I know bank runs.
In 1998, I was Director for Financial Advisory Services (FAS) for Deloitte & Touche in Budapest. FAS was the investment banking arm of Deloitte, and I had been chosen to lead the privatization of Hungary's second largest bank, Postabank. It was run by Gábor Princz, a short, squat man with a bad reputation.
Most Americans are not familiar with postal banks, but many European countries have them. These allow members of the public to deposit and withdraw cash at their local post office, handy for a country with thin banking coverage. As a result, postal banks are the most retail of all retail banks, true widows-and-orphans institutions, the place Hungarian retirees would go to collect their meager pensions in cash.
Postabank should have been a boring bank, but this was post-communist Hungary. One could hear on the street that "Postabank stands behind" such and such an entrepreneur. This would be truly bizarre posture for a commercial bank. You will never hear, for example, "Bank of America stands behind the Stop-'n-Shop." However, we might instead expect to hear that from a private equity or hedge fund. And that's exactly what Princz had done with Postabank: transformed it into his own personal hedge fund. Princz used Postabank to bankroll any number of dodgy characters, many of them with better political connections than business expertise.
As was the case with everything in Budapest, this was widely known and a cause for concern. The Socialist government of Gyula Horn was keen to privatize the bank. The problem was, however, that none of the Big Four accounting firms was willing to sign off on the bank’s audit. Well, none except Deloitte & Touche. Some senior partners at Deloitte had historical ties to the Socialist Party from way back in the socialist days. (There is a direct line from this to Russia's capture of Hungary's foreign policy, by the way.) Over howls of protest from the auditors, Deloitte certified Postabank's books. No doubt purely by coincidence, Deloitte was promptly awarded the mandate to privatize the bank.
I was chosen to be the privatization lead.
A week or two later, at 10 am on February 28th of 1998, I was at my desk when I saw a murmuring among my Hungarian staff, flurries of emails, and hushed phone calls. I naturally asked what was going on, to be told that Postabank was rumored to be failing and depositors were queuing to get their money out. By noon, it was all over. The bank had failed. In that two-hour stretch, Hungary had lost more than 1% of its GDP.
This colorful, if disastrous, story from post-communist Hungary is a tale of a classic bank run, of depositors -- panicked that bank management had lent their money to shady or uncreditworthy borrowers -- stampeding to withdraw their money.
This is not the story of SVB.
No one has accused SVB of reckless loans to bad clients. SVB did not and does not have a non-performing loan problem. Indeed, half of its deposits were invested in US government treasuries and agency-backed (government guaranteed) mortgage securities. Lending money does not get safer than that, and indeed, Investopedia notes that "the interest rate on a three-month U.S. Treasury bill (T-bill) is often used as the risk-free rate for U.S.-based investors." To all appearances, SVB set conservative lending standards.
So why the run on SVB?
It is a story of bad management, not at the commercial banks, but rather at the US Federal Reserve Bank and Treasury. The most important error was one of construction. The Fed believed that the pandemic downturn was the equivalent of the Great Recession and required a ‘go big’ policy. As Fed Chairman Jerome Powell saw it in October 2020:
"Too little support would lead to a weak recovery, creating unnecessary hardship for households and businesses. Over time, household insolvencies and business bankruptcies would rise, harming the productive capacity of the economy, and holding back wage growth. By contrast, the risks of overdoing it seem, for now, to be smaller. Even if policy actions ultimately prove to be greater than needed, they will not go to waste."
This view underpinned the Treasury’s decision to push the Federal Funds Rate (FFR) effectively to zero.
Only twice in the prior century had this happened, during the Great Depression and the Great Recession, with the FFR held at zero not for a few months, but for seven years at a time. Because both qualified as depressions, zero interest rates had virtually no effect on either asset prices or inflation. In the Great Recession, for example, home prices did not regain their 2006 peak until 2017.
