Hacker News Reader

Comment by TheAlchemist | Hacker News Reader

You know what I hate the most in all of it ?

There was no panic to begin with - in the worst case scenario, the cut for depositors was definitely less than 10%, most probably 0.

The Jason Calacanis, David Sacks & co, decided that even that small % of losses were enough to start a bank run in order to get government involved - and make sure they are made whole.

Stress on 'THEY'. Just look at their recent tweets - did Mark Cuban said today that unlimited FDIC is a very bad thing - should 'Never' be a thing ? Amazing stuff. (edit: Here is the link => https://twitter.com/mcuban/status/1635282882259476486?s=20 )

Make no mistake - the 3 banks in trouble (there will be more. Once the fire is lit, it's hard to stop it) were extremely specific and were crypto / VCs banks. It's not random banks that failed. They failed specifically because of business they were servicing.

View on HN

I agree, and I see three fallouts from this event:

- Yellen folded fast, so the market will test her again (moral hazard)

- Powell will hesitate to raise rates further, concerned about hidden stresses building up in the system. This was basic duration risk, and the fact that they missed it entirely will raise the worry of what else they are missing.

- FDIC will probably have to be reformed. You cannot guarantee everything, all the time. Maybe an opt in system where large depositors can buy individual coverage.

> FDIC will probably have to be reformed

Why? Isn't the guaranteed deposits above 250k mainly 'insured' through Fed offering unlimited loans back by the nominal price of bonds owned by affected banks?

I should have explained better. My sense is that right now there is an implicit expectation in the market that the 250k threshold is no longer relevant, and that the FDIC is implicitly insuring everything that looks remotely like a deposit, regardless of the amount.

This doesn't work. The only way to draw a new line in the sand is some sort of reform.

The key issue is that the new line has to be "subgame perfect", meaning it has to be credible that if a bank steps over the new line, it will not simply be bailed out again.

Yes, now I wonder if anyone is allowed to fail. David Sacks complained that if the regionals aren't protected, then everyone will only bank with the big four. Yet if everyone is protected even when the bank makes poor decisions, that's a run on the FDIC itself. So as you say, where is this new line, and who can be trusted to allow those who cross it to fail?

<< Yes, now I wonder if anyone is allowed to fail.

That was one of the questions on the call/presentation I was recently on and no one has a clear answer at the moment ( I suppose it is not a surprise since we can't take Yellen at her word ). Some clarity will be needed and sooner rather than later if stated policy and rules are to be believed to be real policy and rules.

FWIW, odds are, just about every bank by now has either reviewed or scrambling to review their exposure.

I do not envy the weight of Yellen's decisions, because from where I sit it is still hard to tell if it was a 'less bad choice' available.

> odds are, just about every bank by now has either reviewed or scrambling to review their exposure

The lesson they learned is to shout global emergency when your regional bank can't meet withdrawals due to poorly managed finances.

Banks will not be more prudent, in the long run. They've just been taught that the government will protect depositors beyond federally insured limits. So now they can make riskier bets.

Keep in mind that SVB equity was totally wiped out. Everyone most certainly was not protected.

The people who make the poor decisions absolutely took a hit. The Board and management of the bank all took a hit and lost their jobs.

I don’t see a big problem with depositors being made whole. The bonds that backed their deposits are fully intact.

> The Board and management of the bank all took a hit and lost their jobs.

Right after they cashed out significant amounts of SVB stocks.

Don't be too surprised if the FDIC (or perhaps the SEC, since we're talking about equity) forces clawbacks of those payouts, not to mention salary and bonuses in general after December 2021.

What is the case against them? You can't claw back without demonstrating some law was broken. Making a bad financial decision that causes a business to fail isn't against the law unless you somehow profited from it. And, from what's publicly known, SVB's CEO didn't come away with more money from the bank's closure than he would have had he continued running the bank.

The question is, what would it take for a bank to be allowed to fail? The FDIC insurance line of $250k must've meant something.

SVB failed. You should look up the FDIC mission statement; it has nothing to do with drawing lines in the sand.

> The mission of the Federal Deposit Insurance Corporation (FDIC) is to maintain stability and public confidence in the nation's financial system.

I see nothing there about "we will guarantee deposits of any amount over $250k."

The decision to guarantee SVB may wreak havoc on the financial system as banks feel comfortable continuing to make risky bets and thus attracting more customers, knowing that the fed will rescue the customer deposits that backed those bets.

Greg Becker isn't going to jail, and he isn't going to be financially ruined. There are thousands of people willing to fill the role of extracting money from the government. The trick is to not give into their attempts to bend the rules.

It’s possible for the bank to capture a % of the deposit insurance by taking profit neutral risks. This is achieved by just increasing the risk. For example the profit neutral bet:

  0.5: 1
  0.5: -1

And the bet is made of 80% deposits 20% capital.

For the bank the profit before paying interest to depositors:

  0.5 * 1 - 0.2 * 0.5 = 0.4

The bet is profit neutral but the banks profit comes from increasing the risk of triggering the insurance.

For the insurer the cost is: 0.5 * -1* 0.8 = -0.4

If you want to run a ‘scam’ bank that makes money from looting the FDIC fund then having your capital going to zero sometimes is part of the cost of doing business. This is why it’s important for the insurer to try and control risk and insure there is enough capital so the loot equation does not work.

> I should have explained better. My sense is that right now there is an implicit expectation in the market that the 250k threshold is no longer relevant, and that the FDIC is implicitly insuring everything that looks remotely like a deposit, regardless of the amount.

If rich people think that, I think they deserve to be taxed at more than 80%, because that's stupid.

FDIC move was perfectly rational and cost-optimized. Had they not done that, there would have been more bank runs, and they would have to actually pay out 250k/account on many more accounts than just SVB's, while ""saving"" SVB cost almost nothing.

You yourself just explained the reasoning as to why they would think that and seem to agree as well, not sure why you'd call it stupid.

The simplest answer is "excess deposits insurance". Banks (or third-party insurers) could charge a monthly insurance fee for balances over $250k that provides insurance over the FDIC $250k cap. This allows everyone else under the $250k cap to bank without paying fees (the bank already paying insurance on those as per FDIC), and provides greater coverage to those above the cap. Think of it like an Umbrella Insurance policy. Some banks might provide the insurance for their most favored clients if they so choose, or embed the insurance inside other services.

