The run on Silicon Valley Bank (SVB) occurred in the blink of an eye. On Wednesday, March 8, SVB announced that it had sold $21 billion worth of securities, resulting in an after-tax loss of $1.8 billion and that it sought to raise over $2 billion to offset those losses. On March 9, investors and depositors attempted to withdraw $42 billion from the bank, a visceral show of the loss of confidence that followed the bank’s announcements. The next day, on March 10, the bank was seized and placed under the Federal Deposit Insurance Corporation (FDIC)’s receivership. For the public and those not involved in the financial industry or the technology and venture capital industries — who were SVB’s primary customers — it appeared that a bank with some $209 billion in assets had suddenly collapsed overnight.
Of course, that was not the case. Put simply, SVB’s failures stemmed from its investment strategy. As they received billions in deposits during COVID from a strong technology industry, they invested their money in long-term Treasury bonds. These long-term Treasury bonds are widely regarded as risk-free, backed by the full might of the United States government. You invest the money with the U.S. Treasury, and over the bond’s lifetime, you receive periodic interest payments in return. Once the bond has fully matured, you receive back the money you initially invested. Additionally, during its lifetime, you can sell the bond on the secondary market to someone else, who will subsequently receive the payments from the Treasury. In this system, the profit you get from holding the bond is equal to the interest rate at the time of purchase — set by the Federal Reserve — and thus interest rates are intimately (and inversely) related to bond prices.
At the time that SVB purchased their long-term Treasury bonds, the interest rate was near zero, as the Federal Reserve aimed to encourage economic activity during the COVID pandemic. But recently, the Federal Reserve raised interest rates dramatically to attempt to slow down the economy and reduce inflation. Importantly, this meant that the interest rates on the Treasury bonds that SVB held were now significantly lower than the interest rates that investors could get if they simply purchased new bonds directly from the Treasury. Therefore, on the secondary market, if SVB needed to sell their bonds to raise capital, they would be unable to sell them at the same price they bought them — forcing them to absorb losses. If they didn’t have to sell the bonds and could hold them until maturity, they would still maintain a profit.
Unfortunately, as the technology industry regressed and could no longer easily raise new funds, they began to withdraw their money from SVB. This forced SVB to sell their securities at a loss to raise funds, which in turn caused their depositors to panic even further and withdraw more funds: a vicious cycle that ended with the second-largest bank collapse in American history.
SVB’s collapse can be best understood as a bank being overexposed to risk, specifically interest-rate risk, by buying up billions in Treasury bonds in a low-interest rate environment, and insufficiently accounting for that exposure as interest rates rose. When their limits were tested by a simultaneous downturn in the industry with which they did business — forcing their depositors to withdraw money from the bank — they collapsed. Naturally, there follows an important question. Why, after we all had experienced what happens when banks take on too much risk with too little oversight during the 2008 crisis, was this not caught?
One potential reason lies in a 2018 law, signed by then-President Donald Trump: the Economic Growth, Regulatory Relief, and Consumer Protection Act, better known by its quite unaesthetic abbreviation, the EGRRCPA. Among many other things, the EGRRCPA rolled back regulation from the Dodd-Frank Act that required banks with over $50 billion in assets to undergo more stringent stress testing and oversight, raising this threshold to $250 billion. SVB’s CEO, Greg Becker, lobbied strongly in favor of the deregulation, which placed mid-sized banks such as SVB in a realm of reduced regulation and oversight.
While it is impossible to say with certainty if previous, more stringent regulations would’ve caught SVB’s warning signs ahead of time, the EGRRCPA definitely made it easier for regulators to overlook them. Following EGRRCPA, banks of SVB’s size were exempted from mandatory company-run stress tests, subject to only periodic stress tests, especially a problem thanks to SVB’s meteoric growth during the COVID pandemic, and exempted from heightened liquidity requirements, which could’ve led to more adaptability in the face of increased withdrawals. Additionally, the EGRRCPA “shifted supervisory priorities,” making it clear that banks of SVB’s size were not the Federal Reserve’s primary worry any longer — creating an environment where the Federal Reserve may be hesitant to utilize regulatory authority it had, given Congress’s stance in the EGRRCPA.
If there is a key lesson to be learned from SVB’s collapse, it is one that we have learned time and time again — in an industry as essential to the quality of life for millions of Americans as the financial industry, regulation is paramount. Bankers, in pursuit of heightened bonuses and salaries, will always lobby aggressively for deregulation in their industry. Nevertheless, it is an essential responsibility of our lawmakers to protect their constituents from the negative effects of bank runs and bank collapses like SVB, and they should hold firm and maintain effective oversight of the financial system.
Progress in this regard is far from impossible; it happened after the 2008 financial crisis, and can certainly happen again after SVB’s dramatic failure. Already, Senator Elizabeth Warren and other House and Senate Democrats have revealed legislation to undo the deregulation of the EGRRCPA, and President Joe Biden has also postured in favor of stronger regulation. The rest of Congress must follow swiftly, lest we be forced to learn this same lesson yet again in the future.
