The Sudden Demise of Silicon Valley Bank

VIEWS EXPRESSED IN THIS ARTICLE ARE PERSONAL

Silicon Valley Bank (SVB) was established in 1983, in Santa Clara - California, to provide banking services to the mushrooming technology ecosystem that was taking hold in the Silicon Valley. SVB became one of the several new banks launched in California that year.

SVB offered technology companies a range of banking products and sevices: deposit services, loans, investment products, cash management, commercial finance and fiduciary services. Because tech start-ups tend to have more cash in the beginning, and most of the SVB’s income was traditionally made on the deposit side of the business, as a good portion of these deposits were non-interest bearing.  Driven by a boom in venture capital funding, many of Silicon Valley’s customers became flush with cash over 2020 and 2021. Between the end of 2019 and the first quarter of 2022, the bank’s deposit balances than tripled to $175 billion (aided by a small acquisition of Boston Private Financial Holdings).

 

By 2022, SVB the 16^th largest bank in the US and to give a comparison, the deposits of SVB were of similar magnitude to that of HDFC Bank in India!  SVB was also rated a buy by many equity analysts until a few weeks back and was also listed in America’s best banks list by Forbes for 5 years in a row.

On Wednesday, 8^th March 2023, SVB announced a share sale of $2.25bn ($1.25bn to public, $500m of private placement and $500m of Mandatory convertible preferred stock) to shore up its capital.

SVB said that due to the restructuring of the Balance Sheet it sold its AFS securities portfolio of $21bn to raise cash and to reposition its asset sensitivity to interest rates.  This resulted in a realised a loss of $1.8bn due to the sale of its securities and it was seeking to bolster its capital position and improve its asset-sensitivity to rates.

The timing of the stock offer coming on the heels of the collapse of another Californian Bank (Silvergate Bank - that serviced cryptocurrency users), meant that the news of SVB’s stock offering was received with apprehension and, its stock price plummeted at open on Thursday, dropping from $268 to $176.  By market close on Thursday, the stock fell further to close at $106, recording a whopping 60% drop within the day! 

On Friday, 10^th March 2023, trading in SVB stock was halted as it lost more than 60% of its value. Before the end of the day, the California Department of Financial Protection and Innovation closed SVB and named the FDIC as the receiver. The FDIC in turn has created the Deposit Insurance National Bank of Santa Clara, which now holds the insured deposits from SVB.   FDIC deposit insurance covers up to $250,000 per account but the average balance of SVB customers were ~$4million.  So, based on high level estimates, FDIC insurance would cover less than 10% of the insured customer deposits. It remains to be seen how the issue of unsecured depositor payout is resolved.

Within a couple of days, SVB’s market capitalisation of ¬$16bn was totally wiped out, as if it was struck by a lightning!  And a total of $100bn was wiped out from market capitalisation of the banking sector due to fear of contagion.

What were the key issues?

Its normal for Banks to engage in maturity transformation to make a spread by borrowing short term and lending long term, with established limits for such asset-liability mismatches. 

Asset Liability Management (ALM) entails managing mainly these types of risks amongst others:

·       Liquidity risk

·       Funding risks arising from refinancing risk, funding concentration etc.

·       Interest rate risk in the Banking Book (Trading Book managed with VaR limits) while IRRBB is managed with Earnings-at-risk (EAR) measured as a % of Net interest income over 12 months and change in Economic Value of Equity (EVE) measured as NPV of future cash flows with limit for delta.

·       Capital ratios and other related risks e.g., leverage

 

Robust ALM policies ensure mitigation of the above risks, and these are typically mitigated by a multitude of ALM strategies with an appropriate risk appetite thresholds and early warning indicators for each of these risks.

While Basel 3 has an outlier limit of 15% for change in EVE, they have not prescribed any limit for EAR for the outlier test and left this at national regulator’s discretion.

 

SVB, based on publicly available information, did not employ sufficient mitigating strategies for:

·       Interest rate risk in the banking book (IRRBB), as it invested in a long-dated fixed rate securities (USTs, MBS and CMOs with greater than 10 years maturity at a yield of ~1.9%), probably under the assumption that the risk is well covered by non-interest-bearing deposits, without validating those assumptions for customer behaviour in higher interest rate environment i.e., failure to validate and stress the assumptions for customer behaviour under different interest rate environment.  As evident from the customer behaviour, over $45bn of non-interest deposits declined with 2/3rds of it moving to interest-bearing demand deposits. Such a move would cause a massive shift to SVBs cost of funding leading to earnings drag on their low yielding fixed income portfolio (estimated drag of $2bn p.a.).

·       Deposit concentration to a single industry segment, though it did carry a large stock of Liquid Assets Buffer, it failed to diversify its funding base and with most of its funding in demand deposits, it was disaster waiting to happen

While it’s possible to hedge interest rate risk with derivatives and mitigate liquidity risk with sufficient levels of liquid asset buffers, it is impossible to hedge sector concentration of funding sources while sector itself is undergoing stress, which in SVB’s case was funding source (tech and VC segment).

