How did SVB collapse, and what is the fallout? : Planet Money Silicon Valley Bank was the 16th largest bank in America, the bank of choice for tech startups and big-name venture capitalists. Then, in the span of just a few days, it collapsed. Whispers that SVB might be in trouble spread like wildfire through group texts and Twitter posts. Depositors raced to empty their accounts, withdrawing $42 billion in a single day. Last Friday, after regulators declared that SVB had failed, the FDIC seized the bank.

As the dust settles on the biggest bank failure — and bank rescue — in recent memory, we're still figuring out what happened. But poor investment choices, weak regulation, and customer panic all played their parts. We'll look into the bank's collapse to understand what it can teach us about the business of banking itself.

This episode was produced by Willa Rubin, with help from Dave Blanchard. It was edited by Keith Romer, and engineered by Brian Jarboe. Fact-checking by Sierra Juarez. Our acting executive producer is Jess Jiang.

Music: "
I Don't Do Gossip," "Groovy Little Penguins" and "Vision."

Help support Planet Money and get bonus episodes by subscribing to Planet Money+ in
Apple Podcasts or at

Silicon Valley Bank's collapse and rescue

  • Download
  • <iframe src="" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript




It has been a week for the banking sector, for all of us, really. The core of the action happened last week out in California. Mid-week, the tight-knit startup scene in Silicon Valley got wind that their preferred bank, appropriately named Silicon Valley Bank, might be in trouble.


It was unclear if it was really in trouble, but it didn't matter. As soon as some prominent startup folks told other prominent startup folks they were pulling their money out of the bank, quickly became a bank run - $42 billion withdrawn in a single day.

FOUNTAIN: By Friday, regulators were like, this bank has failed. The government seized the bank and said, we'll take it from here.

GUO: And over the weekend, there was a lot of hand-wringing about this. What was going to happen to all the people who hadn't managed to get their money out in time? Should the government bail out the bank? Was this going to be, like, the beginning of a brand-new 2008-style financial crisis?

FOUNTAIN: Sunday night, the government made it clear. They did not want this to spread. They seized a second bank, Signature Bank of New York. They also guaranteed that everyone who had their money in Silicon Valley Bank or Signature was going to be paid in full. And they set up generous credit lines to every other bank in the country to strengthen them and help prevent more bank runs.

GUO: To sum up, in the last week, we have seen the biggest bank failure and the most aggressive bank rescue since the global financial crisis.


GUO: Hello, and welcome to PLANET MONEY. I'm Jeff Guo.

FOUNTAIN: And I'm Nick Fountain. Today on the show, we are going to examine the collapse of Silicon Valley Bank in three ways - what was going on inside of the bank, what was going on with its customers, and what was going on with the people who were supposed to be watching over this whole thing.

GUO: And we'll consider one uncomfortable truth - that banks are, by their very nature, unstable, that instability is built right into the business model.


FOUNTAIN: Again, it's early days, and it's a little premature to give the definitive history of exactly why Silicon Valley Bank collapsed last week. But there are a few obvious places we can poke around to get clues. We can look at the bank. We can look at its customers. And we can look at the regulators.

GUO: We're going to start first with the bank itself. What was the horror lurking in its balance sheets? We called up someone who is really, really good at reading balance sheets - Columbia University law professor Kate Judge.

FOUNTAIN: I feel like most weeks of the year, people don't want to talk to the editor of the Journal of Financial Regulation, but some weeks everybody wants to talk to you.

KATE JUDGE: I mean, like, you never really actually want to have your moment, is the particular irony.

FOUNTAIN: We asked Kate, so what was it about Silicon Valley Bank that made it different from other banks? And she said, well, you got to look at the industry they catered to.

JUDGE: Silicon Valley Bank was a regional bank that really specialized in surveying the venture capital community and startup firms that were backed by venture capital.

GUO: And the specialized community that Silicon Valley Bank was serving just happened to be, like, the biggest economic growth thing that happened in the past 10 years.

JUDGE: Exactly.

GUO: A few years back, tech startups were the hotness. Oh, you're going to make a new Wi-Fi-enabled juicer? Sure. Here's a hundred million bucks.

FOUNTAIN: And part of this had to do with how low interest rates were at the time. Safe, regular investments, like bonds, were paying basically nothing. So investors were like, might as well take some big swings, see if we can make some real money.

GUO: And according to the bank, almost half of Silicon Valley startups, from Roblox to Roku, they were taking all that money that investors were throwing at them, and they were depositing it at Silicon Valley Bank.

JUDGE: So for a lot of the startups, they were using Silicon Valley Bank as the parking spot for all the cash that they had coming in.

