Collapse of Silicon Valley Bank

Based on Wikipedia: Collapse of Silicon Valley Bank

On a Thursday morning in March 2023, Silicon Valley Bank looked like any other day at the office. By Friday afternoon, it was gone—seized by regulators in what became the largest bank failure since the 2008 financial crisis and the third-largest in American history. The speed of the collapse stunned even seasoned financial observers. This wasn't a slow-motion disaster playing out over months. It was a bank run compressed into roughly thirty-six hours.

The Bank That Bet on Tech

Silicon Valley Bank wasn't a household name like Chase or Bank of America. Most Americans had never heard of it. But in the world of technology startups and venture capital, SVB was practically an institution.

Founded in 1983 and headquartered in Santa Clara, California—the heart of Silicon Valley—SVB had carved out a remarkable niche. Nearly half of all venture capital-backed healthcare and technology companies in the United States banked there. The client list read like a who's who of tech success stories: Airbnb, Cisco, Fitbit, Pinterest, Block. When entrepreneurs launched their startups, SVB was often their first call.

The bank offered something most traditional banks wouldn't: it would lend money to companies that had no revenue, no profits, and sometimes little more than a promising idea and some venture capital backing. In exchange for taking on this risk, SVB demanded exclusivity. If you borrowed from them, you banked with them. Period.

This arrangement worked beautifully for decades. As startups grew into giants, they stayed loyal. And SVB grew alongside them.

The Pandemic Money Flood

Then came the pandemic. And with it, an avalanche of money.

The COVID-19 crisis accelerated demand for technology in ways no one anticipated. Companies needed software for remote work. Consumers shifted to online shopping. Biotech firms raced to develop vaccines and treatments. Venture capital poured into tech companies at unprecedented rates.

All that money had to go somewhere. For many startups, that somewhere was their account at Silicon Valley Bank.

In just one year—from March 2020 to March 2021—SVB's deposits doubled. They went from sixty-two billion dollars to one hundred twenty-four billion. By the end of 2022, the bank held two hundred nine billion in total assets, making it the sixteenth largest bank in America.

Here's where things got complicated.

The Interest Rate Trap

When a bank receives deposits, it needs to do something with that money. Keeping it all in a vault earning nothing would be financial malpractice. So banks invest their deposits, primarily in loans and bonds.

SVB's leadership looked at their mountain of new deposits and made a fateful decision: they would invest heavily in long-term bonds. These were mostly U.S. Treasury bonds and mortgage-backed securities—about as safe as investments get in terms of credit risk. The government wasn't going to default.

But there's a catch with bonds that many people don't understand. When interest rates go up, bond prices go down. This isn't a matter of opinion or market sentiment. It's math.

Think of it this way. If you buy a bond paying three percent interest, and then the Federal Reserve raises rates so that new bonds pay five percent, nobody wants your old three percent bond anymore—at least not at full price. To sell it, you'd have to offer a discount. The longer the bond has until it matures, the bigger that discount needs to be.

SVB had loaded up on long-term bonds precisely because they paid slightly better interest rates than short-term bonds. In a world where interest rates stayed low forever, this would have been fine.

But interest rates didn't stay low.

The Fed Turns the Screws

By early 2022, inflation in the United States had surged to levels not seen in forty years. The Federal Reserve, whose primary job is keeping prices stable, began raising interest rates aggressively. They raised rates seven times in 2022 alone, with more increases in 2023.

Each rate hike hammered the value of SVB's bond portfolio. By the end of 2022, the bank's bonds had lost more than fifteen billion dollars in value.

Now, here's a crucial detail: SVB hadn't actually lost this money yet. They had classified most of their bonds as "held to maturity," an accounting category that means you don't have to mark them at current market prices. You can keep them on your books at their original value, and as long as you hold them until they mature, you'll eventually get all your money back.

This is perfectly legitimate accounting. It's how most banks handle their bond portfolios. The logic is sound: if you're never going to sell these bonds, why should temporary price fluctuations matter?

The problem is that "never going to sell" assumes you'll never need to sell.

When Deposits Walk Out the Door

Remember those tech companies that had flooded SVB with deposits during the pandemic? By late 2022, they needed that money back.

The venture capital boom had cooled dramatically. Startups that once raised funding easily now found investors cautious. Companies that had planned to grow fast had to tighten their belts instead. And that meant drawing down their bank accounts to cover operating costs.

