13 ms·
Silicon Valley Bank Failure [pdf]
- belter 4y ago"The liabiity issue: extreme reliance on institutional/VC funding rather than traditional retail deposits While capital, wholesale funding and loan to deposit ratios improved for many US banks since 2008, there are exceptions. As shown in the first chart, SIVB was in a league of its own: a high level of loans plus securities as a percentage of deposits, and very low reliance on stickier retail deposits as a share of total deposits. Bottom line: SIVB carved out a distinct and riskier niche than other banks, setting itself up for large potential capital shortfalls in case of rising interest rates, deposit outflows and forced asset sales..."
- woeirua 4y agoOn the one hand this does provide some clarity. On the other hand it also reeks of “this is why this could never happen at JPM.”
- runeks 4y agoThat's actually quite concerning, if you read between the lines. What they're saying is: "JPM is just as insolvent as SIVB. The only difference is that JPM's customers are less likely to withdraw their funds."
- jcalvinowens 4y agoNo, what they're saying is a far greater portion of their deposits are from depositors below $250K who have no rational motivation to participate in a bank run.
- belter 4y agoYou got it :-)
- tempsy 4y agothe irony of this whole situation is VCs and startups pouncing on the chaos to encourage people to move their money into even more opaque neobanks eg Mercury/Brex/Ramp as if they don’t have the same issues with relying on VC funded startup deposits but even worse in that their balance sheets are hidden.
- Eumenes 4y agoAgreed that these neobanks aren't a wise choice, but I still think the VCs made the right call in advising their portfolio companies - they had a literal vested interest in their success, they had to do something, this eventually was going to happen.
- mrkurt 4y agoWe use Mercury, they're a frontend though. They're not a bank. Our checking/savings are with Choice Financial, and swept into at least four of these banks: https://co-mercury-prod.s3.amazonaws.com/legal/Choice+-+Deposit+Sweep+Bank+List+-+03_09_2023.pdf https://co-mercury-prod.s3.amazonaws.com/legal/Choice+-+Depo... We use their Treasury account type for most of our cash. It's split between Morgan Stanley and Vanguard funds. Now, if Mercury fails it's going to be a pain in the ass to get at the money in these backer accounts. So we keep what we need for 2 months of operations with another bank. We were using SVB for this, now we have to find a new emergency backup bank. Brex and Ramp are similar to Mercury.
- metalspot 4y agoalso use mercury for the same reason. they are a tech company that provides a good interface to baking services at a wholesale price, which is very different than trying to be a bank. for actual banking i see know reason to use anyone other than jpm chase.
- tempsy 4y agothey work with tiny regional banks that act as a white label service if one or more of those tiny regional banks fail in the same way SVB did I don’t see why there wouldn’t be an issue as a customer of Mercury it’s also unknown what percentage of these banks’ total deposits are Mercury deposits now. if it’s a lot then you have the same risk of a bank run on the Mercury side disproportionately impacting the solvency of the underlying bank
- cm2187 4y agoI am surprised they show JPM in all their comparison charts (typically research doesn't cover their own employer). By showing JPM as an outlier on the opposite of the spectrum to SVB, it feels a little bit like a marketing document.
- notyourwork 4y agoJPM is a massive bank, they don’t need to market.
- wojcikstefan 4y agoCoca Cola & Red Bull are massive, they surely have tiny marketing spend… Right?
- notyourwork 4y agoYou’re missing the point. This isn’t a marketing piece. JPM doesn’t need to compare themselves to SVB for any reason, that’s like comparing David and Goliath. What’s purpose would that accomplish?
- Waterluvian 4y agoThey’re doing what every bank has been doing: reassuring everyone that they’re not at risk.
- HPsquared 4y agoTo reassure their creditors.
- htag 4y agoThis is a marketing piece the same way an engineering blog post is a marketing piece. It demonstrates expertise and helps influence others. I can think of two audience members: * Those with cash in SVB, or another regional bank, and want to find a 'safer' bank. * The relatively uniformed that are nervous about banking.
- nonethewiser 4y agoI think we’re too accustomed to startups here to recognize that SVB was actually assuming quite a bit of risk. We acknowledge most banks don’t want to touch startups and that startups will have a harder time banking in the future. Yet I don’t see much consideration for the fact that there is a good reason most banks see startups as risky. It’s just explained away as “they don’t understand .” Also consider the past 10 years have probably been the friendliest 10 years to startups ever.
- bigbillheck 4y agoWasn't the risk 'betting hard that interest rates wouldn't rise for a decade' in 2021?
