SVB made obvious mistakes that say a lot about the California tech culture, and the federal response says worse things about how we are governed now.
Silicon Valley Bank made obvious mistakes that say a lot about the California tech culture, and the federal response says worse things about how we are governed now.
B ank failures are always complicated stories, even when they are, at bottom, very simple. It is worth unpacking a few of the angles of the failure of Silicon Valley Bank, both in public-policy terms and in more basic business terms.
The simple story here appears to start with the fact that SVB invested its money in ways that were inconsistent with being a bank — specifically, a bank with a potentially volatile customer base. When interest rates were low and the bank had a lot of cash to invest (too much cash and too few business opportunities being a hallmark of the 2020-21 period), it loaded up its balance sheet with low-yielding government bonds. These were, for the most part, extremely safe and conservative investments if you could afford to hold them until they matured. But if you might need to sell the bonds in the market to pay back depositors or other creditors, these plain-vanilla investments could be highly volatile, because nobody wants to buy low-interest bonds in a high-interest environment unless they are sold for enough of a discount to deliver a yield that is equivalent to a high-interest bond.
To use a simplified example: If Bond A promises to yield 2 percent (a payoff of 102 on an investment of 100), and the same issuer puts out an identical Bond B promising to yield 3 percent (a payoff of 103 on an investment of 100), then the price of Bond A has to drop from 100 to around 99 to give the same expected return on investment. If Bond C promises an 8 percent yield, Bond A drops to 94.4. In other words, the more available interest rates rise, the more the low-interest bonds will decline in market value even if they are completely safe investments. That, in a nutshell, is interest-rate risk. And it can get steeper than this once you move away from the simplest financial instruments.
If you’re an individual investor looking for a conservative place to park some of your retirement funds, or a university looking for a secure portion of its endowment, interest-rate risk just means that when the bonds are paid off, you got a lower return than if you’d bought something else. But a bank needs liquidity, and the more erratic its inflows and outflows are, the more liquid it needs to be.
SVB bought too many things that were exposed to the specific risks of rising interest rates. It tried to value some of those things on its balance sheet as hold-to-maturity assets (in my example above, valuing the 2 percent bond at 102 instead of 99 or 94) instead of mark-to-market assets, the value of which has to be adjusted for market prices. When it came under pressure to liquidate or at least value those assets at liquid market prices, its insolvency became obvious. That was, it seems, all out in the open, which is why Silicon Valley tycoons and journalists sounded the alarm, triggering a bank run. If you’ve seen It’s a Wonderful Life or Mary Poppins, you know what a bank run is — no bank ever has enough cash on hand to pay everybody’s deposits back simultaneously, because those deposits are used to make loans and investments. Only the mechanics are different in an age of online banking and a bank whose customer base was largely tech companies.
There are a couple of immediate questions that should occur to anyone reading this who is above a certain age and/or has some experience in finance, and those questions should be asked first as a matter of SVB’s business judgment and then as a matter of bank regulation. How did it not see the risk that it might need liquidity and should not be accounting for a significant chunk of its assets as if they’d be held to maturity? How did SVB not see the risk that interest rates could rise, and have in place a hedge to guard against that?
Even before inflation really took off in 2021, there were a lot of warning lights flashing that it could be a problem, and anyone who knows what the Federal Reserve is knows that, if inflation becomes a problem, the Fed will raise rates. This was not ancient history: The Fed had raised rates by a quarter point at nine quarterly meetings in a row from 2015 through 2018, when inflation was a far less immediate threat. I refinanced my mortgage in the summer of 2021 because it was obvious to me at the time that rates this low would not last forever.
Some folks who have been around a little longer noted that SVB made the same basic mistake as Orange County, Calif., in the 1990s, a fiasco that ought to have lingered in the memory of Californians if the bank was actually run by people who were adults in California in 1994. Robert Citron, Orange County’s tax collector and county treasurer, put a heavy bet on a portfolio that assumed interest rates would not rise — and they did, pushing the county into a humiliating bankruptcy and years of litigation. SVB’s risk-management team should have sounded the alarm on this — but instead, the company “operated without a chief risk officer between April 2022 and January 2023.” Who was in charge? Press reports have noted that the bank’s European risk-management head was a publicly prominent spokeswoman for various woke causes; maybe she spent too little time on the core elements of the job, or maybe, being stationed overseas, she just wasn’t kept in the loop on evaluating the bank’s portfolio. But someone should have been — especially as rates kept rising throughout 2022 and the bank could have cut some of its losses or purchased more investments that would appreciate in value if rates went up, thus hedging its risk exposure.
The second major piece of the puzzle is SVB’s customers. Bank Customer 101: The FDIC ensures your deposits up to $250,000, so as a non-wealthy individual or a modest small business, you don’t really need to worry about the soundness of your bank’s balance sheet. But many of SVB’s customers were tech startups and executives who had a lot more than that in the bank. Part of the problem was the incestuous relationship between venture-capital funders and the bank, which meant that many companies were not given a choice — their funders demanded that they keep the funds at SVB. As Jim Geraghty notes:
Silicon Valley Bank was the bank of choice for 1,074 private-equity and venture-capital funds in recent years. Fortune explains that, “As venture capitalists are legally required to ‘custody’ their fund assets at a financial institution, they use ‘custodians,’ which are banks like SVB that will monitor and safekeep the capital and ensure it is not stolen or lost.” In other words, when venture capitalists invest in a start-up company, they require that company to use their bank, in this case, Silicon Valley Bank. In retrospect, it was wildly risky to have almost an entire sector of the U.S. economy doing so much of its banking through one mid-sized bank. . . .
