SVB’s breakdown suggests a degree of muddle among regulators and bank executives.
The collapse of Silicon Valley Bank illustrates not so much a failure of regulation as of regulators. The distinction points to the difficulty of the task that lies ahead: Simply reworking the rules about capital and liquidity requirements won’t be sufficient. What will be necessary is a rethinking of the very notion of risk.
SVB’s breakdown suggests a degree of muddle among regulators and bank executives that’s hard to square with their undoubted dedication and expertise. Only people with years of training and experience, professionals who understand modern finance in minute detail, could fail to grasp what seems obvious to the naïve onlooker — that when the value of an asset goes down, its owner is less wealthy.
SVB owned a lot of fixed-income assets (Treasury bonds and mortgage-backed securities), so when interest rates went up, the price of those assets dropped. In other words, the bank lost money. Experts argue that it’s a bit more complicated: Losses aren’t realized unless assets are sold. Most likely they won’t be sold. Held to maturity, they’ll repay the principal.
In effect, managers and regulators agreed to make that assumption, not just for SVB but for banks in general — then concluded, what’s the problem?
Reports of heavy “unrealised losses” seem to be what started the SVB scare. The bank’s depositors evidently didn’t understand that rising interest rates meant such losses were to be expected and weren’t cause for alarm.
Just to be sure, the mistaken depositors pulled their money out. They had a lot of very big (hence uninsured) accounts, and were an unusually networked and tightly knit group. The run forced the bank to sell its assets, paper losses became actual losses, and technical insolvency became actual insolvency.
Out of nowhere, a middle-sized bank — heavily invested in safe, liquid assets — became an officially recognised systemic threat, raising questions about the safety of many other banks, convulsing global markets, and demanding instant and wide-ranging official intervention in the US and abroad.
A first line of analysis blamed the administration of Donald Trump for excusing banks of SVB’s size from the full force of post-crash rules on capital and liquidity — an appealing explanation that puts the blame in a satisfying place and suggests a straightforward remedy. But two things turned out to be wrong with it. First, as my colleague Paul J. Davies pointed out, US regulators retained discretion to hold SVB and other such banks to the tougher standards; they chose not to exercise it. Second, few paused to wonder whether SVB might in fact have met the stricter requirements if challenged. According to one study, it would have.
The post-crash rules aren’t blind to the perils of interest-rate risk and associated unrealised losses, but they might as well be. The Liquidity Coverage Ratio asks whether a bank has enough “high-quality liquid assets” to pay back deposits at times of severe stress. Another metric, the Net Stable Funding Ratio, weighs the composition of the banks’ assets (loans and investments) and liabilities (deposits and other borrowings) to see whether they’re suitably aligned for liquidity purposes over longer periods. But according to these methodologies a bank like SVB that invests in supposedly safe, liquid securities (as opposed to making loans that are hard to sell if need be) and that relies heavily for its funding on presumptively sticky deposits (as opposed to volatile wholesale borrowing) is unlikely to raise a red flag.
The crux of the matter is that regulators and managers recognised interest-rate risk — what happens to asset prices when central banks move to quell inflation — but never took it seriously. The post-crash mindset was preoccupied with credit risk (bad loans and dumb investments); it wasn’t that bothered about maturity mismatch and unrealised losses, because interest rates obviously weren’t going anywhere.
Under the post-crash Basel III rules, the argument about how to measure bank capital (equity and other funding capable of absorbing losses) gives a telling illustration.
Banks designate their securities as “available for sale” or “held to maturity.” For the first, the difference between market value and acquisition price is written to the regulated bank’s balance sheet, and unrealised losses are deemed to deplete capital. But unrealised losses on held-to-maturity securities aren’t counted. As this different treatment of the two categories was phased in between 2014 and 2018, banksshifted securities from available-for-sale to hold-to-maturity to stop fluctuations in asset prices affecting their measured capital. That is, they chose to ignore, and were allowed to ignore, the implications of interest-rate risk for capital adequacy.
Once the immediate crisis has passed, the Federal Reserve and other regulators have their work cut out. They’ll need to revisit the regulatory distinction between big and small banks, now they’ve affirmed that small banks can cause systemic risk; work out how to weigh unrealised losses in judging how much capital is enough; admit that safe, liquid assets such as long-term default-free government bonds are neither safe nor, in a certain sense, liquid. Most fundamentally, they’ll have to start questioning obvious truths that turn out to be false and grapple with unheeded fragility hiding in plain sight.
There’ll be a bunch of patches, for certain. But it’s looking increasingly likely that they won’t be enough. Despite the difficulties, not to mention the intense resistance they’ll arouse, more radical remedies will be needed to avoid the next SVB.
Clive Crook is a Bloomberg Opinion columnist and member of the editorial board covering economics, finance and politics. Views are personal and do not represent the stand of this publication.