A functional person wants to get more

Silicon Valley Bank

---

The banking business seems easy to me. Use depositor money acquired at a low rate to lend to others at a higher interest rate. Make sure that the rate spread is high enough for you to employ enough people for customer service, but also offer high-fee depository services to pad the bottom line. That’s classic banking.

In reality, the game is much more complicated than that as we’ve all learned in the past 72 hours with the crack-up of Silicon Valley Bank and First Republic Bank in California and Signature Bank in New York. Each of these banks had high exposure to a homogenous client base of technology companies and associated entrepreneurs. So what happened? Why didn’t classic banking rules apply?

In a word, COVID. This virus (which didn’t come from a lab in China, but maybe it did, but probably didn’t, but definitely did) caused a shutdown for the world economy in 2020. In response to the shutdown, governments and central banks pumped the world full of money to overcome any short-term cashflow needs. Much of those funds were invested in VC and PE funds, which the duly invested it on behalf of their Limited Partners into tech companies with the promise of high growth and high investment returns.

Over 50% of venture backed companies (65k businesses), had a depository relationship with Silicon Valley Bank, an institution literally founded to provide services to a historically underbanked section of the economy. When those companies received the VC money, they duly deposited it into their SVB accounts. SVB deposits soared as a result as became America’s 20th largest bank with almost $200b in cash. Then came the problem: what do you do when you have too much cash?

Banks exist to make money; they do this by loaning money at high interest rates, except interest rates until this year have been at historic lows and were going to stay there. Every quarter, the Fed reports a forecast on the long term trend prediction of interest rates. Want to know where rates will be in 5 years, look to the Fed, they’ll tell you.

In 2020, the Fed was reporting that rates would increase by .1% by 2023. Silicon Valley Bank relied on that number and invested B-B-Billions in Treasuries at then current rates. It turns out the forecasters at the Fed were slightly off the mark with their trend analysis. Instead of .1%, rates actually increased by 4.75%, a 47x rounding error.

When interest rates rise, bond prices fall. Why buy a bond paying 1.2% when you can buy the latest vintage paying 4.75%? And here is Silicon Valley Bank making exactly that bet billions of times. When it started trying to true up the balance sheet and free up the invested capital for liquidity needs, it lost almost $2b. So it stopped doing that and instead went in search of new investors to help shore things up until the yield curves righted themselves.

No investors were to be found for such a shitty deal. Things were bad … but not as bad as they were going to be. One of the most influential depositors in SVB was Peter Thiel, a PayPal Founder and the genius behind the Founder’s Fund VC Fund. Peter heard that SVB was having trouble funding deposit withdrawals and he sent a blanket order to get ALL of his money out of the bank - personal, fund’s, companies, everything was wired out. Then he sent an email to all of his portfolio companies and said “Get out while you can.”

Dozens, then hundreds of business began withdrawing funds, and no one was depositing. It was a classic bank run, where the bank simply didn’t have the liquid cash available to fund everyone’s requests. In order to raise that money, it would have to sell securities at a loss which pushed it to insolvency. The FDIC walked in, shuttered the bank and seized it assets. The FDIC then tried to sell the bank over the weekend and it received no offers from any bank (although the criteria set by the feds were stupid and ego-driven). This forced it, the Treasury and the Fed to invent on the spot a new credit facility — backed by taxpayers — to pay off all the depositors and make them whole again.

Case closed. Crisis averted. Champagne corks fly.

"Banking is very good business if you don't do anything dumb."

— Warren Buffett

Was their wrong done at SVB? Was this the result of bad actions by greedy bankers? Initially it didn’t look like it. The bank was just overexposed a specific sector of the economy and made some bad decisions. Them’s the breaks. No harm, no foul.

But.

Maybe the leadership at the Bank should have known something was up. Maybe it did know. After all, in its most recent earnings report in January, the bank revealed its held-to-maturity securities had mark-to-market losses of nearly $16 billion in Q3, against just $11.5 billion of tangible common equity. Essentially, the bank would be underwater if it was forced to liquidate all its assets.

Surely the Bank’s Chief Risk Officer was charged with evaluating the reality and strategy to recover from such a strange situation … except the Chief Risk Officer position was vacant … for nine months. It did, though, have a Chief Administrative Officer who had been in place for 16 years — right after he left his job as the CFO at Lehman Brothers which expired in exactly the same way and precipitated the 2008 financial crisis.

Man, seems like the CEO should know the bank was in trouble and take some action. Well, he did take action. He sold almost $4m of SVB shares for his own account and evidently encouraged his colleagues to do the same, selling almost $10m in stock in February alone. Even knowing the situation, he still ordered that employee bonuses be paid out.

As with all things seen through the rear view mirror, the entire scenario would have been obvious to someone who had the entire picture. No one did, not even the managers. The crisis they created with their betting strategy is one that many, many other banks have adopted, as well. Let that sink in.

"Don't do anything stupd. And don't waste money. Let everybody else waste money and do stupid things; then we'll buy them."

— Jamie Dimon

Numbers obscure nuance.