The 40-year-old Silicon Valley Bank (SVB) was once the fuel behind the start-up flight. More than 2,500 VC firms banked with them, and half of all venture-backed start-ups were its customers. It went belly-up in 48 hours flat, making it one of the fastest crashes in history.  

No one saw the crash coming. Mark it; in 2022, Forbes named Silicon Valley among America’s Best Banks, and Moody’s gave an ‘A’ rating! The rating meant that Moody’s judged it to be of upper-medium grade and subject to low credit risk. 

Initial growth phase

Here’s the background to the fall.

Deposits (basically liabilities) with SVB tripled between 2018 and 2021. With start-ups running out of gas in the wake of the pandemic, there was not much to invest, and a lot of cash lay around. The bank placed this money in long-dated government bonds and mortgage-backed securities. 

SVB had a balance sheet size of $211 billion. Its investment book was $120 billion, of which $91 billion was in mortgage-backed securities (MBS) and $29 billion in t-bonds. An MBS is a collateralised pool of mortgages and transfers credit risk from a primary lender to an investment bank. It is illiquid compared to treasuries and, in the case of SVB, had a duration of 10 years. 

The interest rates on these investments were in line with the market. So far, so good.

Suddenly, The Federal Reserve increased interest rates, and the value of the bonds fell in a heap. Remember, interest rates and bond value are inversely proportional. I will tell you how. 

Suppose you have invested in bonds worth ₹1,000 with an 8 per cent coupon. Now, assume that the market yield or expectations go up to 10 per cent, which happened with Feds increasing the rate. Your existing bonds fetch you ₹80 (8 per cent on ₹1,000). Because you are making only 8 per cent, while you can earn 10 per cent elsewhere, you will want to liquidate your bond. Everyone holding the bond will also do so, and selling pressure will kick in. Hence, the bond price will fall and fall to ₹800, at which point the ₹80 gives a yield of 10 per cent. This would mean a loss of ₹200, which will be temporary and notional if the bonds are held till maturity. 

Now, in accounting, bonds can be classified as Available for Sale (AFS) or Held to Maturity (HTM). If they are held as AFS, they would be valued at market price, which would mean a periodic marking down or up of the value, as the case may be. If classified as HTM, you will book losses only when you sell, and many companies find it convenient to do that. SVB had the bulk of its assets as HTM. 

With investors wanting their deposits back because yields were better elsewhere, these bonds had to be sold to find money cash to repay the borrower. It sold $20 billion of securities to Goldman Sachs in an OTC sale at a loss of $2 billion, and the contagion effect began. More sales followed.

SVB took a hit as its HTM book had grown from $14 billion to $99 billion. Also, the age of the borrowings was shorter than that of investments, leading to an asset-liability mismatch. In accounting, it’s called duration risk. For instance, if you borrow for five years, you shouldn’t be parking the money for eight years because when the liability comes up for settlement, you may need liquid assets to settle. 

The avalanche begins

When SVB started selling its assets, it had to book for its losses. To make matters worse, Moody’s talked about a possible downgrade of the bank’s credit rating, causing investors to panic. Also, as bonds began to lead to capital losses, the bank decided to raise money to shore up. The investors saw that as a warning sign. 

What followed was a stampede to withdraw deposits. It leads to the second-largest bank failure in US history. 

So what was it in summary? 

SVB relied too heavily on corporate and venture capital funding, which are fair-weather friends, instead of retail deposits, which are stickier. It had 99 per cent of its deposits from the corporate sector, of which only 7 per cent were under 250K and were insured. It also had a very high level of loans plus securities as a percentage of deposits. In other words, SVB fashioned out a riskier profile than other banks, setting itself up for a smash in case of rising interest rates, deposit runs, and mandatory asset sales.

What happens now?

In a later day development, the US regulators have chalked out a plan under which all depositors of the failed bank would receive their money.

The Fed will create a new Bank Term Funding Program (BTFP) for institutions impacted by the fall of SVB. The BTFP will provide loans of up to one year in length against US Treasuries and MBS. These assets will be valued at par regardless of the changes in interest rates.

The government will also make available up to $25 billion from the Exchange Stabilization Fund towards this. None of this means that SVB will be bailed out. Shareholders and unsecured depositors will get zilch. 

What next?

Well, they could have ensured an asset-liability match, failing which they could have hedged their risk by taking derivative exposures like interest rate options or interest rate swaps. Apparently, they did not and bet on the Fed keeping the interest rate low. 

This is the first significant case of a fall due to needing to stick to basic principles of corporate finance and risk management. The irony is that most banks have historically failed due to credit risk issues. They were reckless. They did not have a chief risk officer for eight months and paid bonuses days before the fall. That’s a corporate governance failure, as well.

The writers are chartered accountants.