> Banks charge a monthly insurance fee for balances over $250k that provides FDIC insurance over the $250k cap

The thing is .. the FDIC is already funded by banks.

https://www.fdic.gov/about/what-we-do/index.html

"The FDIC receives no Congressional appropriations - it is funded by premiums that banks and savings associations pay for deposit insurance coverage. The FDIC insures trillions of dollars of deposits in U.S. banks and thrifts - deposits in virtually every bank and savings association in the country."

How the banks choose to pass that cost on to customers/shareholders is up to them.

It's instructive to read https://www.fdic.gov/analysis/quarterly-banking-profile/qbp/... for the last available quarter; the insurance fund is on page 24. I can quote you two numbers and you can decide whether they are big or small: there is $128 billion in the fund, and this covers 1.27% of total US banking deposits.

Yes that's correct. I'm suggesting an additional fee that is paid by the account holder who holds excess funds above the amounts that are already covered by the bank in their fees for FDIC insurance coverage. I'm aware of the existing bank-paid insurance that covers the mandatory $250k FDIC coverage up to the cap. The difference that I am suggesting is that the umbrella insurance is paid by the account holder to the bank for excess insurance coverage. This allows insurance coverage for account holders over $250k, of which there is no insurance coverage, optional or of any kind, available today.

I think we've just discovered that account holders over $250k _are_ (or can be) covered regardless?

Surely the logical counterparty for the insurance is not the bank, but the third party insurer? i.e. that people should explicitly have to pay for FDIC coverage themselves?

> of which there is no insurance coverage, optional or of any kind, available today.

This is basically a credit default swap for bank accounts, and if you wanted to insure the reported $450m that Roku allegedly had with SVB, someone would have sold you a product I'm sure.

edit: remembered "insured cash sweep", which is the product that everyone should have been using. See https://www.intrafinetworkdeposits.com/ or https://www.cbhou.com/Resources/Customer-Corner/entryid/237/...

I think this is a case of insurance coverage. The fact that there's some implicit coverage of excess amounts is not sustainable long-term. Just like we buy insurance in the unlikely event that our house is flooded or damaged by an earthquake, so too should companies with significant assets in banks purchase, on their own, insurance to cover the potential possible, but unlikely, situation of a bank bust causing them to lose most of their bank account funds. In this way the government can focus on insuring the general public, and those "too big to fail" can get private insurance to cover their own risks. This is just my opinion and a possibility. Certainly if governments want to come to the rescue or we have a "survival of the fittest" with the loudest, angriest parties getting their way, then we can run things that way as well.

Is FDIC, which is paid for by banks, "the government"?

Thank you, I’ve been confused about whether this product even potentially exists (as opposed to shotgunning your deposit across lots of banks to game the FDIC limit) and this is the first time I’ve seen it properly named.

Right now there aren't many satisfactory options for insuring excess deposits but here are a few strategies (including your aforementioned "shotgunning" approach): https://www.forbes.com/advisor/banking/ways-to-insure-excess...

Insurance won’t matter because the insurer will be bailed out. Remember AIG? They sold insurance (CDS) which everyone bought then when it was time to pay out “oops bail us out please” and they were. So insurance is meaningless because who cares if the risk is priced correctly because the insurer will be bailed out themselves.

The money pipe was opened then and SVB showed it will never be closed. The capture is complete.

Clearly there needs to be some sort of reckoning when it comes to bailouts. That's a political problem for which I don't have any simple answer.

Agree with 1 and 3.

Not with 2 though. The BTFP (not to be confused with BTFD :) ) program is removing completely the duration risk (for a year). This lets the FED raise the rates as high as they want without putting banks balance sheets at risk (barred some other unforeseen consequences).

If we have to buy insurance to insure our money, the financial sector will just game that, too, and make even riskier bets. That we (taxpayers) end up paying for.

This is the gaming of the system that is resulting in the government FULLY covering the bets of the rich. They could have bought insurance, but running a bank is cheaper.

Not necessarily. If insurance companies were allowed to set prices based on the bank you're deposited with, then what's currently a political process could become a market-based process. Have assets that are exposed to interest rate hikes? Higher insurance rates. Have assets that hedge against changes in the interest rate? Lower deposit insurance rates. Won't let us see your asset mix? Assume the worst case scenario.

So in this universe, when you go to get insurance at SVB, you see that it costs you 1% per year, which seems high to you. You go to the insurance company's website, and see that insurance for Chase is only 0.5%, so you switch to them instead. SVB is pressured to change to a less risky strategy by the market, not b the government.

(NB I'm not a libertarian, nor am I recommending this approach; I'm just saying that in theory such a system could work.)

EDIT: Upon reflection, this might just punt the issue down the road: Suppose SVDI (Silicon Valley Deposit Insurance) gives you a rate of 0.1% per year at SVB. Everything is fine until SVB fails, and then it turns out SVDI didn't have enough capital to back their insurance, and you lose anyway.

Maybe there are ways you could fix this, but my preliminary conclusion would be that you need some kind of regulation somewhere.

You can sort of fix this in a free market system through reinsurance which spreads the risk out across multiple counterparties with deep capital reserves. But ultimately the only way to eliminate the risk is through insurance by a sovereign with the ability to levy taxes and print fiat currency.

FDIC already charges different amounts per bank. SVDI would buy reinsurance.

Is "keeping your money in a bank account" supposed to be a "bet"?

(or, as an argument I made earlier, if your company is completely dependent on a single other company to function, such as AWS, is there really any more risk in having only one bank account?)

at this point, who could doubt that AWS itself is faaar too big to fail?

I would say 'yes'. Banks should start offering fully insured accounts (maybe they already do) and those accounts should cost more and possibly have different liquidity profiles.

Fully insured by whom?

A re-insurance company probably, or the FDIC which would charge a premium for the excess insurance.

Why banks specifically? Should there be insurance for AWS and Github accounts?

> maybe they already do

"Insured cash sweep" (which is basically "RAID for banking")

There already is DR and business continuity insurance so non-financial assets have an entire industry around insuring against events there.

Cash sweep is insured but only to a limit, there's nothing magical there and the underlying securities have default risk (or breaking the buck).

North of $250,000 it's a bet. If you're Roku and you have hundreds of millions of dollars in a single bank, that's absolutely a bet. Companies can get private insurance to cover money in excess of the FDIC, but they didn't. And now no one will, because billionaires know they can threaten bank runs and violence to get bailed out.

Potential solution? Make the bank board and C-suite pay for all the insurance out of their own compensation packages. And the the bank fails it comes direct out of their pockets.

When the bank fails, they end up living on the streets for their mistakes, and will never (hopefully) be able to do it again to another bank.

