The big story over the weekend was the scramble over the shuttering of Silicon Valley Bank (SVB), which California regulators shut down Friday after a run on the $200 billion bank. At the core, there are two main issues.

Issue number one is the impact of the Fed’s rate increases on SVB’s capital, which was largely held in longer-term Treasury securities. These are extremely exposed to interest rate risk. As an illustration, when the interest rate is 50 basis points, the promise to pay $1 10 years from now is worth $0.95 – practically a full dollar. If interest rates rise to 4.5 percent, the same promise is worth only $0.64 – essentially a decline in value of one-third. The same thing happened to SVB over the recent past as the Fed hiked rates to fight inflation, making it impossible to sell the securities and meet the demands of customers wanting their deposits back.

Eakinomics readers were doubtless aware that the Fed had raised rates and was promising to continue to raise rates. This raises the question: Why didn’t SVB manage its interest rate risk? It looks like a pure management failure.

Issue number two is the bank’s very narrow customer base. SVB’s customers are nearly exclusively startups concentrated in tech and their venture capitalists. This is extremely undiversified; when one tech firm needs to lay its hands on money it is quite likely that all of the customers will want their deposits at the same time. Moreover, the small group of venture capitalists are sophisticated investors who could easily diagnose issue number one. When it became obvious there was a problem, SVB’s customers withdrew $42 billion from their accounts on Thursday. That’s $4.2 billion an hour, or more than $1 million per second for 10 hours straight. Welcome to the digital bank run. The previous largest bank run in modern U.S. history took place at Washington Mutual bank in 2008, and totaled $16.7 billion over the course of 10 days.

The question that remains is what happens next. Almost none of SVB’s deposits were Federal Deposit Insurance Corporation (FDIC) insured, as its customers overwhelmingly had more than the $250,000 amount the FDIC insures in the bank. Yesterday, federal regulators announced a plan to bail out depositors. No customers will suffer losses, and none will have to wait for whatever money they will get until the FDIC sells off the bank’s remaining assets.

This strikes me as a mistake because a related issue is whether this is an isolated event or a prelude to financial distress elsewhere in the system. Put differently, is there an entire generation of management so used to zero interest rates that it is incapable of managing interest rate risk? If there is a chance of more failures, the precedent set by the SVB episode becomes even more important.

Notice that this is entirely different from the 2007-08 financial crisis. That crisis was global in nature (some of the initial failures were overseas), often was exacerbated by losses in special vehicles that were not even on the banks’ balance sheets, was difficult to diagnose because of the complexity of derivatives and the opaqueness of balance sheets, and was turbocharged by high leverage. This (thus far) is Silicon Valley, is a straight balance sheet mismanagement, and was so transparent that sophisticated customers took their money and ran.

This is how markets work and in my view, the government should let the episode run its course. SVB should fail and the customers who did not diversify their banking or break up their deposits into $250,000 chunks will have learned a valuable lesson.