Hacker News Reader

Comment by LewisVerstappen | Hacker News Reader

So the VC thinkbois basically created a run on SVB. Nice job guys!

Anyway, the only rational voice I've been hearing so far is Bob Elliott (used to be on the IC at Bridgewater/advisor to Ray Dalio and taught multiple courses on the banking system) - https://twitter.com/BobEUnlimited

Probably a good idea to just ignore all the moronic VC threads and fintwit influencers.

It seems like the most likely scenario is that depositors will be made whole quickly and equity holders will get fucked since SVB's balance sheet looks pretty strong. Matt Levine came to the same conclusion.

View on HN

Can you please make your substantive points without snark and name-calling? The latter break the site guidelines and your account looks to have been making a habit of that. We ban such accounts and I don't want to ban you because you've clearly got some good points—so if you'd please review the rules and stick to them, we'd be grateful.

https://news.ycombinator.com/newsguidelines.html

will do. Thanks!

Appreciated!

What were they supposed to do? They knew the clock is ticking. Whoever bailed first will be Ok and the last will be left hiding the bag. That goes for their portfolio companies as well. This is game-theoretic decision making. I am a bit rusty on Nash equilibrium, so I am not sure if this was inevitable outcome.

If you map 'pull all your money' as defect and 'leave your deposits in the bank' as cooperate, you can make a case for it being a prisoner's dilemma, though I don't claim to know if everything would be okay if people didn't pull out.

The Nash is to defect in the prisoner's dilemma. That's not the whole story, mind you, because there's a case for tit-for-tat and similar strategies in repeated games, but this looks like a one-shot game from what I know, so... yeah.

Prisoner's dilemma doesn't have a part where they take all the prisoners from the Silicon Valley Jail and move them to the First Republic jail with a bunch of other prisoners and ask them the same question, do they?

But it is never a one-shot game. It is now a fact that “capital firms Peter Thiel’s Founders Fund, Coatue Management, Union Square Ventures and Founder Collective all advised their startups to pull their cash from the bank”.

Valley is small. There is real loss of wealth now, because SVB is no more. And there’d be likely a ripple effect.

I suspect, there will be a bailout. Silicon Valley has very good relationship with the current White House. It would not be a stretch to make the case for a bailout. All the Government has to do is to provide $10-$30B (?) worth of liquidity of the next 2 - 3 years, until the rates will go down. Then, they could sell the whole fixed income portfolio.

The very strong argument for a bailout is if there isn't one, every company will look very hard at where their cash is, and shifting it around could cause more bank runs. With how fast this went, there needs to be something concrete on Monday.

And they’d make money on the bailout, right?

What makes you think rates are going down in the next 2-3 years? Rates could continue going up and put SVB further underwater

To be fair if there's a run on a bank happening you want to be following the people creating the run so you can be one of the first ones out the door.

Incidentally there's a major moral hazard problem for bank runs in that unlike e.g. stock market or commodities panics, there isn't any ability to "buy the dip" or hold through the panic.

“Becker held the roughly 10-minute call with investors at about 11:30 a.m. San Francisco time. He asked the bank’s clients, including venture capital investors, to support the bank the way it has supported its customers over the past 40 years, the person said.”

The above was the right answer that would have maximized the wealth generating machine.

Agreed. I wish the VC community had come together and financed a stock purchase with SVB to recapitalize it instead of running for the hills. Unfortunately, VCs aren't exactly known for their community-oriented mindset.

Sad thing is that the defectors are clearly the short-term winners in that case.

Long term, the purchasing power of the Valley will be diminished and financial services disrupted. This might result in substantial disruption of businesses and cascades of business failures. Which may in turn affect defectors. But this seems to be the only negative feedback loop.

>the most likely scenario is that depositors will be made whole quickly

Insured depositors (up to $250K) will absolutely be made whole quickly. They'll get their money on Monday.

Uninsured depositors, up to some multiple of the $250K, will probably be made whole, but it will likely take a couple of weeks.

Uninsured depositors beyond whatever multiple the FDIC decides to accommodate (likely somewhere in the millions to tens of millions of $), will probably get some fraction of their deposits back, but it could take several months.

This talking point that all depositors will be made whole quickly is dangerous and almost certainly false. The FDIC needs to liquidate or find a buyer for SVB and the value of the assets isn't enough to make everyone whole, beyond that there is the FDIC insurance fund but it's likely insufficient, beyond that requires Congress passing a law to enable a further bailout.

> So the VC thinkbois basically created a run on SVB. Nice job guys!

You say it like that absolves SVB.

SVB has around $200B under management. The bank run on Thursday pulled $41B out of it. That is 20% of their total AUM. That is an insane bank run. No bank is setup for that kind of bank run.

