Matt Levine: Silicon Valley Bank Ran Out of Money | Hacker News Reader

Matt Levine: Silicon Valley Bank Ran Out of Money

bloomberg.com

37 pointsVagueMag21 comments

I think an undercovered long term impact of Silicon Valley Bank collapsing is that VCs have managed to kill off one of their own unique perks. Once you start digging in to what SVB was doing it wasn't just holding money for VCs and start ups, it was doing all sorts of stuff that was clearly questionable banking practice- but bought it that unique clientelle. Whether it's cheap mortgages to founders, weird venture debt, or turning a blind eye on regulatory compliance. One thing no clearly some VCs didn't think about was "next time I need a bit of flexibility out of a bank, could I get that flexibility from JPM". The answer is clearly no, and the whole Valley is going to have start playing by the rules of mere mortals again.

I’d always ask here what it was about SVB that made clients flock to them and responses were pretty mum. I could believe it that it was things that other banks wouldn’t do.

Our rationale:

1. They had some GUI functions that other banks didn't -- like a lot of fields to optionally fill in for wires (instructions for intermediate internation banks etc). Yeah we didn't use those, but for a power geek its nice to know it's available when we need it. Frankly, they killed most of those with their new "SVB Go" UI unfortunately.

2. More control on users permissions (e.g. approval limits).

3. Diversification -- one more bank is good.

4. Just the name -- we operate from Silicon Valley. :)

None of these seem like great reasons tbh

Actually we have a pressing need now, obviously :)

What would be your reasons?

Stability -- pretty much all top 20 banks should be pretty stable (until they aren't lol)

So I honestly cannot think of anything differentiating besides how modern is their IT infrastructure.

My reasons for picking who I bank with are based on risk & needs. I personally don't have more than $250k in cash, but I bank with who I do because they are convenient (ATMs everywhere across the US, for one) and they are one of the largest financial institutions on the planet.

Yea, but how do you estimate risk? As I said, all top 20 banks are supposed to be pretty stable, and SVB was one of them. Credit Suisse was also very big. If one is not a banker (and we are not), it's kinda hard to know the risk.

ATM is not really a reason for business banking. But Wires are, especially when you need to pay an international contractor in a foreign currency. And businesses usually have more than $250k.

Well, it's true that I probably would not be able to do that to a high level of accuracy. But that's my reason for picking a bank.

I suppose that picking SVB or CS or whatever was still a good choice, because you wouldn't have lost your money!

My understanding is banking for businesses is fairly hands-on and relationship-based, and for the same reason that normal investors don't deal with startups (so you have VCs), normal bankers don't want to either (so you have SVB). Startups are weird. Their relationship with money is weird. No steady cash flow like a normal small business, no collateral like a real estate thing, $100 million check in the door one day, broke the next.

I'm curious how you got that from this article. To quote, "SVB actually did relatively little lending, for a bank, and had rather a lot of safe bonds." That is, they were run like you would expect a small regional bank that doesn't like milking interest out of their region to be run. Straight forward take in deposits and stick them somewhere with silly modest returns.

Are there other sources going into shady things they were doing?

This article touches on some: https://www.nytimes.com/2023/03/17/technology/svb-tech-start...

I'm not sure that is really underlining too many "sleezy" things happening, though? Honestly reads more of "regional bank caters heavily to region it is in" to me.

And some of the numbers are high looking in this, but it literally states "its $9.5 billion fund," when the article here is talking about how it had 100+ billion in bonds. Which were deemed safe.

Let that sink in. Yes, they had 9 billion they were investing. But they had over 120 billion sunk in "safe" bonds. That is, the 9 billion was /supposed/ to be the risky part of their portfolio. Yes, they probably took some excess risk there. But that is definitively not the risk that tanked them. And any complaints in that direction is a side show that has virtually nothing to do with what happened. Unless you want to say they weren't doing it enough.

>a small regional bank

They were in the top 20 biggest banks in the USA. "regional" is a term of art more than used in the conventional way. I think that most people who hear "regional bank" think of one with only a few branches and a relatively small amount of assets (compared to global behemoths). SVB was very much a world-class player in terms of both general sophistication and AUM.

Right, my understanding from following is they should have been run with the sophistication that their size and exposure would have warranted; but were instead still playing the game as if they were the regional bank they started as.

VC's have tech banks in their portfolio, they're probably not planning on sticking with JPM or whatever other wall st. firm long term.

I have to admit that after a lifetime of dealing with banking hours and having to wait days for ACH transfers, it is a bit of Schadenfreude to see the same happen to a bank itself.

The article this points to was submitted here:

How the last-ditch effort to save Silicon Valley Bank failed - https://news.ycombinator.com/item?id=35259527

The lines of credit that startups had at SVB are the most likely contagion.

Yes. Depends on terms. But I know of a few that can have their entire credit line called for SVB as part of the bank's closure if needed. Kind of wild.

More importantly, having those lines of credit available between accounts receivable/payable terms or as a gap between rounds is key.

For a non San Francisco based startup, my previous cos, line of credit could provide 1 year of opex at the time of a funding round

Reply on news.ycombinator.com