When Crypto Crashes, Banks Burn
February 15, 2023
This is part 2 of a series on the 2022 crypto collapse. Part 1 is here.
In part 1, we covered how crypto firms collapsed in a chain through 2022. What happened next is less covered but worth understanding: the damage didn't stop at the edge of the crypto world. It kept going and took down traditional banks with it.
In early 2023, three banks failed in rapid succession: Silicon Valley Bank, Silvergate Bank, and Signature Bank. None of them traded cryptocurrency. But all three had built their businesses around crypto companies, and that turned out to matter a lot.
What it means for a bank to "work closely" with crypto
These banks didn't just have a few crypto clients on the side. Their deposit bases were heavily concentrated in crypto companies. Silvergate, for example, had built an entire payments network specifically for crypto businesses. Signature was similar — a large portion of its deposits came from the same sector.
This matters because of how banks work. A bank takes in deposits and puts that money to work: lending it out, buying bonds, earning interest. As long as customers don't all withdraw at once, the system functions fine. The risk is a bank run — when confidence drops and everyone tries to get their money out at the same time.
When crypto firms started collapsing in 2022, those firms needed cash. They withdrew deposits. Word spread that the banks had heavy crypto exposure. More customers got nervous and pulled their money. The banks couldn't cover it fast enough. That's a bank run, and it's exactly what happened.
What actually happens when a bank collapses
When a bank fails, customers temporarily lose access to their funds. The FDIC, which is the federal agency that insures bank deposits, steps in to take over. For most people with accounts under $250,000, the FDIC guarantees they'll get their money back. But there's a period of uncertainty, and for businesses that kept large amounts in those accounts, recovery could take time or be partial.
For the crypto companies that banked with Silvergate and Signature, the collapse created a practical problem: payment infrastructure disappeared overnight. These banks had built the rails that crypto companies used to move dollars in and out. When the banks went down, those rails went with them.
The coupling problem
Here's what's worth sitting with: these banks didn't stumble into a bad bet. They made a deliberate choice to become the financial infrastructure for an emerging industry. And for a while, it worked well. Crypto was growing, and the banks grew with it.
But deep integration cuts both ways. When you build your business on top of a platform — or build your business around serving one — you take on that platform's risk. If the platform struggles, so do you. If it collapses, you might collapse with it.
This is sometimes called ecosystem lock, and it's usually discussed from one direction: a company becoming too dependent on a platform it doesn't control. What happened here is the same dynamic from the other direction. The banks became so coupled to the crypto ecosystem that the ecosystem's failure became their failure.
What to take from this
Platform dependency is a risk that compounds quietly. It's most visible in hindsight, after the thing you were coupled to falls apart.
The question worth asking, for banks or any business with a concentrated platform bet, is: if the platform we're most embedded in had a bad year, how bad would our year be? If the answer is "catastrophic," that's worth knowing before it happens.
The 2023 bank failures weren't a crypto story or a banking story. They were a platform dependency story. The mechanism is the same whether you're a business relying on a single app store, a startup built on top of a single API, or a bank whose deposit base is a single industry.
Deep integration with a growing platform is a growth strategy. It's also a concentrated risk. Both things are true at the same time.