Here's our latest article on bank safety: "The Fed Lost $113 Billion And Is Underwater By $70 Billion."
As part of our ongoing series of articles on bank stability, and at the request of many of our clients, we wanted to address the major risks we foresee for bank stability in the coming years.
But before we begin, I want to take this opportunity to remind you that we have reviewed many larger banks in our public articles. But I must warn you: The substance of that analysis is not looking too good for the future of the larger banks in the United States, details for which are here.
Moreover, if you believe that the banking issues have been addressed, I'm sorry to inform you that you likely only saw the tip of the iceberg. We were able to identify the exact reasons in our public article which caused SVB to fail, well before anyone even considered these issues. And I can assure you that they have not been resolved. It's now only a matter of time.
We have been discussing maturity mismatch risk in the U.S. banking system for more than 18 months. It already has led to the infamous failure of SVB, and, importantly, it still continues to destabilize the banking system. Despite the fact that many experts and officials are saying that the liquidity crisis in the U.S. has ended and is unlikely to repeat, the liquidity situation in the country's banking system remains challenging, and we have recently published a detailed article on this issue.
U.S. banks turned out to be completely unprepared for a sharp rise in interest rates. The annual Fed’s stress tests show only how the banks would perform in a low-to-zero interest rate environment. It may come as a surprise to some, but the regulator has never officially published a stress test with assumptions of higher interest rates.
The recent data published by the St. Louis Fed shows that even the U.S. banking regulator itself is completely unprepared for a rising rate environment. The chart below shows that the Fed has posted a loss of almost $100B since the beginning of the year, and its total loss since September 2022 has reached $113B. Importantly, these are not paper losses but real losses to the U.S. Treasury and, hence, a part of the U.S. budget deficit.
In this article, we would like to take a deeper look at this quite extraordinary event.
The Fed has not yet published its financial report for 3Q23, and, as such, we have only the 2Q/1H data. The table below shows a breakdown of the Fed’s net interest income (expense). The regulator's interest income for the first six months of the year was $88.4B. However, its interest expense was $141.8 B. As a result, the Fed’s net interest expense for 6M23 was $53.5B.
There also were operating expenses of $4.4B, and, as a result, the Fed’s total net loss to the Treasury for 6M23 was $58.1B. The chart by the St. Louis Fed, which we demonstrated earlier, shows that, as of Oct. 25, this loss was $113B. This implies that the Fed has lost another $50B since Q2.
The reason for this loss is the Fed’s balance sheet. Here's a breakdown of the Fed’s interest income on the SOMA (System Open Market Account) portfolio with average daily balances and average interest rates. Interest income on the SOMA book accounted for 94% of the Fed’s total interest income for 6M23.
Based on these numbers, we can calculate the weighted-average yield on the Fed’s SOMA portfolio, which is 2.0% for 6M23.
Now, let’s look at the regulator’s interest expense. 42% of its total interest expense is interest expense on securities sold under agreement to repurchase. As shown below, the Fed is paying 4.75% on these instruments.
The rest of the regulator’s interest expense is what it's paying to depository institutions. According to the Fed, depository institutions earn interest at the interest on reserve balance (IORB) rate, which is currently 5.4%.
Based on that, we can calculate that the Fed’s weighted-average cost on its interest-bearing liabilities is 5.1% for 6M23. Given that the weighted-average yield on the SOMA book is 2.0%, the regulator has a negative net interest spread, which is more than 3%. Obviously, this is not a sustainable situation, and any commercial bank will fail in the medium term with such a net interest spread.
It's interesting how the Fed names its losses:
In the fall of 2022, the Reserve Banks first suspended weekly remittances to the Treasury because earnings shifted from excess to less than the costs of operations, payment of dividends, and reservation of surplus. The Reserve Banks began accumulating a deferred asset, which represents the net accumulation of costs in excess of earnings and is reported as "Deferred asset—remittances to the Treasury" in the Combined statements of condition.
It is a “deferred asset” rather than a loss. This makes sense from the regulator’s perspective, as otherwise, with a total loss of more than $113B and a capital of $42B, the Fed would have negative equity of more than $70B.
It's important to note that in mid-2022, the Fed published an academic paper saying that it expects some losses due to higher interest rates. The regulator’s base case was that these losses would peak at $60B, while the bearish case was that they would peak at about $180B:
While the deferred asset reaches a peak of about $60 billion in the baseline projection, the tail risk in these projections, represented by the upper edge of the dark-blue area, indicates that the deferred asset could reach as high as about $180 billion.
However, as we see, the Fed’s losses are already more than $110B, and even optimistic forecasts suggest that they will surpass $200B and peak only in 2025.
Another interesting thing from this Fed’s paper is the regulator’s estimates of unrealized losses from the SOMA portfolio:
Under the baseline projection, the unrealized loss position is projected to reach a maximum of about $670 billion, or about 8 percent of the portfolio's par value, by the end of 2022. The unrealized loss position then declines gradually as SOMA securities associated with unrealized losses approach maturity; at the maturity date, the market value of each security returns to its face value. The range of outcomes for the unrealized loss position stemming from much higher interest rates than under the baseline is quite large; for example, as indicated by the lower edge of the 90-percent confidence band which represents a tail risk event, unrealized losses peak at around $1.1 trillion, or about 16 percent of the portfolio's par value, in 2023.
But, as the table below shows, even the Q2 data demonstrates that these losses are already more than $1T. According to the Fed, this tail-risk event, which by definition has a very low probability, has already realized. This reminds us of the 2007-2008 crisis, when some tail-risk events occurred.
One might argue that low-yield securities on the Fed’s balance sheet will expire soon and will be replaced by higher-yielding instruments. But a maturity analysis of the Fed’s assets tells us a different story.
46% of Treasury bonds have maturities of more than five years, and 80% of Treasury bonds have maturities of more than one year. Moreover, 98% of MBS have maturities of more than 10 years.
Some economists say that a recession in the U.S. economy, which is expected by quite a lot of experts, would lead to a decrease in interest rates and would solve this issue with the Fed’s losses. However, a potential recession will likely trigger a major banking crisis given that U.S. banks have a lot of issues, which we discussed in our previous articles. On the other hand, if the Fed’s losses continue to increase, it also will likely be a major issue for the stability of both the U.S. economy and the banking system.
Another takeaway from our analysis is that the Fed’s estimate of its own losses and its assessment of its tail risk events make some question the reliability of the regulator’s stress tests, which it publishes annually for U.S. banks.
As Alan Greenspan was quoted as saying:
“We really can't forecast all that well, and yet we pretend that we can, but we really can't.”
The obvious question here is: how can the Fed assess the safety of the U.S. banks when it can’t correctly estimate the risks of its own balance sheet? We have written a lot about the Fed’s stress tests, but this situation, with huge losses for the regulator, has undermined their credibility even further, in our view. We believe retail depositors should not really on the Fed’s stress test when choosing a bank for their savings.
At the end of the day, we're speaking of protecting your hard-earned money. Therefore, it behooves you to engage in due diligence regarding the banks which currently house your money.
You have a responsibility to yourself and your family to make sure your money resides in only the safest of institutions. And if you're relying on the FDIC, I suggest you read our prior articles which outline why such reliance will not be as prudent as you may believe in the coming years.
It's time for you to do a deep dive on the banks that house your hard-earned money in order to determine whether your bank is truly solid or not. Our due diligence methodology here.