Meanwhile, the tech industry was shedding workers by the tens of thousands. Laid-off employees withdrew their savings. The deposits that had gushed into SVB began flowing back out.

SVB needed cash to pay these withdrawals. And to get cash, they would have to sell some of those underwater bonds—crystallizing billions in paper losses into very real ones.

The Fateful Announcement

On Wednesday, March 8, 2023, SVB made an announcement that sealed its fate.

The bank revealed it had sold twenty-one billion dollars worth of bonds, locking in a loss of 1.8 billion dollars. To shore up its finances, it had also borrowed fifteen billion dollars and planned to raise another 2.25 billion by selling new shares of stock.

In isolation, these moves might have seemed like prudent crisis management. But the timing was terrible. Just days earlier, Silvergate Bank—a smaller institution that catered to cryptocurrency companies—had announced it was winding down operations. The crypto world was still reeling from the collapse of the FTX exchange. Anxiety about banks was already elevated.

And SVB hadn't lined up buyers for its stock before making the announcement. They were essentially asking the market for a vote of confidence without securing that confidence in advance.

The market voted no.

The Bank Run Begins

Word spread through Silicon Valley's tight-knit network with terrifying speed. By Thursday morning, venture capital investors were telling their portfolio companies to get their money out of SVB. Immediately.

Peter Thiel's Founders Fund—one of the most influential venture capital firms in tech—withdrew all of its money from SVB by Thursday morning and advised every company it invested in to do the same. Union Square Ventures and Coatue Management sent similar warnings to their portfolio companies.

In venture capital circles, these names carry enormous weight. When Peter Thiel's firm says move your money, people move their money.

By the end of Thursday, customers had withdrawn forty-two billion dollars. Forty-two billion. In a single day.

To put that in perspective, SVB's entire deposit base at the end of 2022 was about one hundred seventy-five billion. Customers had pulled out nearly a quarter of the bank's deposits in one business day.

The bank ended Thursday with a negative cash balance of nearly one billion dollars. It was technically already insolvent.

Friday Morning: The End

SVB's stock price, which had been around two hundred seventy dollars per share before the crisis, had been plummeting. Trading was halted Friday morning as the bank scrambled to find a last-minute savior.

JPMorgan Chase and Bank of America—the giants of American banking—looked at acquiring SVB and said no thanks. By midmorning, examiners from the Federal Reserve and the Federal Deposit Insurance Corporation, known as the FDIC, had arrived at SVB's offices.

At 11:30 a.m. Pacific time, the California Department of Financial Protection and Innovation seized the bank, citing "inadequate liquidity and insolvency." Silicon Valley Bank was dead.

Regulators estimated that another hundred billion dollars would have been withdrawn Friday if the bank had stayed open. There simply wasn't enough money in the vault to meet those demands. No bank keeps that much cash on hand. No bank could.

The Deposit Insurance Problem

Here's where the story gets really interesting—and really scary for a lot of people.

The Federal Deposit Insurance Corporation insures bank deposits in the United States, but only up to two hundred fifty thousand dollars per depositor, per bank. This protection was created after the Great Depression to prevent bank runs by assuring ordinary Americans that their savings were safe.

For most people, two hundred fifty thousand dollars is more than enough. But SVB wasn't a bank for most people. It was a bank for tech companies.

A startup with fifty employees needs to make payroll. A medium-sized tech company might have tens of millions of dollars in operating cash. These balances dwarf the FDIC insurance limit.

According to SVB's own filings, about eighty-nine percent of its deposits—roughly one hundred fifty-two billion dollars—exceeded the FDIC insurance limit. These were uninsured deposits. If the bank failed and its assets couldn't cover all the deposits, those depositors could lose money.

Suddenly, companies across America realized they might not be able to pay their employees.

The Weekend of Uncertainty

That weekend was chaos. Thousands of startups—including many with no connection to Silicon Valley beyond their choice of bank—faced an existential crisis. They couldn't access their money. They didn't know when or if they'd get it back.

The implications rippled outward. These weren't just faceless corporations. They employed real people who needed real paychecks. Many founders spent the weekend desperately trying to figure out how to make payroll.

Meanwhile, the FDIC set up a temporary bank to reopen SVB's branches on Monday and give depositors access to their insured funds—up to that two hundred fifty thousand dollar limit. For uninsured deposits, the FDIC would pay dividends as it liquidated SVB's assets, eventually returning perhaps eighty to ninety percent of the money. Eventually. Weeks or months from now.