But the pandemic was not a depression, but a suppression. This is the difference between Jimmy getting the flu and Jimmy being grounded. The economy was not sick during the pandemic, it was locked down. Imagine that the parents ground Jimmy and say, "Stay in your room, but here's $2,000 in spending money." It's clear that Jimmy would be in a mood to spend when conditions permitted. In the meanwhile, Jimmy contemplated what to do with easy money, and went out and invested in real estate and stocks. From Jan. 2020 to June 2022, house prices increased nationally by 45%, and this during a period when a chunk of the economy was shut down. Stocks similarly benefitted, with the S&P 500 doubling in value. But this was nothing compared to the tech stocks. Tesla's valuation, for example, increased 25-fold in less than one year. Unlike the Great Recession, zero interest rates exploded house and tech company valuations. This happened because the Fed and Treasury misunderstood the essential nature of the downturn. It was not a depression, and not even a recession by traditional metrics, as the economy bottomed in less than one quarter. It was a suppression, an external force -- covid and resulting lockdowns -- preventing the economy from operating. By mistaking the nature of the downturn, the Fed and Treasury vastly overstimulated the economy.
For SVB, this would prove the perfect storm. Soaring valuations led tech companies to raise capital and deposit the money in, well, Silicon Valley Bank. SVB's deposits quadrupled during the pandemic. But how should SVB invest the funds? Given rock bottom interest rates, SVB took a conservative approach and invested half of its funds in 3-month treasuries and mortgages guaranteed by the US government. SVB did not make unusually bad loans. No one has suggested that they did.
So what happened?
As it turns out, the action is all on the liabilities -- customer cash deposits -- side. As readers will recall, US Treasury Secretary Yellen and Fed Chair Powell assured the public as late as mid-2021 that inflation was 'transitory'. This faith was based on a rejection of the Quantity Theory of Money (QTM), which holds that the price level is a function of the quantity of money in the economy. This concept is best captured by the famous quote from the Nobel Laureate economist Milton Friedman, who said, “Inflation is always and everywhere a monetary phenomenon, in the sense that it is and can be produced only by a more rapid increase in the quantity of money than in output.” Alas, Friedman’s view had fallen out of fashion, most notably at the Treasury and Fed, giving Powell and Yellen the confidence to blow out the money supply by nearly 40%.
QTM would have cautioned that the price level should rise by near 40%, that is, that the US should experience inflation summing to 40% (less real GDP growth) over a period of, say, 2-4 years. Not so, said Yellen and Powell. At worst, inflation would be transitory.
This notion was not entirely unfounded. From late 2007 to just before the pandemic, increases in the money supply should have generated a cumulative 60% increase in prices. Instead, over those twelve years, the US saw only 20% inflation in total. However, that period saw the Great Recession (really a depression), as well as the rise of China and aging demographics in the US, with all this playing out over more than a decade.
By contrast, the Fed injected a massive amount of liquidity into the economy in 2020, with the resulting dynamics closer to those of the 1970s oil shocks. But even here the comparison is qualified. The second biggest injection of liquidity in the last sixty years came in 1971-1972. During a two-year period, the money supply grew enough to imply a 14% rise in prices. By contrast, in 2020, the Fed injected liquidity three times the 1971 precedent -- and all in one quarter! It is hard to overstate just how aggressive monetary and fiscal stimulus was during the pandemic. The result was persistently high inflation, leading to a change of heart at the Fed.
Figure 1
Source: FRED M2SL, GDP, GDPC1, CPI, Princeton Policy analysis
Early last year, the Federal Reserve determined that reducing inflation would require raising interest rates after all, which began in earnest in Q2 2022. By July, it was apparent that the Fed would follow the so-called Taylor Rule or a variant to bring down inflation. A number of economists analyzed the implications, one version of which, labelled “P&P Balanced Approach”, can be seen on the graph below. This suggested that the interest rate on the 3-month treasury bill would reach 4% by year-end 2022 and 4.4% in the first quarter of this year, and so it proved.