This was the "Lloyd's Names" situation: https://www.reuters.com/article/uk-financial-equitas-names-i... ; although they weren't decision makers, they were a special class of investors.

And insurance won’t even matter. Credit default swaps were suppose to be insurance and those holders were bailed out in 2008/9, remember AIG?

In 2008/9, The fact that a solidly red administration (Bush) wrote the check and a solidly blue administration (Obama) delivered it on bended knee to the banking industry made it clear who really ran things.

“Paying your bills is a moral hazard” - guy in the background of a tv interview that birthed the tea party in the US.

You're right. It can't just be about paying for insurance but also has to include undergoing and passing regular stress tests. Just because one buys fire insurance shouldn't give them a free pass to fill their house up with large amounts of flammable material and sparking wires.

They did have insurance... They had significant campaign contributions to the DNC.

> where large depositors can buy individual coverage

Q: Aren't large depositors already free to buy individual coverage from whomever offers it?

FDIC basically forces all depositors to have insurance in a generalistic hand-wavy way.

Seems we also need to force others to have insurance because they can’t be bothered and will just get bailed out — next time probably with taxpayer dollars.

Well then there’s 4:

The regulators failed massively, allowing banks to mask their duration losses. Obviously this needs to change, but they probably need to fix the other banks first (that are still hiding their losses).

The regulators had the regulations changed out from under them by Congress (at behest of the banks). They can’t regulate what’s not in the law.

As far back as 2019 the FDIC was sounding the alarm about regional banks. In 2020, SVB’s risk committee told them to change their asset mix, they were over ruled because it would lower profits.

If we are going to allow lax regulations fine, then the market (including depositors) have to do the regulation, but you can’t blame regulators in that case.

Exactly. Let’s not be so quick to say “the regulators failed”, particularly when the case is that even if the regulators did somehow fail and not have stronger regulations removed by Congress leading to said failure, these financial firms as a business failed first by making poor business decisions.

> but you can’t blame regulators

When I read "regulators failed", I don't think of Joe Regulator phoning it in at work. If Congress creates and oversees regulations, then they are the regulators in my book. This may be what the comment you replied to meant as well.

The rollback of Dodd-Frank was at least nominally bipartisan.

Limiting the feds power is broadly popular with the population. This is one of the outcomes of those desires, Congress representing their constituents wishes seems hard to call a failure just because the outcomes have some negative outcomes (all laws have negative outcomes for someone).

Note that this is a state chartered bank rather than a national one.

Thus, state regulators are the ones with the oversight power and the ability to really do things.

California Financial Regulator Takes Possession of Silicon Valley Bank - https://dfpi.ca.gov/2023/03/10/california-financial-regulato...

> SAN FRANCISCO – The California Department of Financial Protection and Innovation (DFPI) announced today that, pursuant to California Financial Code section 592, it has taken possession of Silicon Valley Bank, citing inadequate liquidity and insolvency. The DFPI appointed the Federal Deposit Insurance Corporation (FDIC) as receiver of Silicon Valley Bank.

> Silicon Valley Bank is a state-chartered commercial bank based in Santa Clara and is a member of the Federal Reserve System, with total assets of approximately $209 billion and total deposits of approximately $175.4 billion as of Dec. 31, 2022. Its deposits are federally insured by the FDIC subject to applicable limits.

---

Note California and "state-chartered" in there.

My (limited) understanding is that a state chartered bank is still regulated at the federal level.

From https://en.wikipedia.org/wiki/Bank_regulation_in_the_United_...:

  a Nevada state bank that is a member of the Federal Reserve System would be jointly regulated by the Nevada Division of Financial Institutions and the Federal Reserve.

> State regulation of state-chartered banks and certain non-bank affiliates of federally chartered banks applies in addition to federal regulation. State-chartered banks are subject to the regulation of the state regulatory agency of the state in which they were chartered. For example, a California state bank that is not a member of the Federal Reserve System would be regulated by both the California Department of Financial Institutions and the FDIC. Likewise, a Nevada state bank that is a member of the Federal Reserve System would be jointly regulated by the Nevada Division of Financial Institutions and the Federal Reserve.

And from https://www.fdic.gov/about/what-we-do/index.html

> The FDIC directly supervises and examines more than 5,000 banks and savings associations for operational safety and soundness. Banks can be chartered by the states or by the Office of the Comptroller of the Currency. Banks chartered by states also have the choice of whether to join the Federal Reserve System. The FDIC is the primary federal regulator of banks that are chartered by the states that do not join the Federal Reserve System. In addition, the FDIC is the back-up supervisor for the remaining insured banks and savings associations.

Right, but in SVB's case, since they are a member of the Federal Reserve System (like the Nevada example), they were regulated by the Federal Reserve in additional to the State.

The California bank in that example is regulated by FDIC because it's not a member of the Federal Reserve System.

In the run up to the 2008-9 financial crisis, Sheila Bair at the FDIC and Brooksley Born at the CFTC were making alarums about the banks and the housing market, but the dudes at the Fed, Treasury, and the Comptroller of the Currency were like everything's fine, go back to the kitchen.

There needs to be some kind of penalty for the C-suites for running a bank with no risk manager for 9 months too.

They had no interest rate risk hedges the entire time.

Edit: Downvotes with no replies. I hate it when that happens.

In this case, the SVB C-suite have all lost their jobs, and any equity (RSUs etc) they may have held has been zeroed.

The next SVB then knows to cash out early and treat the stock as zero-value.

What about claw backs and garnishment?

I think the risk manager think is a bit of a red herring, outside of pointing towards a culture that didn’t care in the first place.

SVB got blown up on a trade that is kind of at the core of what banks do (this is also why every bank is hurting). To even exist the bank has to be stuffed full of people very knowledgeable about rates risk at all levels of management. There should have been no shortage of people at any level, including executive, that understood the risks.

It wasn’t even like this was some weird bespoke product or something, they got blown up by plain treasuries afaiu. This is the sort of thing that’s “introduction to rates 101” material. They didn’t need a CRO to tell them this.

Why not buy a hedge? Their risk committee met 18 times.

Because HoldForMaturity assets can't be hedged? Or it cuts into profit and affects your stock and management bonuses which they weren't used to giving up, and your share losses due to the hedge imperil your company but just in a different way?

Oh to be clear I agree with the OP that they were incompetent/greedy, I just think that missing a CRO is at best a symptom and not a cause. Every single invite to a risk committee meeting should know exactly well what sort of rates risk they had.

As for the hedge, it’s probably a greed/returns question. If you buy then 10 year and sell everything between say 2 to 10 years as a curve hedge, you’ve basically bought the two year with extra steps. There’s no free lunch where you get returns of ten year without the risk on ten year.