> No bank is setup for that kind of bank run.

Banks in Scotland's free banking era used to have a capital cushion of around 1/3 of their balance sheet.

Not saying that this kind of capital cushion is viable in today's regulatory environment. Just to give an example of existing real world banks that were set up for that kind of bank run.

In any case, I agree that scarcely any bank today would be set up for that kind of run. That still doesn't absolve SVB. They put themselves in a position where it's rational for their depositors to run.

Are you sure about that?

I would make me very nervous if every single bank had ~50% of it's balance sheet in rock bottom rate 10Y bonds that have dropped in value 30+% in the last year

I'm pretty sure most banks have assets > liabilities even after rate hikes, which is the root cause with SVB, not the bank run

> I would make me very nervous if every single bank had ~50% of it's balance sheet in rock bottom rate 10Y bonds that have dropped in value 30+% in the last year

This is a separate issue. SVB is likely insolvent because of their MBS issue. But their liquidity available last week is significantly less than their assets.

Correct me if I'm wrong, but nowadays any (sufficiently large?) solvent bank can just borrow from the Fed and survive such a run.

> Correct me if I'm wrong, but nowadays any (sufficiently large?) solvent bank can just borrow from the Fed and survive such a run.

Why didn't SVB borrow then? I understand they were the 18th largest US bank last week. And they had more assets than just 47B that was the bank run. But the FDIC took them over because they didn't have the liquidity to serve the needs of the bank run as of Thursday night.

I think that you can not just arbitrarily borrow from the fed, there are limits.

It's because they weren't solvent.

I think that isn't how it works. The fed just yesterday added the ability for banks to borrow nearly unlimited amounts against their assets. I do not think this existed previously. It is an emergency measure.

You missed the important part - they added the ability to borrow while valuing collateral at cost. Previously it had to be valued at market price, which was not sufficient for SVB (because it was insolvent on the mark-to-market basis).

> So the VC thinkbois basically created a run on SVB. Nice job guys!

On purpose. Billions of their wealth depended on FED's easy monetary policy (ZIRP, NIRP, QE infinite) for last 30 years or so.

For likes of Gary Tan, Peter Thiel and Cathie Wood (and Ray Dalio) creating moral panic and trying to force FED's hand to reverse normalising monetary policy is the only possible way to keep status quo and their wealth.

Is The Fed Listening? | ITK with Cathie Wood - https://www.youtube.com/watch?v=jvL2Q0cJLyo

30% of YC companies exposed through SVB can’t make payroll in the next 30 days - https://news.ycombinator.com/item?id=35100743

Peter Thiel’s Founders Fund Advises Companies to Withdraw Money From SVB - https://www.bloomberg.com/news/articles/2023-03-09/founders-...

It definitely seems to be done on purpose, but they didn't tell all their friends (David Sacks for example). For Thiel this was a win-win thing, his companies are all fine while hurting the competition and he gets to whine about government from new angles. Also quite in keeping with his soulless character

SVB created a run on themselves by betting against rising interest rate.

Nope, wrong. Their balance sheet is strong and they are fully solvent. This is completely different from an FTX situation.

They DID make a bad bet, but they were still solvent as a bank. The liquidity crisis was caused due to poor communication on SVB's part and typical VC herd mentality.

Read the full article by Matt Levine for a breakdown.

They weren't solvent, 100 billion in 2033 dollars doesn't equal 100 billions in 2023 dollars.

Solvency and liquidity are two different things.

2033 dollars lost a lot of value due to interest rate increases. Higher interest rates means that 2023 dollars would convert to more 2033 dollars, so now if you want to sell 2033 dollars for 2023 dollars you would get less back.

If the interest rates weren't increased they would just have sold their 2033 dollars and everything would be fine, a bank doesn't go under just because it takes a while to get billions of dollars, the whole issue is that they ran out of assets due to their assets having lost value.

That might be true, but it's got me thinking. Suppose you owe $100M and had $80M of cash. That would be considered both involvent and illiquid, correct? Now suppose you go out and buy $80M worth of 10 year treasuries with the cash you have. At current rates you will have $115M in 10 years, plenty of money to pay of your liabilities of $100M. Does that mean that if you're "insolvent" you can magically make yourself "solvent" by buying a bunch of bonds?

No, because you need to value the bonds at their current market value, which would presumably be 80 MM. The fact that you believe an asset will get you a certain amount of money doesn’t mean it’s worth that now.

If the debt is due in ten years, it won't be worth $100m, and you can presumably just pay it off with the cash.