Figure 2
Source: Federal Reserve 2022 Stress Test Scenarios, 2023 Stress Test Scenarios; David Papell and Ruxandra Prodan, “The Fed Fell Behind the Curve by Not Following its Own Policy Rules” in Econbrowser (July 2022); KPMG Insights on Inflation; CNBC
Nevertheless, the Fed never subjected the banks to a stress test at anywhere near this level in 2022. The highest level the Fed tested was 0.7%, not the 4.0% which proved to be the case. Indeed, the Fed was testing at 0.5% at mid-year when the 3-month rate was already at 2.7%.
If the Fed was failing miserably in its mission, the banks were most certainly not. The finance department of any bank will run profit and loss numbers weekly, if not daily. They are all familiar with the Taylor Rule and were well aware, by mid-year, of the likely path of interest rates going forward. They must have been alarmed, because these same banks had issued a tsunami of mortgages around 3% interest rates in the orgy of lending from Q1 2020 to early 2022. If the cost of deposits rose to 4.5%, and the banks were earning 3% on mortgages, they would be structurally loss-making, with virtually no prospect of escape. I can only imagine that this was brought to the attention of Secretary Yellen and Chairman Powell not once, but a dozen times, from July onward -- including from the Fed's and Treasury's own analysts.
And this brings us full circle to SVB. The graph below shows the underlying reality of SVB and other commercial banks. From May 2022, the Treasury rate began to climb, rapidly pulling away from bank interest-bearing deposits like savings and money market accounts. As of this past week, banks were paying on average 0.54% interest on existing deposits even as treasuries were offering 4% (pp) more. In a bank like SVB, one with sophisticated corporate clients, depositors will begin to remove the funds from the bank and invest them in much higher paying treasuries. Of course, the bank could offer higher interest rates, and most offer around 4% on new savings accounts today. But this hardly helps, and may hurt. Many retail clients may not appreciate just how low their interest rate is. Therefore, raising the interest rate will eventually raise it even for clients who are willing to accept a lower rate. Consequently, raising deposit rates may actually make the bank worse off than allowing some portion of existing clients to remove their funds.
In the case of SVB, the rapid rise in interest rates led to a kind of structural run on deposits. As the bank was unable to offer competitive terms, depositors removed their money. This forced SVB to sell its assets, those treasuries and agency-backed mortgages, to pay for redemptions. Of course, with the dramatic rise in interest rates, these assets could only be sold at a loss. When the asset liquidation started, SVB's clients lost faith in the bank and triggered a more traditional, panic-driven run.
Figure 3
Source: FDIC National Interest Rates and Rate Caps
There is nothing particularly special about SVB, regardless of what you might read. It was more exposed than other banks due to its affiliation with the tech industry. Nevertheless, the regional banks, those who make a living issuing mortgages to homeowners in their communities, are similarly at risk. For this reason, the Mid-Size Bank Coalition of America, which represents more than 100 lenders, called on the Federal Deposit Insurance Corporation to put backstops in place and broaden its protection for smaller banks. A group of economists underscored the risk, noting that “almost 190 banks are at a potential risk of impairment to insured depositors.” Nor is that the upper limit, as the study authors further caution that, “if uninsured deposit withdrawals cause even small fire sales, substantially more banks are at risk.”
*****
Any number of articles in the media have impugned the integrity of bank management, notably at SVB. Even more are calling for ever stricter regulation.
The current crisis is not a regulatory or moral hazard problem. It is entirely the making of the US Federal Reserve, abetted by the US Treasury. Once again we see exceptionally poor analysis at the Fed and Treasury.
It is high time for the Fed to run appropriate stress tests to determine how much the banks can bear. Short-term interest rates must be set to a level compatible with the survival of the US banking sector. That may mean higher inflation for longer, which is exactly the implication of exploding the money supply by 40% three years ago. The Fed and Treasury must assume responsibility for their mistakes, rather than trying to foist their errors onto the banking community.