The failures of Silicon Valley Bank, Credit Suisse Group AG and others have gotten financial authorities thinking again about bank runs and liquidity regulations — and whether some rules ought to be tweaked to make the system safer.
Mostly, though, the debate misses the point. The only way to guarantee an end to bank runs is to get rid of banks. That isn’t as flippant as it sounds: It has been one of the ideas often proposed since the 2008 financial crisis, under the name of narrow banking. It’s a idea I don’t buy, as I’ll explain.
But first, the liquidity rules. Liquidity coverage ratios are meant to ensure a bank has enough cash-like assets to cover a drain on its funding in a stressed situation over a period of 30 days. It involves estimating outflows from depositors, lenders, borrowers and counterparties to derivatives trades among other things. The cash-like assets a bank holds are the “fractional reserve” in the classic understanding of banks, as the financial blogger Frances Coppola put it in a smart definition recently.
For example, under the international Basel standards, a bank must be able to pay out at least 3% of retail deposits if they are fully insured, that insurance is credible and the deposits are in accounts linked with other services or where salaries are paid. These conditions are not easy to meet entirely and in practice the lowest assumed outflow rates are 10%. For rich clients with large deposits, the minimum is higher and for large corporates it’s at least 40%, again with conditions attached.
Different countries’ regulators often impose higher outflow expectations than these minimums. Maybe they could lift them a touch further, but this is all shifting proverbial deckchairs. Liquidity rules only protect banks against fluctuations in customers’ needs for liquidity caused by economic or financial stresses outside a bank. Bank runs are totally different beasts: A run is a panic that — whether or not it had a rational cause — will almost certainly be ruinous. And typically happens much faster than 30 days!
A bank run is possible as long as there are some deposits that will require a bank to sell longer-term or risky assets quickly in order to pay them. That always leaves a chance of being the last person trying to cash out and finding the bank comes up short. The only way to guarantee no run is to force banks to match all deposits with assets that have cash-like qualities: For instant-access deposits, that means central bank reserves and very short-term government bills. Such a bank is a so-called narrow bank — in reality no bank at all. It’s little more than a utility for making payments – it makes no loans and creates no money for the economy.
The main way to help prevent runs has been deposit insurance. To strengthen that, covered limits could be increased. The flood of money pumped into the US economy after the 2008 financial crisis and during the Covid pandemic led uninsured deposits to more than treble in value between 2007 and 2021, Travis Hill, vice-chairman of the Federal Deposit Insurance Corporation, said in a recent speech. To remain effective, deposit insurance could be indexed to inflation or money supply growth.
A bad idea would be to get governments to insure all deposits. Putting taxpayers explicitly on the hook for all this risk would require much stricter regulation of what banks can do with their funding. In my view, the logical destination of that road is again narrow banks. We might as well make all our payments through – and hold all our deposits at – the central bank. (If central banks create their own cryptocurrencies, maybe that’s what we’ll end up doing.)
If all banks are narrow banks, how does anyone borrow money? The answer is something like bond funds: Credit managers raise a bunch of money in some form and lend it out at various terms and rates of interest. Of course, credit funds are also vulnerable to runs. To make them more stable, they could be grouped by duration, so that the lifespan of the loans they make are similar to the withdrawal terms set for investors.
But there’s an obvious problem with this setup: It is very inflexible. People and companies have to be fairly certain of their own future spending needs to lock up a good share of their money in longer-term funds. In contrast to credit funds, banks can create loans on demand — they don’t need to prefund them, they just need to ensure they attract enough funding by the end of each day. In a narrow bank world, the supply of credit could be more constrained.
Meanwhile, people like bank deposits because they give each of us the option to suddenly change our spending patters – to meet an unexpected cost, for example. That optionality has a value, which we pay for by accepting low interest rates on the most flexible deposits. As a group, people and companies tend to spend deposits in steady, predictable ways, but any individual is free to behave differently.
Deposits work for much the same reason that insurance works: If people’s needs aren’t too closely correlated, they can always be met, just like it is extremely unlikely that everyone’s house burns down at the same time. However, you can only get cash from your insurer if your house actually burns, whereas if you are worried a bank might fail to make future payments on your behalf, you can go straight to one that looks more reliable.
Looked at this way, another defense against bank runs could be to force each bank to attract diverse depositors and so reduce the chances that they all do the same thing at the same time. That would have made a difference for SVB and Credit Suisse.
Like many other things in finance, depositor liquidity is mostly a useful fiction. It is an option with value, but it only works so long as most of us don’t use it. A bank run occurs when that fiction is exposed and no incremental changes to depositor insurance, diversity, or liquidity coverage ratios can fix that. As long as we have banks we will have bank runs – and they can only really be stopped by a government stepping in.
Bloomberg News provided this article. For more articles like this please visit
bloomberg.com.
More Insurance & Annuities Topics >