You could imagine some imperfect hedge might be better but sort of same problem.

Did SVB do anything irregularly? How would the pre-2018 regulations have changed anything?

From what I’ve read, they would have to mark the portfolio to market (i.e. report its value in terms of current market prices), not say “we intend to hold this to maturity” and account it at face value. Like, the latter actually makes more sense to me, but only if you actually don’t need to sell which might be true for a university endowment (all assets, no liabilities or at least very predictable liabilities) not for a hedging institution.

Wouldn’t it have just made them insolvent earlier then? How does it actually prevent the problem?

Requiring that they provide more information earlier encourages them to behave more safely (by exposing them to scrutiny backed by sound evidence). And if they still go down, at least it happened sooner and there are fewer depositors to be harmed.

Not necessarily, since the gap in finances accumulated over time. They would have had to acknowledge asset price drops on their balance sheet much sooner. Basically they were allowed to build up a bigger integration risk with their accounting if they ever did have to sell due to not having to MTM in the mean time.

Was the accumulation duration larger than the regulatory mark-to-market period? As I understand it, SVG was heavy on US Treasuries, which lost a great deal of value due to recent interest rate hikes. As a result of a higher than expected deposit withdrawal rate, SVG sold US Treasuries for less than what they would have received if they had held them to maturity.

If the gap in finances was discovered earlier, would they have just converted more equities and other long-term higher-yeild instruments into treasuries and then go under because they couldn't get returns to match their deposit interest rates?

Had they been marking their holdings regularly they would have sold their depreciated holdings sooner when losses were smaller, instead of waiting until they had to to cover withdrawals.

Deposit interest is not fixed so that wouldn’t make them go under, they’d have to reduce their interest paid.

Banks are supposed to take duration risk like this - they borrow short-term (deposits) to lend long-term (business loans and mortgages). The thing that's new in the current environment is that the fed has hiked rates so aggressively that even banks that took sane levels of duration risk are in danger of insolvency if their assets get marked to market.

You have that backwards. You borrow long term, then lend short term. Trying to keep enough liquidity to match withdrawal rates to "keep money working".

Lending long (10 year) on money deposited for only a year is a recipe for disaster.

Every other institution borrows long to lend short. That's how money is usually made in a financial company, like a hedge fund or something similar. Banks are the counterparty (in a cosmic sense) to that trade: they borrow short to lend long. It is an inherently unstable business model, but it works if you trust them enough.

Yellen has always been too quick to fold, and we saw that in her time as fed chair. Wall Street and silicon Valley insiders know this, which might be why they immedieately tried to put their PR campaign's focus on her (rather than Jerome Powell, who is a little bit steadier at the poker table).

Yellen was the 11th President of the SF Fed [1] and the current president is her protégé [2]. Greg Becker (SVB CEO & president) was in SF Fed board of directors until Friday [2]. He successfully lobbied for less regulations for his bank. The SF Fed looked the other way.

[1] https://en.wikipedia.org/wiki/Janet_Yellen

[2] https://en.wikipedia.org/wiki/Mary_C._Daly

[3] https://www.reuters.com/markets/us/ceo-failed-silicon-valley...

Are you a trader? I traded global rates markets from 2012-2020 and yellen was probably the best central bank head I saw. She was the most stable and consistent. I would argue she was the least beholden to market pressure out of any bank head, except perhaps Draghi, who just did not care much the last couple years. She was also the least likely to react to congress or press questions and cause unnecessary market volatility.

Yeah this previous comment is probably just good ol' fashioned misogyny. There's no consensus that Yellen is quick to fold from the previous crisis - these people also have real responsibilities so it's not necessarily 'folding' anyway to respond to stakeholders.

There is also no evidence that the previous comment was made due to misogyny.

I was in Wall St for the end of her tenure. I admired her handling of the beginning of her time as Fed chair (not raising rates too early when Wall Street was yelling for it), but IMO she ended up folding to pressure from politicians and Silicon Valley types to keep rates low for far too long. She was very good at not listening to the market (which is arguably a bad thing - when your rate target is 0-25 bps and rates are trading at ~20 bps, maybe you should listen?) or to the traditional Wall Street types, though.

Wow you managed to load a ton of sexism into a short comment!

Parent commenter mustn't have received the memo that women are immune to criticism?

I'm truly curious why you perceive the parent comment to be sexist? Apart from using "her" pronouns, I don't perceive any mention of gender... What am I missing?

If you saw sexism in that comment, then I'm pretty sure you're the one with the sexism issue

Yellen “folded” bc risking the broader financial system in order to give it hard to a bunch of tech bros is itself a hazard of moralizing.

Changing the rules during the game (full deposit insurance, one off advantageous loans to troubled banks) also erodes the moral by standing firmly behind reckless and risky (risky for the masses not self!) behaviour and entities, unlike the rest! For the rest the ordinary rules apply.

Rules don't metter from now on or what?!

They didn't change any rules - they just followed the processes they already had - which was to declare the bank systemically important.

Tell it to other cases where deposits are lost above the insurance limit, tell them please to their face that the rules are the same and not changed ad-hoc.

I would, but I’ve been looking for a case where were over the limit deposits were lost and can’t find one. Also, if it’s before 2008, the laws were significantly changed and the rules wouldn’t be the same but not ad hoc either.

Except that there were no blanket federal guarantee in previous failures BEFORE some private institution acquired the remians of the failed bank and just then they CHOSE standing for full the deposit if they thought so (most of the cases). Not always, not for everyone. Private cas to case agreement not new central rule! The rule was only up to the limits of insurance and so The Enloe State Bank, Covenant Bank, NOVA Bank, Jasper Banking Company, Eastern Shore, Home Savings of America, and several more (all after 2008) where clients did not have the balance above the insurance limit. Do you have some incentive here attempting to mislead?

Banking stocks are down as unlike the 2008 bailout capital is now at risk. Rules may matter but the Trump signed roll-back of Dodd-Franks removed these. This created an invitation to load up on risky assets which on aggregate are estimated 600B under water. We have removed rules preventing systemic risk. Now we face systemic risk. Crying for narrow rule adherence is not situational adequate.

its just going to keep getting worse and worse because there's no competitive advantage to responsible behavior. At this point you're a fool if you don't take unreasonable risks with your money.

> You cannot guarantee everything, all the time.

Why not? What’s keeping the Fed from working with Treasury and the FDIC to provide unlimited liquidity as needed for deposits? No doubt there will be higher order consequences, but I don’t know of any operational reason why it isn’t possible so long as the current monetary regime exists at all.