- senttoschool 4y agoI don’t think startups will have a harder time banking in the future. This isn’t even the fault of startups. It’s a complete risk management mistake on the side of the bank. Buying 10 year low yield securities and not hedging them against rising rates. Plenty of banks would love to have the deposits of startups and VCs. I bet a bank like Mercury or some other ones will grow to take SVB’s place.
- awill 4y agoThis is exactly right. This has nothing to do with the startups using the bank
- JumpCrisscross 4y ago> has nothing to do with the startups using the bank The proximate cause is short-term funding from risky depositors, i.e. start-ups. The ultimate cause was insolvency. (Yes, not illiquidity, simple illiquidity is solved by the Fed’s discount window.)
- metalspot 4y agono. the problem with long term AAA securities was only 100B of their 200B book. the dead loans to startups are a much bigger problem and why nobody would buy them.
- rvz 4y ago> The liabiity issue: extreme reliance on institutional/VC funding rather than traditional retail deposits Unsurprising how the bank failed due to the VC driven mania and now that the pyramid scheme collapsed right in front of everyone; taking the majority of startups in the tech industry with it. Now we will see how unprofitable these startups really are and have been fully dependent of constant VC cash. A very big lesson to learn about risk.
- workOrNah 4y agoELI5?
- kfarr 4y agoSVB had an unusually risky customer profile while taking (debatably) risky bets with deposits from said customers.
- ericpauley 4y agoThis really sheds clarity on the situation. SVB was in bad shape long before the run, and there is no apparent next domino to fall. FDIC limits are very well understood and relatively easy to work with (despite the rampant FUD about “who’s going to use multiple bank accounts”, deposit sweep programs are highly available and convenient). This is a risk management failure by depositors (in addition to the bank of course), and should be treated as such.
- siftrics 4y agoThis is not risk management failure by depositors. Depositors can and should assume that regulations prevent banks from assuming outsize risk like this. This is a policy failure of the regulators that oversee banks. Banks should not be allowed to have so little cash on hand, especially when we knew with high likelihood the fed would raise rates.
- JumpCrisscross 4y ago> Depositors can and should assume that regulations prevent banks from assuming outsize risk like this Small depositors, yes. Institutional depositors, no. Not all banks are equal. SVB was borderline investment grade before it collapsed. Treasury advice strikes me as low-hanging fruit VCs could have guided their companies on. Instead, most universally recommended SVB because the priority was reducing friction, not risk.
- ericpauley 4y agoExactly. 250k to me seems like a very logical threshold to expect at least some sophistication. Deposit sweeps get you to 3M many places after which it seems perfectly reasonable to expect people to manage treasuries. If Bogleheads can do it VCs can. Then again, you have people like Mark Cuban who clearly don’t know about basic cash management (https://twitter.com/mcuban/status/1634413306948603905 https://twitter.com/mcuban/status/1634413306948603905), so maybe American lack of financial literacy has truly trickled all the way up.
- moremetadata 4y ago[dead]
- 6d6b73 4y agoDoes any one knows if VCs have contracts with startups where they have to deposit X amount weekly or monthly? Now if they can't because of the SVB debacle, can the startups sue them? This would put these VCs in even worse situation - not only they could be out of their money deposited at the bank but now they owe even more money to the startups.
- egl2021 4y agoThe VCs don't "owe" money to the startups: they "buy" equity with their money. The VCs aren't "out" the money deposited by the startup at SVB; that money was already exchanged for equity in the startup. The VCs aren't happy because they and the startup both expected that the money-equity exchange meant that the startup would have working capital, so potentially the value of the equity that they got has fallen. This is a problem for both parties.
- 6d6b73 4y agoYes they buy equity but do they always pay upfront? I would imagine that they have some contracts structured in a way that allows them to pay in rates based on performance etc. Also, what if some vc signed contract last week with payment due this week and they can't pay. I dont expect a lot of cases like that but i suspect few did.
- notmindthegap 4y agoMaybe a stupid question: if banks can collapse from a bank run, shouldn’t the entire model be questioned? A bank run is simply when a threshold number of customers decide to withdraw their cash, with every right to do so. With social media + frictionless mobile banking, the entire notion of teetering your model on mitigating the risk of a “bank run” seems anti-customer, regressive, and unsustainable.
- yCombLinks 4y agoMost depositor's money is insured by the government, so there is no reason people would panic withdrawal their money
- LandR 4y agoIn the UK, you are only covered up to £80k though... I could understand people wanting to get at least money over £80k out, but also how long does it take to get access to your cash if you have to go through the government insurance procedure. Is it days, weeks, months ? I have no idea and wouldn't want to have to find out.