That wasn't good enough for a company that needed to make payroll on Tuesday.

Contagion Fears

The bigger fear wasn't just SVB. It was what might happen next.

Signature Bank in New York, which had significant exposure to cryptocurrency firms, was wobbling. First Republic Bank in San Francisco was facing a surge of withdrawals. Investors were scrutinizing every regional bank's bond portfolio, looking for hidden losses.

The nightmare scenario—a cascading series of bank runs spreading from one institution to the next—seemed suddenly plausible. It had happened before, in 2008, when the failure of Lehman Brothers triggered a global financial crisis. And it had happened in 1929, when bank panics helped transform a stock market crash into the Great Depression.

Government officials spent the weekend in intense discussions. By Sunday evening, they had reached a decision.

The Government Steps In

On Sunday night, March 12, the Treasury Department, the Federal Reserve, and the FDIC issued a joint statement. All depositors at Silicon Valley Bank would be made whole. Every dollar, not just the insured two hundred fifty thousand. The same protection would apply to Signature Bank, which regulators had also closed that weekend.

The money wouldn't come from taxpayers—at least not directly. It would come from the FDIC's Deposit Insurance Fund, which is funded by assessments on banks. In essence, the banking industry would pay for the failure of its members.

Critically, the government wasn't bailing out SVB's shareholders or management. The bank's stock was worthless. Its executives lost their jobs. This wasn't 2008, when banks were kept alive with government support. SVB was dead and would stay dead. Only its depositors were being rescued.

President Biden emphasized this distinction in remarks the next morning. "No losses will be borne by taxpayers," he said. "The management of these banks will be fired." He explicitly rejected the term "bailout."

Whether this distinction matters is a question economists and politicians continue to debate. The government's willingness to protect uninsured depositors—deposits that were explicitly, legally uninsured—created a precedent. Some argue it was necessary to prevent a broader panic. Others worry it encourages excessive risk-taking, since depositors now expect protection regardless of the rules.

What Went Wrong?

In the aftermath, investigators and analysts picked through the wreckage looking for explanations. Several factors emerged.

First, SVB had concentrated risk in multiple ways. Its depositors were overwhelmingly tech companies and venture capital-backed startups—an industry that experienced a boom and bust together. When the tech sector needed cash, everyone needed it at the same time. The bank also concentrated its assets in long-term bonds without adequately hedging against interest rate risk.

Second, regulatory changes in 2018 had reduced oversight of mid-sized banks like SVB. Under the Dodd-Frank reforms passed after the 2008 crisis, large banks faced stringent requirements including frequent stress tests—simulations of how they'd perform in a crisis. But lobbying by the banking industry, including by SVB's own CEO Greg Becker, led to the Economic Growth, Regulatory Relief, and Consumer Protection Act, which exempted banks under two hundred fifty billion dollars in assets from some of these requirements.

SVB, with about two hundred billion in assets, fell just under this threshold.

Third, internal warnings went unheeded. A 2021 Federal Reserve review had found deficiencies in SVB's risk management. The bank received six citations and was placed under enhanced scrutiny in 2022. Federal Reserve officials met with SVB's senior leaders in the fall to discuss their ability to raise cash in a crisis. They found SVB was using flawed models that assumed rising interest rates would actually help the bank—a dangerously wrong assumption.

Fourth, the bank's risk officer position was vacant during a critical period. The chief risk officer stepped down in April 2022, and a replacement wasn't named until January 2023. During that gap, interest rates were rising rapidly and the bank's vulnerabilities were growing.

The Human Element

Some details from the collapse feel almost surreal.

On the morning of Friday, March 10—hours before regulators seized the bank—SVB employees received their annual bonuses. The payments went out automatically. Management either didn't think to stop them or couldn't in time.

On February 27, less than two weeks before the collapse, CEO Greg Becker sold twelve thousand shares of company stock worth 3.6 million dollars. He did this through a prearranged trading plan filed with the Securities and Exchange Commission, a mechanism designed to allow executives to sell stock without being accused of insider trading. These plans must be filed in advance—Becker filed his on January 26.

Was this coincidence or foreknowledge? The timing raises questions, though prearranged trading plans are common and the January filing date suggests Becker may not have anticipated the March crisis. Regulators examined the transaction but brought no charges.

The Aftershocks

SVB's parent company, SVB Financial Group, found itself in an absurd situation. The holding company was locked out of its own headquarters in Santa Clara because those offices were shared with the bank, which was now under FDIC control. They had to relocate operations to New York.