When you choose to keep $3.3 billion in a particular bank, as payment technology firm Circle did, you have an enormous incentive to keep an eye on the financial health of that bank . . . streaming service Roku had $487 million in Silicon Valley Bank; defunct crypto lender BlockFi had $227 million; and Roblox, the California-based online-gaming platform had $150 million. BuzzFeed “said it had about $56 million in cash and cash equivalents at the end of 2022, majority of which was held at SVB.”
As noted, there is at least one defensible reason for that: It made it easier for the funders to keep tabs on the startups. But it made the whole edifice highly dependent upon SVB’s soundness. Silicon Valley techbros had a huge financial interest in keeping an eye on that soundness, or — alternatively — in paying to insure their deposits.
Why didn’t they? Partly hubris and a failure of imagination, and partly that they calculated (correctly) that their politically influential industry was too big to fail and would be bailed out if anything went wrong, so they could reap the rewards while things were going well and then socialize the losses later without pre-paying for insurance. While the specific economic causes are different, the psychology at work here seems reminiscent of the collapse earlier this year of Sam Bankman-Fried’s crypto exchange FTX and its impact on people who used the exchange as a custodian for their crypto investments, even though the whole point of cryptocurrency is supposed to be its independence from exposure to institutional failures.
Was it totally unforeseeable that a bank could fail? I question how this basic reality of banking can possibly be news to anyone over the age of 30. Is 2008 really that far beyond the horizons of human memory? Even tech startups should have somebody on their board with enough gray on their head to recall Orange County and the S&L crisis of the early 1990s.
There is a more general point here about California’s youth-obsessed tech industry. If your business — or your political campaign, journalistic enterprise, or army, for that matter — is run by people who don’t even socialize much with anyone over 40, you are probably failing to prepare for some risks that will be obvious to people with a longer perspective. In fact, this is probably the single thing older people do best (often to excess): warn younger folks about the risk that things that have happened with some regularity in the past are apt to happen again. Things such as, “Hey, remember inflation?” or “You know, banks can go under.” This is similar to the problem that we saw in 2008, when too many people in the mortgage-backed-securities industry used financial models whose datasets did not go back to the last crash of the housing market.
It was not just the bank and its customers. The bank’s auditors at KPMG kept certifying it as a going concern all the way to the end. While the auditors can defend the accuracy of the bank’s financial statements — really, all of this was out in the open — a major auditing firm is supposed to have the necessary gray hairs and gravitas to be able to sit the top people at the bank down when it is obvious that they have no meaningful risk management.
The going-forward question for bank regulators is almost certainly going to be a crackdown on hold-to-maturity accounting by banks. But there is also a looking-back question, one with echoes of 2008. Recall that the big Wall Street watchdog types such as then-New York attorney general Eliot Spitzer and then-Connecticut attorney general Richard Blumenthal were, in the years before 2008, focused on the wrong problems. Regulators liked doing the easy headline-grabbing work of calling out comparatively petty conflicts of interest (Spitzer obsessed over Wall Street’s allocation of IPO shares and the integrity of its equity-research reports) rather than the hard work of stress-testing the actual financial soundness of the banks. Any time there is a crash, we get demands for more regulation, which only spreads the regulators thinner and tempts them more to exercise their penchant for seeking publicity by attacking petty venality or pursuing unrelated side issues such as workplace diversity and climate change, and that draws their attention ever further away from the whole reason why we have bank regulators at all.
Then, there is the bailout. The Federal Reserve has effectively announced that it will guarantee all deposits, the legal limit to FDIC insurance be damned, and the FDIC will pay for it out of its insurance fund. Now, I am not a Randian free-market purist: It is entirely appropriate for a central bank to take a role in a faltering bank or brokerage house to ensure an orderly wind-down, stave off panic, and prevent valuable assets from being sold at fire-sale prices. Bank regulators should be like the fire department, who show up and stay long enough to douse the blaze.
There is even a case to be made that, as a matter of ex ante policy, we should have higher deposit-insurance limits for the bank accounts of companies than for individual depositors, on the theory that a company losing its shirt can lead to a lot of innocent employees going unpaid or being laid off. (Of course, such a rule would require additional red tape to ensure that wealthy individuals don’t just make deposits through a shell corporation with no employees).
But there are three big problems with saying that this policy should apply here. One, the federal government is bailing out all of SVB’s over-$250,000 depositors, not just the ones that are companies needing to make payroll. Two, there is a tradeoff in raising the insurance limit, as it creates costs to the bank that are passed on to depositors. If you let people collect on insurance they never paid for, and which other similarly situated depositors also didn’t pay for (so the benefits go only to the people whose bank failed), what you’re doing is the opposite of insurance.
Third — a point that seems to be overlooked in this whole discussion — what the Fed and the Treasury are doing here is, effectively, changing the law, and they did so over a weekend without even considering asking Congress to play some role in lawmaking. At least the Bush-era Wall Street bailouts that triggered so much popular opposition on the right from the Tea Party and on the left from Occupy Wall Street were blessed by Congress after a vigorous debate in which the first version of TARP was rejected by the House of Representatives (which may as well now be renamed the House of Cable Commentators on the Real Government). Just as in the student-loan amnesty and other Biden executive actions, the message is that the executive alone dispenses favors, and can rewrite the law at will. Nobody is even going through the motions of pretending that we’re a nation of laws at this point.
Finally, we see already the downstream effect that is likely to follow: The fight against inflation will be called off, while it is still running at 6 percent. Instead of eating our vegetables and getting better, as we did under Paul Volcker between 1979 and 1982 to tame inflation for a generation, we will try the same halfhearted moves that were tried and failed for half a decade before him. The Fed and the Treasury could plausibly argue, in 2008-09, that the financial system needed more liquidity; today, it needs the opposite, but the regulators only have a hammer, so they will treat everything as a nail.