Silicon Valley Bank: What Really Happened?
I have been buying Short-Term Treasury Bonds (T-Bonds). As I mentioned in my previous post, they are an attractive investment yielding 5% with minimal risk. I am putting emphasis on Short-Term as Short Term bonds are easier to hold until maturity, thereby alleviating my concern about the effect of interest rate-driven changes in the price of bonds (duration risk). If the interest rate goes up, let’s say to 10% (I hope not), my bonds mature in less than 12 months (and sometimes less than 6 months) and I can reinvest my capital at 10% now instead of 5%. The Fed has increased the rate from 0.25 to 4.75% in less than 12 months. With stubborn inflation, there is no end to sight when the Fed will stop tightening. As the interest rates can go beyond 5%, I will keep buying tranches of short-duration T-Bonds.
Fig1: Fed increasing interest rate from 0.25%in May’22 to 4.75% in February’23
What if I buy T-Bonds with a longer duration, maybe maturing in 10 years and the interest rates go up? Well, I have to wait until the interest rates go down or bear the loss. Interest rates have the biggest influence on bond prices (except defaults, of course). It is important to understand this.
How do change in Interest Rates change the Bond Price?
Imagine I lend $100 to buy a bond that is yielding 1% (Annualized Yield) for the next 10 years (Duration). After 10 years, once the bond matures, I would be paid $110 (1.01^10 i.e., $1 in yearly interest payments and $100 as the repayment of my initial capital compounded annually). Now, if the interest rates go up, let’s say to 5%, lenders will be expecting to get paid at least 5% for their capital. The effective return on the same $100 bond will be $162.8 (1.05^10). To match the 5% yield, the bond price will have to fall by 32%. The bond will be priced for $68 instead of $100 to yield the same 5% [$68*(1.05^10) = $110]. The bondholders of 1% yield bonds will have to take a 32% haircut on their bond portfolio (marked to market).
Now that we understand the impact of rising interest rates on bonds, let’s understand what really happened at SVB.
SVB Balance Sheet Summary:
Liabilities: SVB owes $196B to its lenders (Liabilities). $13B of it is owed in the next 12 months, $9B is owed after 12 months. However, they have $173B is cash deposits from customers/ corporations that can be withdrawn anytime.
Assets: SVB has $14B in cash and $212B in Assets. They have enough cash ($14B) to service their short-term debt ($13B) without selling anything. They have an additional $26B debt that will mature within the next 12 months (CDs, T-Bonds) and can be reinvested or can be sold without much loss. They can easily service $40B withdrawals(25% of the deposits). What happens if more than 25% of the deposits are withdrawn?
Once more than $40B withdrawals are requested, SVB will have no option but to sell $91B Held to Maturities Securities. What are these you ask? They are Mortgage Bonds Swaps (MBS) and US 10 Year Treasuries Bonds that SVB bought. All of these were long-term bonds maturing in more than 10 years. Given these were bought at low yields, their current market price declined. As per the Generally accepted accounting principles (GAAP) account rules, assets are not priced as per the market price. In a rising interest environment, fixed income assets are shown at inflated prices. A 1% move up on a 10 yr duration bond is ~10% - on the bond price. This led to SMB selling their securities at a lower price than they were on their balance sheet. Since SVB was 11:1 leveraged (owning $11 for $1 of equity), a 9% loss on these results in a total wipeout of the company’s assets.
On Thursday, March 9th, 1 day before its collapse, SVB declared losses and tried to raise $2.25B by selling their stock. This created a trickle down effect that led to accelerated withdrawals requests, leading to its ultimate collapse.
There are other reasons that aggravated its collapse as described below:
Declines in average client fund balances - Since SVB mostly caters to venture-backed early stage startups, the VC money dried up. With no additional source of funding, and increased cash burn, the deposits decreased by $16B from January to March.
Existing Deposits are getting more expensive to hang on to - Customers are demanding higher interest on their deposits, leading to higher interest expenses on deposits.
Investments in Venture Back Debt - 10% of the consumer deposits were used to offer loans to startups. Since the startups are finding it impossible to have an up round (Stripe valuation dropped from $95B to $50B). These startups have no collateral and SVB incurred losses/will incur losses on these loans.
Poor Risk Management - In a rising interest environment, buying long-term bonds is a colossal mistake. Not having a Chief Risk Officer and not hedging long-term bets by diversifying ultimately led to the wipeout.
I will conclude with a quote from Warren Buffett - “Only when the tide goes out do you discover who's been swimming naked.” Everyone is a good investor in a bull market. While I am happy that the Fed has decided to backstop all the depositors in SVB, I don’t expect the buck to stop here.
To learn more about some of the underlying models and calculations I used, feel free to reach out to me at karanbathla07@gmail.com.