An even more tangled problem: all of SVB Financial Group's employees had technically been on Silicon Valley Bank's payroll, not the parent company's. The holding company had been providing employee benefits to bank employees. Now the FDIC, the bridge bank, and the holding company had to figure out how to untangle payroll systems while thousands of people's livelihoods hung in the balance.

One week after the bank's failure, SVB Financial Group filed for Chapter 11 bankruptcy.

Finding a Buyer

The FDIC's first attempt to auction SVB's assets on March 12 attracted just one bid—and it wasn't even from a bank. Major institutions like Bank of America, JPMorgan Chase, and Goldman Sachs all declined. PNC Financial Services and RBC Bank considered bids but backed away.

Why the reluctance? Acquiring a failed bank is complicated. You inherit its problems along with its assets. And SVB's problems were substantial—not just the bond losses, but the reputational damage and the challenge of retaining customers who had just experienced a traumatic bank run.

Eventually, First Citizens Bank of North Carolina agreed to acquire the remains of SVB. The deal, announced in late March, included about seventy-two billion dollars in assets purchased at a discount. The FDIC estimated its Deposit Insurance Fund would take a hit of roughly twenty billion dollars—a cost that would ultimately be borne by the banking industry through higher assessments.

Global Ripples

SVB's collapse wasn't just an American story. The bank had subsidiaries overseas holding nearly fourteen billion dollars in deposits.

In the United Kingdom, the Bank of England moved to place SVB's British arm into insolvency proceedings. The threat to British tech startups was significant enough that regulators worked through the weekend to arrange a rescue. HSBC ultimately bought the UK subsidiary for one pound—a symbolic price that reflected the uncertainty about what the assets were actually worth.

In Canada, the Office of the Superintendent of Financial Institutions seized SVB's Canadian operations. Unable to find a buyer, regulators eventually restructured it into a bridge bank.

SVB had a joint venture in China with Shanghai Pudong Development Bank. That operation, governed by Chinese regulations and chaired by a Shanghai-based executive, was insulated from the American collapse and continued operating.

The Lessons

What does SVB's collapse teach us?

That bank runs can happen with terrifying speed in the digital age. In the 1930s, a bank run meant lines of people waiting outside branches. In 2023, it meant forty-two billion dollars transferred out via smartphone apps in a single day. Social media and messaging apps spread panic faster than any regulator can respond.

That concentration is dangerous. SVB had concentrated its depositor base in one industry and its assets in one type of investment. When tech companies all needed cash simultaneously, and when interest rates moved against the bond portfolio, the bank had nowhere to hide.

That regulatory relief has costs. The 2018 law that reduced oversight for mid-sized banks was sold as eliminating unnecessary burdens on smaller institutions. But SVB wasn't small—it was the sixteenth largest bank in America. The reduced scrutiny may have allowed problems to fester that more intensive oversight would have caught.

That deposit insurance limits may need rethinking. The two hundred fifty thousand dollar cap made sense when most bank customers were individuals. But in an economy where businesses—including those employing thousands of workers—keep millions in their operating accounts, the limit seems anachronistic. Yet raising it creates moral hazard: why should depositors bother evaluating their bank's health if they know they'll be protected regardless?

And finally, that confidence remains the foundation of banking. Banks work because people believe they work. The moment that belief wavers, even a solvent bank can fail. SVB might have weathered its bond losses if depositors had remained patient. Instead, the rush for the exits created the very catastrophe everyone feared.

A Snapshot of Fragility

Silicon Valley Bank's collapse wasn't caused by fraud or wild speculation or exotic financial instruments. It was caused by something much more mundane: a mismatch between the duration of its assets and the volatility of its deposits, combined with a rapid rise in interest rates that almost no one had predicted.

In other words, it was caused by a bet that interest rates would stay low. When that bet went wrong, everything unraveled.

The episode revealed how quickly the modern financial system can destabilize. It showed that even forty years of successful banking can end in forty hours. And it demonstrated that in an era of instant communication and mobile banking, the old rules about how fast a crisis can develop no longer apply.

For the startups that nearly lost their operating capital, for the employees who spent a weekend wondering if they'd get paid, and for the regulators who had to decide in hours whether to backstop deposits that were explicitly uninsured, the collapse of Silicon Valley Bank was a stark reminder: the financial system, for all its sophistication, remains built on something as fragile as trust.