- Gare 4y agoI was suprised to learn that in the US deposits are insured up to $250K since 2008. In the EU it's only 100K euros and it seems low.
- compiler-guy 4y agoIn the US, typically the bank is closed Friday afternoon and depositors have access to their insured money Monday morning. In this case, depositors lost access to their money in the middle of Friday morning and will have access to their insurances deposits Monday morning. So a very brief outage in the typical case. And about five or so extra hours in the SVB case. An FDIC takeover is efficient and well oiled.
- zhte415 4y ago> In the UK, you are only covered up to £80k though... Per bank.
- Scubabear68 4y agoAn important stand out quote to me here: “ It’s fair to ask about the underwriting discipline of VC firms that put most of their liquidity in a single bank with this kind of risk profile“. I really don’t understand why these firms didn’t use at least two banks for their deposits. Surely these tech firms have heard of single points of failure being problematic?
- mixdup 4y agoIn other articles I've read where some of these VCs had "levers" into the bank where they could keep track of what their companies were doing. Not sure what exactly that means, or if it was legal, but it makes sense.
- mrkurt 4y agoIf you get any kind of loan from SVB, you're required to keep your cash with SVB. I think it was entirely reasonable for Series B and earlier startups to keep all their money in SVB. It was wrong, in hindsight, but reasonable. Bank failure is not the thing that's going to kill most startups. SVB just failed spectacularly, and it sure seems like it's not going to put anyone out of business.
- Scubabear68 4y ago“If you get any kind of loan from SVB, you’re required to keep your cash with SVB”. Really? How would this be enforceable or even discoverable?
- s1artibartfast 4y agoDiscovery might be difficult but enforcement is pretty simple. If you have a loan and break the terms of service, they can call your loan due. It's kind of like the terms of service on a home mortgage loan.
- woeirua 4y agoThis whole incident should disabuse anyone from believing that VC money is “smart money”.
- didip 4y agoIt is important for banks to always have a balanced portfolio and regulations to force them to have it. Otherwise, it’s weakness will soon be recognized by some billionaires, Thiel in this case, and they will take advantage of this weakness to become even richer.
- dboreham 4y agoOn reflection, one wonders why all bank deposits don't have insurance. I'm guessing the answer is: something something make more profit... E.g. my businesses are required to carry liability insurance. I have to do that because we have big company customers who made it a condition of doing business with them. So why do big companies hand $nB over to another company for safe keeping but not require insurance?
- mixdup 4y agoThe only truly viable deposit insurance is from the government, because they can print money What good is private insurance from an insurance company if they go insolvent because of massive numbers of bank failures So, now the answer to your question is a political one: are we willing as a society and government to potentially put that much taxpayer money at risk. I think it's a perfectly valid discussion to have, but it's one we haven't had yet, and one we refuse to have until the shit has hit the fan
- schrodinger 4y agoIsn't it up to the depositor in a way? Normal people are 100% protected up to $250k, and then large investors can choose between boring, low interest, extremely safe banks vs high interest risky banks. Although as the analysis pointed out, the higher interest was nominal in this situation so it was a poor choice.
- chewz 4y agoThis is rather silly explanation of what happend, especially from JP Morgan... Everyone who have ever managed bond portfolio knows that he must hedge interest rate risk. And every bank is doing that. SVB didn't. Since April 2022 till January 2023 SVB had vacant position of Credit Risk Officer.. And the explanation is simple - SVB's former head of risk, Laura Izurieta had left after 1Q2022 when looses from bond portfolio started to grow. She has probably already realized the final outcome as this is ABC of risk management (and in April 2022 the path of rate increases had been already set in motion by FED). Now look at the timing of insiders selling shares of SVB...
- trailrunner46 4y ago100% agree, taking huge unhedged duration risk and then loading up on negatively convex MBS again unhedged is so crazy.
- runeks 4y ago> Everyone who have ever managed bond portfolio knows that he must hedge interest rate risk. And every bank is doing that. How are other banks hedging interest rate risk? And how is the opposite end of this hedge hedging their position?
- m3kw9 4y agoInterest rate swaps maybe. Svb would be on the side that is getting a floating rate payment, while they pay a fixed one to the other side, so when the rates increase it would offset the loss in bond price. Thing is I’m not sure if they need to hedge if the bonds are for 1 year because they will get the money back anyways
- JumpCrisscross 4y ago> how is the opposite end of this hedge hedging their position? Plenty of financial functions create natural short interest rate exposure. Like SVB, they have no business speculating on rates, so they hedge it away. (The Fed is also involved in the repo market.)