FOUNTAIN: Yeah, between 2019 and 2021, Silicon Valley Bank's deposits more than tripled, from $60 billion to $190 Billion, which sounds great, right? More money. But it was also kind of a problem for the bank because they now had to figure out what they were going to do with all that money. And that is where they started to get into trouble.

GUO: Here's how a bank works. Two sides to the business - on one side are all of the bank's customers. People like you, me, all those tech companies, were depositing money into the bank. We're basically giving the bank a short-term loan because we can take our money back at any time. The bank turns around and puts our money into long-term investments, basically loans it out for long periods of time.

FOUNTAIN: How do banks know that they can squirrel away all our money in these long-term investments? Like, I'm allowed to pull my money out at any time, right?

JUDGE: You are allowed to pull your money out at any time. So if you think about any bank, the nature of a bank is to be an inherently fragile institution.

GUO: This is the essential problem of being a bank. If everybody were to go to the bank to ask for their money back, the bank wouldn't be able to pay all of us back - not right away - because most of that money is locked away in those longer-term investments. Banking is risky.

FOUNTAIN: And that means that one of a bank's most important jobs is to manage risk, try to game out what would happen if a bunch of its customers wanted to pull out their money all at once or what would happen to its investments if - I don't know - the Fed started raising interest rates all of a sudden.

JUDGE: Interest rate risk is a classic bank risk.

GUO: Silicon Valley Bank seems to have misjudged both sides of its business. It invested a huge chunk of people's money in long-term government bonds and other securities, basically locked up that money in investments that would be really hurt if interest rates went up.

FOUNTAIN: Here's the other related misjudgment. You know who else would be hurt by rising interest rates? The tech startups that made up such a huge part of Silicon Valley Bank's customer base. Remember all those venture capitalists who were throwing gobs of money at tech companies? Higher interest rates meant those investors could now get decent returns in way safer investments. Forget the robot juicer.

GUO: So both sides of Silicon Valley Bank's business were hurt by rising interest rates. And one of the big questions is, why didn't it manage its risk better? The simple answer is that hedging its risk would have meant that they made less money. And they didn't want to make less money.

FOUNTAIN: But still - but still - was Silicon Valley Bank doomed to fail? Maybe not. They conceivably could have weathered the interest rate storm. But obviously, they did not weather that storm.

GUO: Which brings us to the next part of the action - the customers of Silicon Valley Bank and the part they played in this whole drama. Because that thing about banks being inherently fragile institutions? Customers are not unaware of that. And this, Kate says, brings us to the matter of trust.

JUDGE: So long as everybody trusts the bank, it can actually really be stable. You know, one person's making a deposit; somebody else is making a withdrawal.

FOUNTAIN: But if customers, for whatever reason, stop trusting the bank, that is trouble.

JUDGE: The fragility that arises is, once I'm afraid that my bank isn't quite as healthy as I thought or I'm afraid another depositor is going to want all of their money back, and maybe another depositor is going to want all their money back - well, then, now I'm suddenly worried.

FOUNTAIN: And last week, that is exactly what went down with Silicon Valley Bank's customers, but on steroids because a lot of the bank's customers belong to this tight-knit community of tech people, who all know each other.

GUO: So as soon as someone started pulling money out...

JUDGE: Everyone else said, well, if they're taking their money out, I'm not going to stick around and take any chances being left holding a bag that might not have quite enough cash in it.

GUO: Was this the first bank run that was done by, like, group text?

JUDGE: It does seem like it might be the first bank run that was done by group text, but it might be the first, but not the last. I mean, I think one of the lessons that we've learned is that in a high-tech environment, you're no longer going to wait for people to be physically standing outside the door of their bank demanding their money back.

FOUNTAIN: Yeah, they could just do it from their phones. Thursday of last week, customers withdrew $42 billion out of Silicon Valley Bank in a single day.

GUO: Even if a bank is healthy, once a bank run starts, it's already too late to solve the problem. The only way to address the problem of bank runs is to short-circuit them before they start - to stop people from panicking in the first place.

FOUNTAIN: During the Great Depression, the government figured out that the way to do this was by creating deposit insurance. They said, if your bank fails, don't worry. We got you covered. We'll make you whole again. Federal deposit insurance now covers depositors up to $250,000.

GUO: And the idea is if there is an ironclad guarantee that you are going to get your money back no matter what, there's no incentive anymore to rush down to the bank to take your money out at the first sign of trouble. And if all the customers feel that way, then bank runs can't even start.

FOUNTAIN: But in the case of Silicon Valley Bank, there was one teensy-weensy problem. The bank's customers, largely Silicon Valley startups, had parked not just $250,000 in their account. They had parked much, much more, like millions and millions.

GUO: Most of that money was uninsured. So Silicon Valley Bank's customers, they had a lot to lose if the bank went under. That meant that once people heard about this bank run, it was a thousand percent rational to join in, try to get their money out, too.

