Punched in the Mouth by Inflation
Time for Reality in Bank Balance Sheets
The banking crisis testifies to the genuine menace of inflation, which distorts the value of future promises denominated in dollars.
In recent years, central bankers and others became complacent, as if inflation would necessitate merely a proportional adjustment, thus (modestly) rising prices would be offset by modestly rising wages. Who would really care?
If everyone lived only in the present tense, so it would be.
Alas, financial markets are constructs for discounting future payment streams. Inflation does not much affect the value of current assets, but it greatly diminishes the value of future assets. In short, when inflation shot up to 9%, financial markets could not but help be rocked.
Michael Barr, the Fed’s vice chairman for supervision, has blamed the failure of Silicon Valley Bank on bad management. But it’s part of the job of regulators to assume that private actors will behave badly. Because we know that some of them will.
Yes, it was the height of recklessness on the part of SVB to bet that interest rates would remain quiescent. But it was virtually guaranteed that someone would be exposed to the long end of the borrowing curve. The inflation caused by imprudent deficit spending and, for too long, an overly accommodative Fed, was bound to upset somebody.
And it was inevitable that the somebodies would include banks. In the world as we know it, banks are machines for turning short-term deposits into long-term assets. SVB and other banks have taken a lot of heat for investing in (longer-term) bonds.
But virtually all bank assets are longer term. This is the economic purpose of banks—to collect retail savings and convert them into loans to start-ups and shopping centers. When inflation erodes the value of those assets, the game doesn’t work.
We now have evidence of its cost. As computed by the FDIC and reported in our previous post, due to rapidly rising interest rates, securities held by all U.S. banks have suffered a $620 billion loss in market value. But banks own more than securities. They own commercial and real estate loans, residential mortgages, business loans and so forth.
A study on bank losses hot off the press by a quartet of banking experts at, respectively, the University of Southern California, Northwestern University, Columbia and Stanford, calculates that, due to the scourge of suddenly higher interest rates, the value of all banking assets—not just tradable securities—is approximately $2 trillion less than book value. Two trillion, by the way, is approximately the equity value of American banks.
Two caveats: some banks partially hedged their rate risk, which would mitigate estimated losses, and many continue to pay less-than-market rates on deposits (again, somewhat offsetting losses).
Nonetheless, these numbers mean that the collective market value of bank equity is far less than reported, and in some individual cases is surely less than zero. For nearly half of U.S. banks, the paper asserts, the marked-to-market value of assets would be insufficient to meet liabilities.
Gregor Matvos, one of the authors, told me that one of their conclusions is that in some respects, SVB was not an outlier. Fully 10 percent of banks have greater unrecognized losses, on a percentage basis, than SVB. The failed bank was an outlier in terms of its reliance on uninsured deposits—in which category it landed in the top 1%. Thus, SVB was particularly vulnerable to a run.
Matvos et al relied on prices of traded funds (such as ETFs) to estimate mark-to-market losses. They did not, of course, examine every shopping center loan, nor do they claim that their numbers are precise.
But precision is hardly needed; the point is, the gap between the book value of banks and the marked-to-market value of assets represents a problem. Commercial loans cannot be easily marked to market, but securities, such as bonds, can be.
After the Savings & Loan crisis in the 1980s, similar to the SVB affair in the key respect of being driven by inflation-cum-soaring interest rates, it was recognized that delayed recognition of losses, permitted by accounting practice, played a key role in the crisis. Lawrence White, a member of the Federal Home Loan Bank Board, wrote in 1991, “The revamping of this accounting framework—a switch to market value accounting—is the single most important policy reform that must be accomplished.”
Ensuing legislation failed to make it happen. The Financial Accounting Standards Board, whose rules are generally incorporated into regulatory standards for bank capital, similarly blinked.
Instead, for purposes of calculating capital, banks are allowed to separate securities into two buckets, those that are “available for sale,” which must be marked down on the balance sheet, and those they intend to “hold to maturity.”
When a bond held to maturity suffers a market loss, the loss is not deducted from equity, although the figure is disclosed. In other words, the balance sheet professes to asset values that are inflated compared to market values.
The benefit of marking down assets every quarter is that risky or reckless behavior is incrementally discouraged. As reported results worsen, market pressures force an adjustment. In the 1980s, recognition was delayed. When the S&L crisis was finally admitted to, it cost taxpayers $100 billion. In something of a parallel, when SVB was forced to sell securities and recognize a loss it triggered the run, finis.
Matvos, the paper’s co-author, says raising capital standards is a simpler and better fix than tinkering with accounting rules. I think both remedies are warranted.
There is a growing chorus for switching to mark-to-market accounting, possibly through legislation. But absent action from Congress, it will not happen soon.
Jack Ciesielski, a longtime accounting observer, says marking to market is a hot discussion topic, however, he dispiritingly adds, “There is no proposal on the table” at FASB. Institutional morass is one problem; regulatory capture is another. The regulators like the status quo (delayed recognition) because the banks like it.
Ciesielski does not like it. “It [marking to market] is reality. The securities are worth just what the market says they are. Banks may say they have the intent and the ability to hold until maturity but as we’ve seen – everyone has a plan until they get punched in the mouth. SVB got punched in the mouth by the Fed.”