- oarabbus_ 4y ago
- zhiQ 4y ago“It’s fair to ask about the underwriting discipline of VC firms that put most of their liquidity in a single bank with this kind of risk profile.”
- trailrunner46 4y agoThe chart titled “Impact of unrealized securities losses on capital ratios” really shows just how inadequate the tier 1 capital ratio is (what regulators use). Ignoring the impact of unrealized losses in assets marked as held to maturity is crazy. Seems like a regulator problem to me, no bank taking deposits should be able to make high duration and negatively convex (from high MBS holdings) without hedges.
- molsongolden 4y agoThe other thread was wild and the conclusions in this memo disagree with many of the assertions that commenters have been throwing around–accusing depositors of recklessly banking at SVB to chase higher yields and accusing SVB of playing fast and loose with risky investments. JPM says neither of these were the case: > “At the end of 2022, SIVB only offered 0.60% more on deposits than its peers as compensation for the risks illustrated below; in 2021 this premium was 0.04%.” > “The irony of SIVB is that most banks have historically failed due to credit risk issues. This is the first major one I recall where the primary issue was a duration mismatch between high quality assets and deposit liabilities.”
- ummonk 4y agoGoing by stock market losses, some of these charts, and Twitter sentiment, looks like SBNY, WAL, and FRC are next on the chopping block for bank runs.
- jeffbee 4y agoI don't see the same red flags in FRB's filings that were present in SVB's filings. SVB had almost half their assets in held-to-maturity securities. FRB has 12%. SVB had triple the short-term credit from FHLBs compared to FRB.
- ummonk 4y agoIt doesn't have the HTM red flags around unrealized gains but it has similar red flags about a panic-prone undiversified and uninsured depositor base. Solvency is irrelevant when depositors panic and start a run on the bank anyway.
- jeffbee 4y agoI was not able to find any filings indicating what fraction of FRB's depositors and deposits are or are not insured. Do you happen to know where to find those?
- poof131 4y agoWhat I find most disingenuous in this whole saga is the conflating of small business payroll depositors with all depositors. Circle & USDC rely on the interest rate earned on the stablecoin deposits for their business and SVB was providing that with poor risk management. With a $3B deposit (or more since they likely moved money out and partially caused the collapse), Circle should have been doing additional risk management beyond SVB, not just collecting interest. Now taxpayers are supposed to cover that failure?
- nullc 4y agoSmart guy meme: "The portion of our stablecoin reserves that don't exist can't be lost in a bank failure!"
- woeirua 4y agoGood analysis. Everyone should be very worried about these charts. In short, a lot of banks are sitting on assets that have significant unrealized losses, which is very similar to the situation leading up to 2008. If there is some event down the road that forces the banks to dip into their HTM assets to cover withdrawals or losses, then we could be looking at yet another systemic crisis. The cause will be different. But the banks are in a precarious position. Perhaps more importantly, this puts a ceiling on what the Fed can do to fight inflation. If they keep jacking rates the magnitude of the unrealized losses will increase.
- fach 4y agoA child-like regulatory question: Why aren’t HTM portfolios for retail banks frequently marked to market? Wouldn’t this force more accurate accounting in the event a bank needed to sell bonds in a capital crunch?
- notmindthegap 4y agoWhy aren’t most assets marked-to-market, at least in some regular fashion?
- onlyrealcuzzo 4y agoSo you can cheat taxes & take on excess debt?
- hirako2000 4y agoNot an HTM certified, not even a practitioner here. But it's also an accounting question. The way it's done: asset is valued per maturation. It would mudd the water even further I think to have the quarterly reflect the "at current market" value. Things go up and down. What matters is how you close (realised gains or loss), and, at SVB's demise: your cashflow. No cash? Then sell your assets. Creditors/depositors ain't gonna wait. Realised loss? That's your balance going down in true accounting terms now. Accountings get disclosed? Those quarterly release are mandatory for all publicly trading company: that could cause worry among investors, and in the case of a bank a panic a bank run. What a market valuation made at each quarter could achieve is give a momentarily evaluation of what a fund's assets is made of if they were to liquidate right then, or that day. But it isn't how it works* It does work for goods though, amortization is accounted for and a well oiled exercise. Futures, bonds, stocks? Good luck with that. If we could predict the future the game would be a whole lot different.
- One4QualityNewS 4y agoNot quite sure how much to look into but the operating of the UK side seems flawed. One Director left in Jan, and the particular Director has previously had 35 businesses all of which were lending companies and paid significant dividends. There seems to be a trend. I dont think this is poor risk management, I think this was very intentional.