peak around 5.50% by August. And on the date of receivership, OIS forwards priced a peak of 5.30%. And now, a week later, OIS sees a peak at 4.83%, anticipating no more hikes in Fed funds. The Fed could do a hawkish pause on March 22, leaving fed funds unchanged explicitly to see the impact of SVB on bank credit but warning of its commitment to beat persistent inflation. Or the Fed could do a dovish hike, raising fed funds 25 bp but indicating willingness to pause on signs of deflation in the wake of SVB. Stephen Stanley is anticipating 25 bp at the March FOMC, so a dovish hike looks more likely.
Fed RRP balances closed in the last few days at $2.11 trillion, down $82 billion from a week ago. Despite apparent Fed expectations that attractive rates in money markets will draw funds away from the RRP, it clearly is not happening yet.
Settings on 3-month LIBOR have closed Friday at 500 bp, down 14 bp on the week. Setting on 3-month term SOFR closed Friday at 485 bp, down 27 bp on the week. The spread between these benchmarks is 23 bp wider since March 7.
Further out the curve, the 2-year note closed Friday at 3.84%, down 75 bp on the week. The 10-year note closed at 3.43%, down 25 bp on the week.
The Treasury yield curve has finished its most recent session with 2s10s at -41 bp after hitting -108 bp on March 8. The 5s30s finished the most recent session at 13 bp after hitting -46 bp.
Breakeven 10-year inflation finished the week at 210 bp, down 18 bp in the last week. The 10-year real rate finished the week at 133 bp, down 7 bp in the last week.
The view in spreads
Volatility has rocketed in the aftermath of SVB, with the MOVE index reaching levels last seen in 2008 and 2009. Nominal par 30-year MBS spreads to the blend of 5- and 10-year Treasury yields closed Friday at 156 bp, down 4 bp on the week. Par 30-year MBS TOAS has closed Friday at 50 bp, tighter by 1 bp on the week.
The view in credit
Most investment grade corporate and most consumer balance sheets look relatively well protected against the likely impact of Fed tightening. Fixed-rate funding largely blunts the impact of higher rates on both those corporate and consumer balance sheets, and healthy stocks of cash and liquid assets allow these balance sheets to absorb a moderate squeeze on income. But other parts of the market are funded with floating debt. Leveraged and middle market balance sheets are vulnerable, especially with the sharp tightening of bank credit in the wake of SVB. As for the consumer, subprime auto borrowers, among others, are starting to show some cracks with delinquencies rising quickly. Some commercial real estate funded with floating-rate mortgages have started to show some stress, too.