Powell and Yellen put a system in place over the weekend where all banks can borrow against their bonds at par, thereby preventing any other bank runs.

The duration problem is that banks have bonds/CDOs on their books marked-to-maturity, but when faced with outflows banks are forced to sell them at market prices (much lower because interest rates went up). This turns an illiquidity problem into an insolvency problem if a bank bought too many low interest long duration bonds.

Last week, when a bank had balance sheet consisting of bonds with a market price of $75 but valued at $100 marked-to-maturity they would get into trouble when faced with too many withdrawals. Today, they can borrow $100 from the Fed at a modest interest rate with the bond worth $75 as collateral. This fixes the bank run problem, because the bank can go to the fed to get the money they need to cover any outflows. This deal only covers quality bonds and only those bought before last week. This fix is not an invitation to start lending recklessly, it's just a stopgap measure to allow banks to lose money slowly on the bad debt they already have.

This all means Powell can continue to raise interest rates without having to worry too much about bank failures.

So, the solution to bad assets is more debt? How are the banks going to cover that debt when their customers are fleeing and interest rates are rising? I fear this is only kicking the can down the road by a few months (or years) before even worse fallout occurs when the failed banks have to make substantially larger interest payments.

Edit: a better term might be Zombie Banks.

Welcome to Japan circa 1990s which is now why Japan has the economy it does today.

The idea is that when customers know a bank has effectively an unlimited line of credit to pay for outflows that there is no reason to panic. Conceptually, the credit line works perfectly even if it's never used.

I agree it's not a solution, but it's smart and effective stopgap measure.

Why would customers flee if there's no risk of the bank running out of money?

In a constant intereest rate environment the bonds' market value will tend towards par value the closer they get to maturity

I disagree. I worry that this structure may effectively tie Powell's hands and prevent him from raising rates further.

As you describe it, banks can post collateral with a market value of $75 in return for $100 of loans from the Fed. This is not a panacea. There is a reason why this is not normally done.

The problem arises when this loan has to be repaid. At that point, the bank will have to repay $100 to recover $75 worth of collateral. There is a gap between what the bank must repay and the value of the collateral it gets back. This means that, at that point, banks may find that they are insolvent and unable to repay the loans. This is the banking version of "jingle mail", sending the bank's keys to the Fed instead of repaying the loan.

The higher the rates, the lower the value of the collateral, the greater the gap, and the more likely the banks are to default on their loans when they come due ...

... silly me, of course not ... at that point the banks will expect the Fed to bail them out again and cover the gap through some "repurchase" voodoo where the banks can "repurchase" their collateral at market value and have their loans forgiven.

Banks generally do well in higher interest rate environments. It's in a ZIRP climate that banks struggle. By kicking the can down the road banks can use the profit they're making from the business they do today to pay off the dumb debt they acquired in the past couple of years.

In the US alone banks are underwater to the tune of 650bn (that we know of). It will take a while to pay off.

The banks might get an outright bailout -- it wouldn't be the first time -- but nobody wants that. Not Powell, not Yellen, not Biden, and certainly not you or me.

> Banks generally do well in higher interest rate environments.

It’s not about a high vs low interest rate environment. It’s about a falling versus rising interest environment. In a rising interest rate environment, a banks bond holdings is continually losing value.

> The problem arises when this loan has to be repaid. At that point, the bank will have to repay $100 to recover $75 worth of collateral.

This is not much of a problem if the Fed allows the loan to be rolled until the bond matures. Then, at maturity, the loan's principle is paid by face value of the bond.

The only issue is that, in the meantime, the bank does have to pay interest to the Fed on the loan. This is a tightening; it's just not as tight.

The face valve of the bond cancels the principle, the bond's coupon cancels some portion of the loan interest, and the only thing left at the end is a cashflow from the bank to the Fed representing the remaining interest. Heck, the Fed could make this cancellation explicit, create a bond representing this cash flow, and sell it to somebody else.

This also doesn't prevent Powell from raising rates further. If anything, it eliminates a reason why he couldn't.

Do the banks actually have to repay the loans or can they just sign their bonds over to the Fed?

> The problem arises when this loan has to be repaid. At that point, the bank will have to repay $100 to recover $75 worth of collateral.

Assuming an absence of default, and the Fed being able to wait being paid back until the bond matures, then the collateral will be worth its par value (ie. $100).

> This deal only covers quality bonds and only those bought before last week.

> This all means Powell can continue to raise interest rates without having to worry too much about bank failures.

Do you think banks will stop buying bonds? Otherwise I don’t see how the latter follows from the former.

> - FDIC will probably have to be reformed. You cannot guarantee everything, all the time

You absolutely can (should is the question)?

The second question is whether you want depositors to care about the financial health of their banks or not. I'm on the no side of that conversation. So you let regulation take care of the capital requirements and so on whilst allowing people to feel safe with deposits.

its impossible to guarantee all deposits

Not for the United States government, the issuer of fiat currency. They can guarantee all deposits. Money is a concept.

> You cannot guarantee everything, all the time.

But it's not everything, not even close. It's only the amounts deposited in the standard class of bank accounts. Companies with large amounts of cash keep very little of that in bank accounts, it's in money market funds and other cash-equivalents.

It's certainly feasible and possible to have the "risk" part of the bottom of the risk-reward curve to be at the same level as the viability of the federal government. The risk isn't quite 0, but if the federal government collapses you have bigger problems anyways.

Especially in a high interest rate environment. Bank accounts essentially have a 0% nominal rate, which in a high interest rate environment is a very negative real rate so there is lots of margin for implicit or explicit insurance.

> Companies with large amounts of cash keep very little of that in bank accounts, it's in money market funds and other cash-equivalents.

https://www.nytimes.com/2023/03/13/business/svb-collapse-com...

"Roku, the maker of the streaming media player, said in a U.S. Securities and Exchange Commission filing on Friday that roughly $487 million, or 26 percent, of its $1.9 billion in cash was tied up with Silicon Valley Bank"

(this was not very clever of them! But it seems to have been encouraged among the SV community?)

> > Companies with large amounts of cash keep very little of that in bank accounts, it's in money market funds and other cash-equivalents.

Companies above a certain treshold (and all citizens for that matter) should be able to bank with the Fed, and have peace of mind that the money they deposit there is not being put to work in any way, shape or form.

What developed over the last week is a victory for those who long for CBDCs .

Companies can already buy recurring 3 month treasuries to hold their cash.