 

What went wrong with SVB?

SVB’s Key metrics were robust and on surface the key metrics were strong!

 

SVB was rated BBB by S&P and most leading investment banks and equity brokers had a buy rating on the stock with a target price ranging from $325 to $400 (Wednesday close was $268), implying a potential gain in value of 20% - 50%!

 

But, the devil is in the detail!

 

SVB’s investment in Fixed rate securities portfolio grew significantly in 2021, by a whopping ~$82bn, mostly in long-dated fixed rate securities (maturity > 10years).  The overall yield on these securities were ~1.9% when its cost of funds was less than 10bps.  What seemed to be a great idea in 2021 when Fed Funds Rate were 25bps, quickly unraveled as FED raised rates through beginning of 2022, from its lows to end the year at 4.5%.  The inflation expectations and interest rate forecasts were completely undone by the Russia-Ukraine war causing supply disruptions across several economies, fueling a runaway inflation.  This led to many Central Banks raising its policy interest rates sharply and continually to tame inflation, though the inflation still remains stubbornly high.  The mantra of higher for longer came into being, i.e., interest rates will remain higher for longer.

 

Now none of this new!  The Savings and Loans Crisis (S&Ls) in the 1980s was triggered by a similar reason, when Paul Volcker hiked policy rates aggressively to tame inflation.  S&Ls had issued long-term loans at fixed interest rates that were lower than the newly mandated interest rate at which they could borrow. When interest rates at which they could borrow increased, the S&Ls could only attract more deposits by offering higher interest rates which led to liabilities that were higher than the rates at which they had loaned money. The end result was that about one third of S&Ls became insolvent.  History repeats itself!

 
How did it impact SVB?

Fixed rate bond prices fall in value when interest rates go up as the market yield adjusts to the new normal. The Interest rate risk manifested itself by way negative market value on SVB’s Available for Sale (AFS) securities portfolio of $26bn, and this fed through to its equity via Other Comprehensive income leading to erosion of its equity capital and capital ratios.  Despite this, SVB had strong capital ratio of ~16% at the end of December 2022. 

 

The Held-to-Maturity (HTM), the larger portion of their investment portfolio at $91bn, does not require a mark-to-market as per accounting standards.  To that extent, the drop in value in this portfolio (estimated at $17bn) remains buried, and it flows through gradually into P&L in the form of negative interest earnings over the life these securities i.e., a slow bleed of ¬$2bn each year, if interest rates were to remain elevated through this period, which is typically unlikely, and hence the negative earnings drag would moderate over time.  Unlike AFS securities, HTM securities cannot be sold as sale of even a single security taint the portfolio resulting in redesignation of the portfolio to AFS.  This will lead to the recognition mark to market through OCI which will deplete its capital base, thereby wiping the entire equity (as the estimated MTM on HTM is a negative $17bn.

 

If there is a potential buyer of SVB, they would write down the HTM portfolio, which would result in a negative equity position!  And if there are further asset write-downs, it’s inconceivable that depositors would be made whole.  A recovery rate of ~80c is the most likely outcome for the unsecured depositors, absent any form of a bailout.

 

So, where did it go wrong for SVB?

There were four main reasons for the sudden collapse of the Bank:

1.      Interest Rate Risk in Banking Book (IRRBB) not managed robustly - Significant exposure to Fixed rate risk and soft assumptions for non-maturing deposits. With a portfolio yield of 1.9%, as Fed Funds rate increased from a low of 25bps in January 2022 to a high of 4.25% in December 2022, bonds prices slumped leading to build up of losses in the bond portfolio.  Meanwhile the cost of funding increased as customers shifted their deposits to higher interest accounts from non-interest-bearing accounts. For AFS securities the MTM losses impacted OCI leading capital shortfalls.  While negative earnings drag on the HTM securities would bleed through P&L over time.

2.      Depositor concentration – While SVB had some diversification with 38k customers with balance over $250k accounting for almost $150bn in deposits, it had concentration from source of these funds to the tech firms and VC funding client segment.

3.      External factors resulting in cash drain for tech firms, as VC capital funding activity slowed while techs firms cash burn increased leading to withdrawal of deposits causing liquidity squeeze.

4.      Poor management decision to sell AFS portfolio and crystallise the loss through P&L of $1.8bn and simultaneously launching a stock offering of $2.25bn, when none was required, on a day when another Bank had collapsed (Silvergate Bank that serves crypto currency), thereby causing a market panic followed by loss of customer confidence and a digital Bank run!

 

Detailed analysis

While all banks invest their surplus liquidity in Treasuries and bonds, SVB was an outlier with its investment strategy with its portfolio increasing to 57% of its total assets against an average of 24% for US Banks.