In that context no it’s the same thing. It’s profitability and balanced books (liabilities matching assets) that are different. A business can be profitable and insolvent if cash flow temporarily dries (e.g. you can’t hold employee salaries because customers take too long to pay invoices). A viable business needs both profitability (long term) and cash flow/solvability (short term).

Unmatched assets and liabilities (like FTX) is not insolvency, it’s fraud, and mechanically causes insolvency when the cash is gone after the run.

They are, and this is a full-on solvency issue, not just a liquidity one. Think of it this way: suppose you have an investment that pays out $100 in 2031, plus a low rate of interest that due to interest rates increasing substantially in 2022 is now below what you can get from savings accounts or ultra-safe investments like Treasury bonds. That investment may have been worth $100 in 2021, but it's worth substantially less in 2023 because someone with $100 in 2023 could put it into those higher-interest investments and end up with a lot more money by 2031. However, banking regulations allow banks to not immediately recognize that drop in the value of their reserves by claiming they'd hold the investments to maturity.

Matt Levine article: https://archive.is/GafKl

I read Matt Levine's write-up: as always, a delight!

However it didn't strike me as him making a case that SVB is solvent.

His thesis is that they'll bail out the depositors to avoid other runs.

Yes, that's my reading as well. Which sounds rather different to me than 'Their balance sheet is strong and they are fully solvent.'

Their balance sheet is weak, but they kept going because some accounting tricks allowed them to defer mark-to-market on their losses.

I’m not saying they are insolvent. Their bet led them to having to raise money, and that gave enough signal for their depositors to pull money. Don’t blame VCs, their reactions were rational in this situation.

Only inasmuch as they knew other VCs and depositors would be panicking in response to the signal.

There's another possible stable equilibrium where no one panics because they know no one else will panic and that the bank is solvent.

Well SVB can't raise the interest rate they pay to depositors much without taking a bath, and I don't see why someone would want to keep their money in the bank once it is paying a notably lower interest rate than most other banks. Maybe some more time or wider spread of people wanting to withdraw would've kept them alive long enough to salvage something. But they put a huge amount of their money into this bond that they will be stuck with for a long amount of time. I dunno bank run almost just seems inevitable here.

No, the bank is actually insolvent.

> On March 8, 2023, the Bank announced a loss of approximately $1.8 billion from a sale of investments (U.S. treasuries and mortgage-backed securities). On March 8, 2023, the Bank's holding company announced it was conducting a capital raise. Despite the bank being in sound financial condition prior to March 9, 2023, investors and depositors reacted by initiating withdrawals of $42 billion in deposits from the Bank on March 9, 2023, causing a run on the Bank. As of the close of business on March 9, the bank had a negative cash balance of approximately $958 million. Despite attempts from the Bank, with the assistance of regulators, to transfer collateral from various sources, the Bank did not meet its cash letter with the Federal Reserve. The precipitous deposit withdrawal has caused the Bank to be incapable of paying its obligations as they come due, and the bank is now insolvent.

Source: https://dfpi.ca.gov/wp-content/uploads/sites/337/2023/03/DFP...

> Nope, wrong. Their balance sheet is strong and they are fully solvent.

The fact that they sold assets for a loss indicates that they are not.

The fact that the FDIC took over the bank indicates that they are not.

The fact that the bank is winding down with no other bank willing to buy them indicates that they are not.

What fact would you put forward to support your assertion that the bank is fully solvent. Because right now you are saying the opposite of the market, the FDIC and the banks competitors.

If they were fully solvent it would be pretty trivial to find a buyer to keep the bank to Silicon Valley startups going.

The fact that no one will take on their liabilities is very damning.

I think what they meant to say is they were solvent, till the run.

I partially agree with you... except "The fact that they sold assets for a loss indicates that they are not." is not true. Firms sell assets at losses all the time -- that doesn't mean they are not solvent. That means they had a loss, but their assets were still greatly in excess of their liabilities (as reported by the FDIC 211b vs 195b).

What they were not able to do was have enough cash on hand to deal with a bank run.

The assests they sold recovering as soon as intrest rates drop is as close to a sure thing as you get in finance.

The only reason to realise the loss is they needed the money now.

>as soon as intrest rates drop

Which could take a lot longer than people seem to expect. If you look at the past 100+ years of economic history, 2009-2021 is a complete anomaly. ZIRP is not the normal state of affairs, nor are negative real rates. There was an attempt at returning to normalcy from 2017-2019 but then Covid hit. Starting last year the Fed is attempting once again to return to a normal monetary environment. Rates are likely headed higher than people think and will remain there longer than people think.

What you are saying reported by FDIC as 211B vs 195B was December 31st IIRC

And it appears that wasn't market to market

Since then they have had $40B in withdrawals causing them to sell all liquid assets at a loss and not being able to furnish withdrawals

The California regulator has explicitly stated they are insolvent: https://dfpi.ca.gov/2023/03/10/california-financial-regulato...