> Companies above a certain treshold (and all citizens for that matter) should be able to bank with the Fed, and have peace of mind that the money they deposit there is not being put to work in any way, shape or form.

Banks being able to lend out money is generally considered a good thing though.

There certainly are alternatives though for depositors that want to have more control in how their money is put to work: money market accounts, short term treasuries, etc

> > Banks being able to lend out money is generally considered a good thing though.

Banks are lending not their own money but depositors.

In a perfect world every citizen , small LLC and startup would be able to issue their own bonds if they need credit for purchases or entrepreneurial efforts

> Companies with large amounts of cash keep very little of that in bank accounts, it's in money market funds and other cash-equivalents.

Could banks learn something from companies here, and drop the volatile long term bonds from their assets in favor of money market funds?

> FDIC will probably have to be reformed. You cannot guarantee everything, all the time.

Why can't you when you can literally print money? Just banks will have to be forced to pick up the tab for that, by limiting their money printing ability so that in case of emergency it can be safely printed.

Dollar deposits in a bank should be 100% safe operation regardless of the amount.

And investors can burn. Every investor is reaponsible for their own risk.

Ever since the initial Covid shock wore off, this whole thing just feels like 2008 in slow motion.

2008 started back in 2006 and wasn't done until 2009

> FDIC will probably have to be reformed. You cannot guarantee everything, all the time.

FDIC does not guarantee everything all the time, that's the whole reason for what happened at these banks being controversial.

Yellen folded? What does this really mean? The “market” will test her how? The market is only interested in pnl. Insolvent banks are a great place to make a trade. Hardline central bank policy (Japan) would be a place to make a trade. Treasury secretary and deposit insurance? I don’t know what you see there.

"They failed specifically because of business they were servicing."

I don't think so. All banks are facing major markdown issues right now due to the Fed raising rates. SVB and the crypto banks were just canaries in the coal mine.

First Republic also needed a bailout even though they're a lot more diversified in their client base.

Edit: before you downvote, please read this analysis by our friend patio11 - https://www.bitsaboutmoney.com/archive/banking-in-very-uncer...

TLDR: The U.S. banking system (as a whole) has $620 billion in unrealized losses due to the Fed raising rates. This is way bigger than SVB and crypto.

Right. So the next important question(s) is:

Why didn't The Fed anticipate the downside of their actions?

Follow up by: Isn't The Fed supposed to prevent bank failures, not cause them?

Mind you, there are other (loop) holes that enable such things. Nonetheless, are these random-y bank failure or failures of The Fed and its associated regulatory friends?

So the Fed is supposed to keep its interest rates at 0 forever just because some banks based their business decisions on the assumption that interest rates will stay at 0 forever?

_Everybody_ in banking is supposed to be aware that if the interest rate goes up the spot or mark-to-market value of a bond goes down. This is not something complicated like Black-Scholes.

Everybody in banking is able to see the inflation prints and aware that increasing inflation may cause the central bank to raise rates. A bunch of people failed to spot this and have since been sacked and/or lost their equity in SVB.

They didn't fail to spot it. They just didn't anticipate them to be so drastic in such a short time span. It's barely a week ago Jerome Powell said their own prognoses were still wrong.

In other words, white collar "criminal-esque activity" gets a free pass.

In addition to that piece, Matt Levine also did an excellent analysis https://www.bloomberg.com/opinion/articles/2023-03-14/svb-to...

…due to the Fed raising rates

Translation: due to the banks not being sufficiently stress-tested to assure resiliency under reasonable rate fluctuations.

>unrealized losses due to the Fed raising rates.

so they expected rates to stay at near zero indefinitely? It's obvious the banks did a bad job on their analysis. It seems like everyone but these banks knew the spigot of easy money was going to stop sooner or later. It just wasn't sustainable.

It's psychologically difficult to do nothing when faced with inflation and near zero interest rates. The banks should have just sat on their hands and made peace with losing money for a year or two. It's certainly preferable to buying bonds that plummet in value the moment interest rates go up. But that's what they did, because when faced with inflation it feels like you have to do something, and this is something, so they did it.

People have been saying rates must go up for a long time but it took 10 years. That’s long enough for a bank to go out of business if they try to wait it out.

All banks have markdown issues, but SVB had liquidity issues due to the businesses they were serving, i.e. they served cash burning machines so naturally hit liquidity issues. The other banks, like First Republic, also have issues but might not have had a problem if not for the loss of confidence caused by SVB's collapse.

> They failed specifically because of business they were servicing.

This is so disingenuous. They failed because they made risky investments, interest rate hikes put their investments in jeopardy, and when they tried to raise extra capital (responsible thing to do), people were alerted of the problem and left the bank (responsible thing to do).

IMHO it’s both. Although the risky investing probably played a bigger part.

I used to think that it was entirely about the risky investments, but $42 billion left SVB (which managed about $200 billion in total) in 2 days. Literally no bank today could withstand a bank run for 20% of its asset base.

Per their mid quarter update[1] there was ~20bn outflowing per quarter, so the question I have is if they had to sell their MBS for 20bn at a 2bn loss, that seems it would only buy them an extra quarter or so of liquidity. Surely they couldn't do that every quarter and remain solvent? Did this trigger their insolvency or accelerate it?

[1]: https://s201.q4cdn.com/589201576/files/doc_downloads/2023/03...

Bank runs are the rational thing to do in the face of an insolvent bank whose deposits are decreasing. SVB mainly serviced startups who would need to withdraw deposits in the current macro climate so there is a good chance they would eventually run out money. As soon as people realize that it's about the correct game theory choice to pull the money and run.

It's part of the story. They had a non-diverse group of interconnected customers prone to herd behaviours led by a small influencer elite. Few banks represent such a high risk of a run.

These guys only started all caps tweeting after the bank was closed on Friday. The bank run started on Thursday when $42B (20-30% of all deposits) were withdrawn.

Yet their sensationalistic tweets probably caused bank runs on other banks, which is the article’s argument.

The run they were tweeting about was a wider one on all regional banks. Did not happen at time of their tweets, they were speculating about the future.

Why was it the VCs impacted by the SVB collapse, and not the wider banking world, who stood to lose more, who were loudest to make that claim?

There are numerous accounts of 200+ group chats between SV CEOs and VCs all deciding to withdraw their money too, FWIW.

This weekend, J. was Tweeting about:

1. firearms

2. people needing to be terrified

3. bank runs

If you're of the opinion that their Tweets (ignoring what may or may not have happened in certain private chats leading up to) did not lead to the public start of SVB's bank run, maybe? How much pressure did this add on regulators to act NOW to avert an IMMEDIATE and CATASTROPHIC COLLAPSE?