 

SVB was a bit slow to react to the shifting sands in the business environment of its client segment.  The VC investment activity slowed in 2022 and the cash burn of tech firms increased since Q3,2021.  It missed the alarm bells, and these were sounding louder since Q2/2022, by way of increasingly negative client funds position as shown below:

 

 This general weakness in SVB's client segment, together with the rising interest rates right through 2022 resulted in a significant shift from non-interest-bearing demand deposits to interest bearing deposits, as shown below:

  

The decline in deposit balances and a shift to interest-bearing demand deposits meant that SVB was beginning feel the pain of liquidity and earnings squeeze as it would have to offer higher interest rates to attract deposits which would cause a spike in its cost of funding.  This alone would cause an earnings drag of ~$2bn p.a., wiping out its profitability entirely.

 
What could SVB have done differently?

In some ways, SVB brought this collapse upon themselves!

1.       It was ill-advised to sell the AFS Securities portfolio ($21bn) and crystallise the loss through P&L

Their March 8^th presentation shows that they had ample liquidity and the ability to meet their cash needs through available REPO lines:

 

By selling the AFS portfolio of $21bn, SVB crystallised a lost of $1.8bn. This MTM loss was already feeding through to their equity via the OCI line. And liquidity could have been raised via REPO, without having to report a huge headline loss of $1.8bn. Yes, there would be some collateral haricut but that’s not a huge constraint for SVB.  They could have closed the interest rate risk on the portfolio by hedging with fixed pay Interest rate swaps. 

 

The AFS security sale process to make it asset sensitive is a red herring! Their intent was to raise cost free equity to shore up their net interest income and this whole asset side restructuring seems like smokescreen for that. They were looking improve net interest income by raising equity to camouflage the negative bleed on the HTM portfolio.

 

2.       The need for capital is questionable and the timing of capital raise was awful

SVB had strong Capital Adequacy Ratio at 16% and hence there was no need for a capital raise, at least at this moment.  And the timing of the announcement was awful, coming close on the heels of another bank failure in the Valley (Silvergate).

 

Unfortunately, the capital raise never got done. The bank chose to announce its balance sheet restructuring the same day that Silvergate Capital announced it is going into voluntary liquidation. As a result, customer fear turned Silicon Valley Bank’s manageable deposit outflows into a flood. 

 

Founders Fund, the venture capital fund cofounded by tech billionaire Peter Thiel, on Thursday advised companies to withdraw holdings from SVB, due to concerns about financial stability, Bloomberg reported. Stability concerns reverberated across the banking sector, hitting banks base in the Western U.S. particularly hard.

 

We’ve never really had a bank run in the digital age. Here’s how it looks:

People posted screenshots of Silicon Valley Bank’s website struggling to keep up with user demand.

Is there a risk of a contagion?

There has been a panic reaction in the market with more than $100bn wiped off the market capitalisation of banks in two days!

 

Though, my view is that broader risk of contagion across the banking sector is unlikely.  Some of the smaller banks and fintechs may witness a flight of their higher value deposits which could potentially lead to failure of some of the smaller banks but banks with diversified business model and large customer base will remain unaffected as there is ample liquidity in the banking system.

 

What are the lessons?

Each crisis brings with it an opportunity to introspect and learn.  So, in summary these are the learnings from the SVB saga:

1.       IRRBB - Without a doubt, Regulatory oversight on IRRBB is likely to be the main theme for 2023. Global regulators and central banks will have an increased focus on IRRBB and banks will do well to revisit their IRRBB model assumptions and validate them, and set risk appetite limits for both EAR and EVE and a regular assessment with interest rate simulations.

 

2.        Don't put all your eggs in one basket – This one applies to both assets and funding sources.  Investments should be laddered to allow for 15% - 20% to mature each year to manage both liquidity and rate risk.  There should also be maturity tenor limits, more so when the funding source is largely short-term.  EAR and EVE limits should be set in line with the ability to absorb shocks.  Diversification of funding sources is key and its should be not just by number of customers but also across industry segments, and maturity profile.

3.    Never go for a capital raise if you are solving for liquidity – Liquidity risks cannot be met with capital.  The solution for liquidity risk is to hold liquid assets and reduce refinancing risk by having a proper asset-liability gap limits.

 

CONCLUSION

 

The pressure to enhance shareholder value drives banks to take risks that appear low at inception but is not always fully tested for severe stress scenarios.  Regulatory oversight together with stronger governance where the board members have a very good understanding of risks provides a good level of safeguard in most banks. But quite often, I see a great focus on P&L performance and there is very little probing on the Balance Sheet and Notes to the accounts, where many skeletons can be buried! (e.g., MTM on HTM bonds).

 

Banks are in the business of taking risks, but they have a primary fiduciary duty to keep their depositors’ money safe.  

VIEWS EXPRESSED IN THIS ARTICLE ARE PERSONAL.