How is a bank spunoff back again after a fed takeover?

The FDIC will probably sell the carcass to another bank in the next few days.

The fact that a bank can be sold off in "the next few days" yet it takes the average person 30 days minimum to close on a house...

If you are willing to sell at a discount, you can sell faster.

But more importantly, you probably don't have as many highly paid bankers, lawyers, accountants and regulators trying to push the deal through. Nobody could afford that.

They’ve also created new banks in the past (presumably when they can’t facilitate a shotgun marriage).

E.g. the first big bank failure in 2008, and the 3rd largest failure ever (after WaMu and now SVB), IndyMac became OneWest

https://en.m.wikipedia.org/wiki/List_of_bank_failures_in_the...

> On March 19, 2009, a seven-member investor group, IMB Holdco, led by Steven Mnuchin—which included billionaire Christopher Flowers, John Paulson, Michael Dell, and George Soros—purchased Independent National Mortgage Corporation (IndyMac Bank) of Pasadena, California for $13.65 billion from the FDIC and created OneWest from the remains of IndyMac

https://en.m.wikipedia.org/wiki/OneWest_Bank

In this case they created a new bank. From an outsider's perspective, it implies the bank run was either too sudden, or SVB would have been too expensive, for FDIC to be able to run an expedited auction.

In one of the earlier two threads a poster linked to this segment of 60 Minutes: https://www.youtube.com/watch?v=TAE8i40A5uI . (I would link to the comment directly but I can't find it right now. Apologies to the poster.) In case of a relatively small bank, the FDIC had run a secret auction a few days earlier, and had an acquirer ready on the day.

SVB was pretty big. Lining up a buyer that can absorb and handle a balance of ~$200B might take a while, and interestingly there may be political reasons to discourage depositor haircuts. Buying at firesale prices favours those with deep enough pockets. Buying at par (or near enough) needs more than just money - it gives advantage to those who can do extensive diligence really fast.

Wait should I buy SVB?

Trading in their shares is halted.

However, in principle, if you believe in the efficient market hypothesis, and if SVB shares would still be traded, they would be traded at a (low enough) price that makes buying their shares reasonably thing to do.

Curious what the efficient market hypothesis has to do with this?

Seems pretty opposite to me -- I would think if you are a value investor then the price may have dropped low.

> Curious what the efficient market hypothesis has to do with this?

Let's back away from SVB, because it's not currently traded.

In 2020 the car rental company Hertz went bankrupt. Nevertheless, the stock traded a low but non-zero price.

You can either say that people were crazy, or if you take the efficient market hypothesis to heart, you can see it as an indication that there was a positive probability that Hertz stock would be worth something after the bankruptcy.

As for the value investor perspective: for SVB you would try to come up with your own valuation, compare that to the market price (and thus market capitalisation), and if the latter was low enough, you'd buy.

As an efficient market guy, you'd skip the first step, and just assume that the market price is probably fair enough; and just buy an index fund that includes a small fraction of SVB.

(SVB used to be in the S&P 500.)

If you're an efficient market hypothesis, you don't believe there is any value that can be had because it's completely built into the market, so there's no reason to pay attention to the news and just buy a fund in the first place.

This likely depends on which version of EMH you're referring to, and what information is "efficient", but typically speaking you don't care if things drop or rise because you believe that it's efficiently priced.

Can you help me understand where you and I differ?

Yes, details depend on exactly what version of EMH you'd subscribe to. I think we mostly agree.

One fairly sensible version of EMH is the notion that even an efficient market pays you to take certain risks. Eg an insurance company might be paid to take weather risk when they insure crops. If there's enough competition, that risk will be efficiently priced.

People can spend extra effort to learn more about crops or the weather, and thus get a better handle on the probabilities and variances. At some point you hit diminishing returns. Different people have different amounts of productivity, ie how much effort leads to how much improvement in understanding of risk. The market price will be driven by the most productive people and companies; because they hit diminishing returns last. (To give a silly illustration: I could spend a lot of effort trying to understand the weather better, but because I have no clue, that effort wouldn't do me much good. An expert will do much better. And so it's the experts who effectively set the price in competition with each other.)

The same applies for credit risks that a bank faces when it makes a mortgage loan.

In an efficient market, the risk for holding SVB stock would be priced efficiently. But you'd maybe want to investigate whether it's the kind of risk you want in your portfolio. Or, if your investigative abilities are at the productivity frontier, you might also want to just check out SVB in general, to see if the market price is slightly off compared to your expert judgement.

Reply on news.ycombinator.com