However, is it really a stretch to say that those Tweets may have downstream effects on other banks?

SVB failed because of an interest rate bet gone bad. If they invested in short duration treasuries instead of long duration, none of this would have happened. The main difference is their bonuses would have been lower the past few years.

They couldn’t afford to keep treating founders and VCs like personal royalty[0]: the yields were too low. But they feared what would happen if they stopped. So they made a bad bet to cover the extra expense.

[0] Video of Jason Calacanis expounding on his love of royal treatment by SVB: https://twitter.com/one4thecashbag/status/163533710637676953...

There is always no panic until the panic starts!

Even if the worst case scenario was only losing 10% of deposits, that's a very good reason to remove your money.

But in reality, everyone knows this will trigger a bank run, which will lose much more, so the only thing that matters is to get out ahead of it.

This is just the inherent logic of the situation, not the fault of whoever realized it first.

"This will cause a bank run, so you gotta a make a run on the bank to get your money first."

I mean, I kind of get it. That first 10% could very likely be people who know something you don't. So you use it as an indicator. But then the very perception of instability causes that same instability.

Jason Calacanis had a personal interest in SVB

https://twitter.com/one4thecashbag/status/163533710637676953...

> there will be more. Once the fire is lit, it's hard to stop it)

For now bonds yields have come down significantly so most banks in a similar situation should be in a much safer position until/unless yields start going back up again.

A bank should keep the money secure. Anything >0% loss is unacceptable. They were greedy and had a duration mismatch. Interest rates rise and the value of the longterm bonds go down. The regulators should have stepped in and stopped that. Why didn't they?

No one should be made whole. Looks like the stock holders will have a 0 and bond holders will take a significant cut. Depositors will be safe.

The executives should all be in prison for this. Only have to do this once. The other banks would wisen up and get their act together.

What about the CFOs and wealthy individuals that where just bailed out ?

Shouldn't they were in prison (the CFOs, for wealthy individuals - it's just their money) too ? They were fully aware of the 250k$ limit of FDIC insurance.

I'm not really qualified to comment on these things so correct me if I'm wrong, but I thought the FDIC insurance was a minimum insurance, meaning that everyone regardless of balance is insured up to 250k and will get that 250k (or whatever they had on the accounts less than 250). In the case of the ones with more than 250k, they'd still be made whole once all the bank's assets are sold/redistributed, right?

Not really. They would be paid as much as possible from selling assets. That does not imply they would be paid everything. They would probably lost some money over that limit.

Dumb question but is just keeping customers funds without investing them (even in bonds) and just collecting fees not a viable business model for banks? Is it a necessity or greed?

In that it's theoretically possible of course.

The problem is inflation and the lack of interest rates means that just holding cash is a losing proposition (inflation will eat away at it).

With inflation running at 9%ish percent if you hold 20K for 10 years at 9% inflation, it's real worth at the end will be ~8.5K.

That risk belongs to the depositor though.

No, the banks goal is to make money, if their assets keep devaluing it is also their problem.

This idea is fairly popular with libertarians and people have tried to set up so-called narrow banks that do this, but the Fed blocked them from doing so because they consider it to be too important that banks invest their deposits and make loans. That is, the non-existence of the kind of bank you're talking about is the direct result of government intervention, and in fact pretty much all this crisis is (the bank failures were caused by the Fed's interest rate policies, banking regulations pushed banks to focus on default risk over duration risk, etc).

The only sources I can find about the Fed taking any kind of action against narrow or full reserve banks is on low-credibility websites with ideological agenda.

Can you describe what the Fed blocking action was and where I can read about it? I'm tempted to conclude that it doesn't exist without additional evidence.

I'm pretty sure most of the credible, non-fringe discussion of this was on paywalled financial news websites unfortunately.

What, like FT and Bloomberg? I didn't see anything like that.

The Fed literally banned a full reserve bank last month: https://www.yahoo.com/lifestyle/crypto-bank-custodia-taking-...

Also there was a narrow bank proposal the Fed banned in 2019: https://www.bloomberg.com/opinion/articles/2019-03-08/the-fe...

The Fed and other regulators greatly fear banks that do not justify their overreach (and also make existing banks look bad)

Thanks. The Yahoo article makes it seem more like the Fed was opposed to the cryptocurrency aspect than the narrow banking aspect. But the Matt Levine writeup is more apt.

The Fed always give all kinds of excuses but the real problem is that they are afraid that banks that can demonstrate that they are run-proof without government aid might outcompete government-backed banks.

Or, maybe, just maybe, you form healthy economies via the movement of money, and not by letting it sit in a box. You think that may be it since it follows standard, non-politically charged, economic theory?

This is very alarming. I want a place I can store dollars in gold equivalents and redeem them for dollars at will. I don't want new loans to be derived from my deposits. This is hardly an extremist position.

Well, then you expose yourself to gold price risk. Until the fed loosens their position on narrow banks or directly provides deposit accounts to consumers, this won't happen.

However, thanks to the yield curve being inverted, you can approximate this and make decent interest by rolling 4-week T Bills and putting all of your purchases on a credit card (effectively giving you net 30 payment terms).

> dollars in gold equivalents and redeem them for dollars at will

These are not the same thing and I would expect people on a tech website to know that it's not 1971 any more.

Simple solution: buy gold and store it.

100k$ of gold as 100g bars + their certificates of analysis/authenticity fits in a small lunch box. Bigger bars allow greater storage density, but are harder to offload.

The problem is you now are exposed to fluctuations in the gold market, and offloading gold in any sizeable amount is a pain in the arse.

Realistically, is it hard to sell a lunchbox worth of gold for its fair market value?

Depends on where you live. Generally, the answer is no. Gold is easy to move in pretty much any quantity. Local coin shops want gold they can move easily. In some areas that is gold in any form. In others sovereign coins are more popular than bars.

> This idea is fairly popular with libertarians

Which is ironic because it entirely relies on the Fed paying interest on reserves, which they never did until 2008 when they started as part of the bank bailouts.

Or it assumes away inflation, which in the dogmatic libertarian view is also a government-created ("archopathic"?) problem.

I can see why the government would do this - doesn't capitalism require a steady supply of capital?

Question is level of these fees. If they were to charge let's say 0.5% on any deposit of any scale it might work.

With current levels of normal account fees, likely less so. There is lot of staff and operational expenses even if these were set at minimum.

Keep the funds in what?

