Monetary Policy & Inflation | US
- As of the end of end-2022, Silicon Valley Bank reported a 15% Tier 1 capital ratio but fell under receivership on Friday due to a bank run and fire sale of its bond portfolio.
- SVB is unique for its concentration in tech, large bond portfolio, and large deposit losses.
- Yet, to a much lesser extent than SVB, all banks face run risks, partly due to post-GFC regulatory tightening.
- The government plans to make all depositors whole and provide long-term bank funding with collateral valued at par; those are likely to stabilize markets.
Like these insights? Join our free newsletter for regular updates and trade ideas.
SVB Felled by a Bank Run
On Friday, 10 March, the FDIC placed Silicon Valley Bank (SVB) under receivership after a run on its deposits.
Depositors’ concerns over losses triggered the run. SVB lending was concentrated in the tech sector, and capital flows to the sector have dried up since the Federal Reserve (Fed) started raising rates. As a result, SVB lost deposits, and the quality of its loan book came under scrutiny (Chart 1). In addition, a large share of SVB deposits was above the FDIC insurance limit of $250,000 and, therefore, was more likely to leave the bank if doubts over its solvency arose.
As of 31 December 2022, SVB reported a Tier 1 capital ratio of 15.4% (i.e., the bank was solvent). This number, however, did not account for unrealized losses on SVB bond holdings. Because three-quarters of SVB’s bond portfolio was held under the Held To Maturity category, they did not have to be marked to market. Losses would have entirely wiped out Tier 1 capital if they had been marked to market.
To meet depositor withdrawals, SVB had to liquidate its bonds and realize the losses. On Wednesday, 8 March, SVB announced it needed to raise $2.25bn to make up for the losses. This effectively triggered the run and put the bank in receivership.
Silicon Valley Bank was the 16th largest US bank with assets of $209bn on 31 December, about 1% of total domestically chartered banks assets. It was the second largest US bank to default after Washington Mutual in September 2008: WaMu’s assets were $307bn or 3% of total domestically chartered banks assets.
Silicon Valley Bank’s Problems Are Not Systemic
SVB’s problems are not representative of the broader US banking sector. First, SVB’s exposure to tech was unique. SVB played a unique role in Silicon Valley: it built its business around tech startups, companies, founders and VCs. Its disappearance will have profound consequences on tech startups.
Second, compared to other large banks (the Fed provides balance sheet data for the 25 largest domestically chartered banks), SVB stood out with its fast asset growth and higher bond exposure (Chart 2). As a bank catering to the tech sector, SVB and its clients strongly benefitted from the extraordinary Fed stimulus during the pandemic.
Third, Silicon Valley Bank suffered larger deposit losses than other large banks (Chart 3). At the same time, it had a lower loan-to-deposit ratio than other banks. However, this did not protect SVB from a run because its non-loan assets were bonds, presumably unhedged against interest rate risks.
All US Banks Face Run Risks
For all the uniqueness of SVB’s problems, all US banks are exposed to the same perils, if to a much lesser extent. Essentially, the post-GFC regulatory tightening has required banks to hold more bonds (Chart 4).
At the same time, the regulators did not foresee the ongoing inflation surge and associated increase in yields. For instance, even though they entail very different interest rate risks, Treasury Bills and bonds have the same weighting for the computation of Liquidity Coverage Ratio, meant to pre-empt precisely SVB fire sale-induced bankruptcy. In addition, the Fed’s latest stress test scenario for banks forecast, in the base case scenario, that the 10yr yield will peak at 3.9% in Q1 2023—and fall to below 1% by Q2 2023 in the adverse scenario.
And fundamentally, bank runs are part and parcel of the banking business model that consists of funding illiquid assets—loans—with short-term liabilities, deposits (the economists Douglas Diamond and Philip Dybvig won the Nobel prize for formalizing this). Unless banks move to a more utility-based business model (such as narrow banks), they will remain prone to runs. In addition, in today’s era of widespread social media and instant payments, runs can happen much faster than previously.
For now, however, the joint Fed-FDIC-Treasury plan announced on Sunday greatly reduces the risk of a repeat of an SVB-type crisis.
Treasury/Fed/FDIC Intervention to Stabilize Markets
The Treasury Department, the FDIC and the Federal Reserve have a plan to stabilize any possible systematic volatility when the market opens on Monday. We think it works quite well in dealing with the immediate problem and maybe even in the medium term.
The plan consists of the following:
- FDIC to complete the resolution of Silicon Valley Bank and Signature Bank, which was closed on Sunday. All insured and non-insured depositors will be fully protected at both banks.
- The Fed will provide liquidity to the depository sector by creating a new credit facility: Bank Term Funding Program (BTFP). This will provide:
- Up to one-year maturity loans to depositories against Treasuries, agency debt, MBS, and other high-quality securities.
- Most importantly, there will be no haircut; the collateral will be valued at par.
- The BTFP facility will be backstopped with $25 million in funding from the Treasury Department’s Exchange Stabilization Fund, a Treasury facility often used in financial rescue operations.
The Fed also tried to deal with moral hazard by stating that ‘No losses associated with the resolutions will be borne by the taxpayer. Shareholders and certain unsecured debtholders will not be protected. Senior management has also been removed. Any losses to the Deposit Insurance Fund to support uninsured depositors will be recovered by a special assessment on banks, as required by law.’ The last sentence implies that banks’ deposit insurance assessment could rise if asset sales do not cover the cost of making all SVB and SB depositors whole.
In addition, the Fed expressed that it does not expect any drawdown from the ESF facility. We think that means any losses would be recovered from a special assessment of banks. And although this was not mentioned today, in the long run, the regulators could ask banks to raise capital to cover unrealized losses on their bond portfolios.
The new facility is to last one year but could be extended. In the long run, will the Fed make all depositors whole in future cases of bank failures? Today’s rescue steps suggest so. At the same time, an increasingly explicit guarantee of all deposits, irrespective of size, could see banks become more like utilities and credit and market risks get moved to non-bank financial institutions.