… none of which is to say that this made SVB particularly irresponsible or deserving of what happened this week. It really does look like a confluence of bad timing w/r/t increasing interest rates and some unfortunately timed bond buys. IANAB, but it seems like all the money is there, just tied up in bonds that won’t mature for ten years, leaving them illiquid.

If I understand this correctly, SVB will be bought by a larger fish, and we’re back to business as usual, albeit more consolidated (enormously bad, imo, but a different bad).

Per Matt Levine, their fatal and unforgivable mistake was putting their assets in fixed, low-interest, HTM investments whose value was directly correlated to the financial health of SVB’s depositors, so the bank had to write down assets when depositors needed their money.

In hindsight, it would have been better to keep cash. Or even better to find an asset inversely correlated with interest rates, even if it would have earned less money when rates were low.

It was a lack of beta that killed them, and while there doesn’t seem to have been wrongdoing, it sure looks like a classic “failed to understand their own business” mistake.

They were also a victim of massive success: their deposits skyrocketed and they did not have anywhere to put the money. This is not unlike the same problem wealthy individuals had for the past 10 years if they felt like the stock market was over-valued; nothing else was paying any return. Now imagine your depositors start increasing their withdrawl rate and you're locked in 10 year bonds at < 2%. Uh oh...

It was the greed of seeking high returns even though their value would drop for the same conditions that would cause depositors to need their money.

The bank would still exist, and just have been less profitable in previous years, if they had hedged against interest rate increases.

Thanks, that's the explanation I was looking for.

> inversely correlated with interest rates, even if it would have earned less money when rates were low.

I'm not too savvy but isn't that what I-series bonds are for?

those have very low limits - 10k per tax identifier

I see. So useful for companies who are considered too high risk for "regular big banks" --- but they're all clumping together so it should be your second choice, only if you couldn't get a loan from a regular big bank.

No. SVB was not a bank for high risk people. They were the 16th largest bank in the United States. They were a regular commercial bank. First Republic, Northern Trust, etc are all very large banks that cater to specific clientele.

It’s not that they catered to high risk clients it’s that they understood that the definition of risk was not universal. What may look risky to Bank of America isn’t actually more risky if you know the nuances of startups.

Or put another way, the odds that a new commercial Bank of America company goes on to become a 10 billion dollar company is probably 1 in a billion. The odds that an SVB customer would go on to become one is probably 1 in a thousand.

But the important thing is that SVB didn’t fail because of bad credit or bad loans - something you’d expect from riskier clients. SVB’s problem was that they had more deposits and cash than they could handle, they literally had too much money and the fed has raised interest rates too fast for them to absorb.

> more deposits and cash than they could handle

Do you know why couldn't they handle it? Banks usually have to manage their Credit, Interest Rate, Liquidity, Price, Foreign Exchange, Transaction, Compliance, Strategic and Reputation. So I wonder whether SVB just wasn't sophisticated enough to handle these risks.

If they're the 16th largest bank in the US, it seems to me that they should have had the chops to "handle" large amounts of cash...

It is unclear if any other bank could have handled it either.

It took two factors.

3x the money you bank in 2021 specifically

Have well networked clients that carry out a bank run together.

Three things I can see:

1. way more deposits than they historically managed, with no-where good to put it.

2. failure to do what banks do, which is arbitrage their position, putting a huge amount into long-term low interest bonds

3. looking for loopholes in liquidity & stress tests, and reporting exemptions that might have raised earlier alarms

#3 is inexecusable but the other points are in the "bad at your job" category.

Their problem was that they tied up too much of their money in illiquid low-yield bonds. Pure mismanagement.

this is the ultimate cause, but they also used loopholes to avoid reporting obligations other banks complete, and did not highlight the liquidity issues or stress-test that say... increased withdrawals might highlight. These aspects should be investigated deeper.

agree. it's always such systemic / procedural lapses that allow problems to go unnoticed

It's not even about getting a loan.

It is about having a company checking account for payroll and somewhere your investors can put money.

I don’t think it’s so much “high risk” as it is “non-traditional.” In a normal new business, you have a factory, and all the machinery is assets, and if it goes under you can sell the physical trucks and widget presses.

That’s just not how America has worked for a long time now, and I don’t think most folks would describe American businesses as any riskier than in say Serbia.

This isn’t a “bad guy” story, as far as I can tell.

> This isn’t a “bad guy” story, as far as I can tell.

I agree. There are no bad guys here. There's just a bank that took and chance and it bit them.

If there are bad guys it’s the VCs who coordinated the bank run. Even still, each of them was acting rationally and not maliciously, but they’re all savvy enough to know that racing for the exits is going to kill the bank.

If SVB's activity can be described as "yield chasing", there is no way clients, whatever their actions were, deserve more blame than the management.

Eh, if “clients” and “management” were totally distinct, I’d agree. My understanding, and I’d welcome correction from those more knowledgeable, is that the SVB was hardly independent from the VC powerhouses.

> I don’t think it’s so much “high risk” as it is “non-traditional.”

Funny how tables have turned recently. Now it's other way around more often. If your start up is actually manufacturing something, a physical item, the banks deem the business as non traditional... Go figure...

> because tech startups usually lose money at the beginning of their lives

Oh what a 2010 thing to say!

2023 startups lose money for all their lives. SVB was running a bad business banking almost exclusively to startups that have never had positive cash flow.

Right, and this is why they fundamentally failed. If their main depositors were traditional businesses and individual accounts they wouldn’t be an issue today. They invested their money as if they had low-risk depositors but a start-up is a high risk depositor——they don’t have a proven business model so it is hard to forecast their behavior.

To me this is classic SV echo chamber thinking. There’s a reason traditional banks are more wary of lending to start-ups and it isn’t because they hate money.