FOUNTAIN: Now, Kate says there is a reason why Congress decided to set the limit at $250,000 because folks with more money should theoretically be more financially savvy. Surely, these titans of the innovation economy could spot a risky bank and avoid it, right?

JUDGE: There's this core tension where, on the one hand, we like the idea of capping it at $250,000 because we want the larger, more sophisticated depositors to potentially engage in a little bit of monitoring so that if they don't trust their bank...


JUDGE: ...They're moving it away. Exactly. What we worried about with deposit insurance is moral hazard, right? Like, this is the big drawback whenever we have insurance, is, well, nobody's going to care if their bank's stable or not if the government's going to make them whole. And so we say, like, that's perfectly fine. That's the right conclusion to draw for all these small depositors. But these really sophisticated players, like, we don't necessarily want to protect them.

FOUNTAIN: That's the system we have. And on some level, this system is kind of silly, right? Like, every business owner in America, when they're deciding where to put their payroll account, is supposed to be an expert at combing through a bank's balance sheets? Seems unrealistic.

GUO: Regardless, all of that theory goes out the window when a bank actually fails and people actually lose money and fear about the financial system starts to spread.

FOUNTAIN: OK. So you could blame the bank for making poor investment decisions. You could, I guess, say that the depositor should have done more due diligence. But wait a second. Wasn't there supposed to be someone keeping tabs on all this? Like, what about the regulators?


GUO: After the break - would stronger regulation have prevented the collapse of Silicon Valley Bank? We hear from someone who tried to sound the alarm years ago.


GUO: Every once in a while, as a society, we are reminded that banks are, at their core, these fragile, unstable businesses. That is why banking is one of the most tightly regulated industries in America. Our last reminder of this fragility came in 2008, during the global financial crisis. One lesson we learned was that when banks start to fail, it's already too late. Robust regulation needs to be in place already before the trouble starts.

FOUNTAIN: That is what this last chapter is about - the regulations that were supposed to prevent this and whether they were strong enough.

GUO: So after the global financial crisis, Congress passed the Dodd-Frank Act. This established some new, more stringent regulations for banks above a certain size.

SAULE OMAROVA: In the Dodd-Frank Act, Congress said, we will look at large banking institutions above $50 billion in assets.

FOUNTAIN: This is Saule Omarova, law professor at Cornell University who specializes in financial regulation.

OMAROVA: Fifty billion dollars in assets was set as a threshold.

FOUNTAIN: Sure, why not? 50 billion. Sounds like a big number. Seems dangerous.

OMAROVA: That's exactly right.

GUO: Big, systemically important banks would be subject to more supervision - more looking over their shoulder, more reining them in when the bank started making risky moves.

FOUNTAIN: And not too long after Dodd-Frank was passed, banks started to lobby to scale it back. Among other things, they wanted to redefine what was considered a big bank, shift that line from $50 billion in assets to 250 billion.

OMAROVA: There was a general - I'm going to use that word vibe at the time...

FOUNTAIN: Ooh. We love the word vibe. Yes, go on.

OMAROVA: ...That, oh, Congress and the Dodd-Frank Act went too far in stifling innovation by overregulating banks. And look at us now. For some reason, the credit isn't flowing.

FOUNTAIN: One noteworthy proponent for redefining what counted as a big bank was Silicon Valley Bank CEO Greg Becker. His bank was getting right up to the cusp of getting categorized as a big bank and being hit with all those big bank regulations.

GUO: In a letter to the Senate Banking Committee in 2015, Becker told them Silicon Valley Bank didn't need all these new regulations. They were already good at managing risk. If Congress made them follow the new rules, it was just going to hurt all the bank's customers, like, all of Silicon Valley. It's going to hurt the innovation economy itself.

FOUNTAIN: And on the other side of that argument, at a Senate hearing in 2017...


OMAROVA: Senators, thank you for the opportunity to testify...

FOUNTAIN: ...Was Saule. She told senators, banks just want deregulation so that they can make more profits by taking bigger risks with their customers' money. And that is dangerous. Don't let them.


OMAROVA: I urge you to keep focus not on what banks want for the sake of their own profitability, but on what the American economy and the American people need.

FOUNTAIN: Did you have a feeling in that hearing that your argument wasn't taking the day, that these regulations would be rolled back?

OMAROVA: Yes, I did. Oh, absolutely. It was absolutely clear.

GUO: In the end, the bankers got their way. In 2018, lawmakers, both Republicans and Democrats, significantly rolled back Dodd-Frank. Large banks were now defined as those with over $250 billion in assets, up from $50 billion. Worth noting in its latest annual report, Silicon Valley Bank said it had $212 billion in assets. They were still, for regulatory purposes, a medium-sized bank.