There's about $2trn in banknotes https://www.uscurrency.gov/life-cycle/data/circulation and from my earlier reading of the FDIC statements there's about $24trn in total US bank deposits. So if you want to have everybody keeping their own individually serialled banknotes that's a non-starter.

You can keep it at the central bank, whether that's "postal banking" or "the deposit window" or "a CBDC", but that seems to be politically unpopular for incomprehensible reasons.

> A bank should keep the money secure. Anything >0% loss is unacceptable.

You're wrong about this part at least. Borrowing short and lending long is what banks do (among other things). The alternative is either:

- banks asking for 10+ year time deposits to match your 10+ year mortgages , or

- banks only willing to do mortgages that are < 1 year

The risks can be managed somewhat, and SVB *definitely* were too greedy (and stupid), but you're mistaken if you think the other banks are qualitatively different than SVB in their exposures to interest rate risks...

That's why most bank's stocks are down. Most people don't think they will fail, but recent events do highlight that they have a bunch of long term securities that lost value.

This just looks like the system working. California swiftly pulled the plug and FDIC is sorting it out. Lessons will be learned, buy the dip, move on.

> were extremely specific and were crypto / VCs banks. It's not random banks that failed. They failed specifically because of business they were servicing.

Agree.

Now that the crypto industry in the US have no banks supporting them, perhaps it is time for crypto completely die (and it should) which should reduce the amount of VCs fueling the ponzi scheme.

There are still plenty of banks that serve blockchain and crypto companies: https://cointelegraph.com/news/svb-and-silvergate-are-out-bu...

Most of the scams in crypto probably don't need direct banking services because they aren't running long-term businesses, just quick marketing schemes to pump their token value before cashing out. Unless exchanges world wide are shutdown, this are unlikely to stop.

There are plenty of non-scam blockchain companies that aren't hurting anyone, just building interesting tech. I don't see any reason other than personal bias to want those startups to be debanked since they aren't any worse than the typical startup.

In this case it's the banking system having a crisis (perhaps revealing some Ponzi-like elements). Surely that is a boost to the crypto bros, in the abstract at least.

It is.

Bitcoin went up a fair bit during the SVB/SBNY crisis.

The bank had long term bets if government bonds that after the rate change massively weakened their value, threatening the bank’s capitalisation levels. This has nothing to do with their clients.

This has everything to do with their clients. Do you know any other bank recently which faced 42B $ of withdrawals ?

Thing is, this bank was servicing people basically juicing the system of free money. This cut both ways.

What's the direct relationship between these two. Are you saying if all of those deposits were kept in JPM the same would have happened? Why?

The size of JPM would've made it so a bunch of VC funds abruptly withdrawing all their money would look like a tiny drop in the bucket rather than a bank-destabilizing balance sheet issue.

JPM banks in the trillions; SVB, 200 billion.

Then a better questions is why would they suddenly decide to withdraw their money if the had deposited it at JPM?

The withdrawals happened because the client base was sophisticated enough to realize (based on publicly available information), and interconnected enough to communicate to each other, that the bank was insolvent. But it's not the client base that caused the bank to be insolvent. That was entirely the fault of the bankers.

The bankers did a very poor job here, for sure ! But it was far from insovlent - it was insovlent if everybody decided to pull out their funds at once - which is what happened. Not a single bank could withstand that.

But so did the client base - we're not talking about small depositors here - if I read the documents correctly, Circle had >8B $ in SVB and they moved 5B $ before the week-end. It was a handful of wealthy individuals that organized this bank run.

> it was insovlent if everybody decided to pull out their funds at once

Which is the definition of insolvent.

A normal bank just goes to the Fed as the lender of last resort but, apparently, they won’t lend money to insolvent banks so here we are.

> Not a single bank could withstand that.

Wrong. Banks whose assets are worth more than liabilities can get unlimited low interest loans from the fed.

It depends.

If Company A wants a bank account with x% interest and instant access to all their deposits at any time (which might be impossible to provide)

and Company B offers them such a bank account, wagering that they won't use that instant access at the same time as everyone else

who is at fault?

Does it matter ?

They are companies - they just go to court with that. They are or should be prepared to deal with this stuff.

What I'm saying is that this bank run has nothing to do with small depositors withdrawing their money here.

Gotta keep those 2020-2022 vintages on a slow decline so they can raise the next fund.

SVB was the only bank (or maybe one of the only banks) that allowed anyone to set up a bank account with them without visiting the country - they were a part of Stripe Atlas.

That sounds great in a libertarian, everything goes world. But the reality is that there are plenty of bad actors and they will abuse it to launder money and commit fraud. Every other bank was conservative enough to avoid this by enforcing some basic in-person kyc/aml processes.

But not SVB. If you were banking with them, you should have asked yourself at least once: why is my bank doing something no other bank does? And is my money as safe there as it would be with a more conservative bank?

Why take a random small program being run at the bank, and imply that it was somehow linked to their failure, when their failure is well understood and has nothing to do with the Atlas program?

If your bank is willing to take on excess risk in one area, it's reasonable to assume that it's also willing to take on excess risk in other areas.

The very fact that it was catering to startups was high risk behavior.

And let's not pretend that extending loans to 15 year old "startups" that have never made a dime in profits and have no clear exit path was a sound business.

Their problem wasn't customers defaulting on loans, wasn't criminal money laundering, wasn't crypto meltdowns.

Their depositors were simply more flighty than average.

Who is pretending what about some made up thing that didn’t happen?

The failure of SVB had nothing to do with extending loans to their client base.

The client base WAS a risk factor.

It's business 101 that you need a diversified customer base. If you're too heavily concentrated in one industry, you're entirely exposed to any downturn in that industry.

In this case, things were made even worse since SVB's customers were VC-funded startups, and VCs have an outsized say in the decisions made by VCs. Particularly in 2023 when we've seen startups shed more and more equity to VCs.

If I was running a business similar to SVB, I would have asked myself if I was too over-exposed to VC-funded startups, and if those VCs were always going to play nice.

> why is my bank doing something no other bank does?

Well, that's an easy answer in techbro circles: because it's the smartest one, duuh.

> SVB was the only bank (or maybe one of the only banks) that allowed anyone to set up a bank account with them without visiting the country

There are plenty of others, Ally being the next off the top of my head. This doesn’t, at all, free you from AML or KYC.

I wonder if this behavior will lead banks to be even less likely to take on startups as clients in the future.

edit: in an attempt to be more clear, I thought the reason SVB was used by startups was because "regular" banks were not open to startup finances. Is this not the case?

Reply on news.ycombinator.com