FOUNTAIN: And this raises this really interesting question that a lot of really smart people are going to spend the next several months trying to answer. Would Silicon Valley Bank have collapsed had it been subject to the enhanced big-bank regulations?

GUO: Here are some things that would have been different if Silicon Valley Bank had been categorized as a big bank under Dodd-Frank. Regulators would have been more aggressively poking around in its business, might have had to rein in its risk-taking, maybe set more money aside for a crisis like this one. And it would have been subject to annual stress tests from the Fed.

FOUNTAIN: Those stress tests are kind of interesting, actually. They're basically just, like, little banking role-playing games.

OMAROVA: Imagine that there is some kind of a war breaking out in Europe, and there is huge inflation. And there are earthquakes, and the aliens are taking over.


OMAROVA: Things of that kind.

FOUNTAIN: That sounds super stressful.

OMAROVA: I know. That's the point. I'm just joking about the aliens, but you get the idea. So there are different scenarios that are supposed to play out - potential bad things happening all at once.

FOUNTAIN: And these simulations try to figure out, would the banks survive? And if not, how can we make them stronger?

GUO: The Fed actually publishes these scenarios right on their website. In 2022, the Fed tested for massive unemployment, a collapse in asset prices, plummeting GDP.

FOUNTAIN: But it's funny. They actually didn't test for a spike in interest rates, which was the thing that took down Silicon Valley Bank.

GUO: Still, Saule said the point is all of these stronger regulations together would have made Silicon Valley Bank more careful, less likely to take on the kind of risks that could lead to a bank failure.

FOUNTAIN: Saule, I think that you're - what you're trying to say - but you're afraid to say it, but now is your opportunity to say it on national radio - is I told you so.

OMAROVA: Well, I did tell you so.


OMAROVA: I told you so. Everything I've written was precisely about this.

FOUNTAIN: However you want to distribute the blame, what we saw last weekend is what happens when some combination of bankers, customers and regulators fail to keep a crisis from occurring.

GUO: But the story obviously didn't stop there. Once a bank like Silicon Valley Bank gets into as bad a spot as it did, the government has to decide whether it wants to switch roles - from regulator to rescuer.

FOUNTAIN: And our banking regulation expert, Kate Judge, says the government very clearly decided to put on the rescuer cape, at least for the depositors.

JUDGE: The Fed, the Treasury and the FDIC came in, and they said, we are going to make sure that you are made whole. Every single cent that you have as a depositor is going to be fully protected.

FOUNTAIN: All those tech companies with their millions of dollars of uninsured deposits, they were going to get all their money back though, we should say, the investors in the bank, like the bank's shareholders, they're probably going to take a big loss.

GUO: It's funny. Even though Silicon Valley Bank had managed to keep itself from being regulated like a big, systemically important bank, the government ultimately decided, actually, we do need to treat it like a big bank, at least now that it has to be rescued.

FOUNTAIN: And so to keep what had happened at Silicon Valley Bank from happening at banks around the country, the Fed created a new loan program, a really generous program, because there are some other banks who took on similar risks and maybe didn't do a great job at managing those risks either. Now those banks will be able to lean on the Fed and get their hands on some cash if their customers decide they want to pull their money out.

GUO: How much of this is like the Fed expects banks to use these new lending services? And how much of it is just, like, a confidence booster to let everybody in America know, we're going to stand behind the banks, they're going to be fine?

JUDGE: It's potentially both.

FOUNTAIN: Kate says what the Fed is doing goes back to this fundamental truth about the nature of banking. Perceptions are just as important as reality. The reality of Silicon Valley Bank is that it was a bank that, yes, had clearly made some bad choices. But it only really got into trouble when people's perceptions about it started to shift.

GUO: And that fundamental truth about the importance of perceptions, it applies to the entire banking system. That includes the regulators themselves. And maybe that's why they chose to make their rescue plan so extra. The regulators want to maintain this perception that they can keep the crisis from getting any worse so that we'll all go back to not worrying about the banking system and how fragile it really is.


FOUNTAIN: We're going to be covering this story for a long time, I think. If you have any questions about what just happened, what's going to happen in the future, let us know. You can email us at You can also find us on social media and many places @planetmoney.

GUO: This episode was produced by Willa Rubin with help from Dave Blanchard. It was edited by Keith Romer, fact-checked by Sierra Juarez, and engineered by Brian Jarboe. Our acting executive producer is Jess Jiang. I'm Jeff Guo.

FOUNTAIN: And I'm Nick Fountain. This is NPR. Thank you for listening.


Copyright © 2023 NPR. All rights reserved. Visit our website terms of use and permissions pages at for further information.

NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.