LONDON — Governments should consider an international moratorium on the development of central bank digital currencies (CBDCs) due to the financial stability risk they could pose for small jurisdictions, the former director of research at the Federal Reserve Bank of New York, Stephen Cecchetti, told central bankers in Finland on Thursday.
“I believe there should be something equivalent to… a CBDC treaty — like the Nuclear Test Ban treaty — across the world where everyone agrees not to issue CBDCs,” said the American economist, who is now Rosen Chair in International Finance at the Brandeis International Business School. “Because as soon as a large jurisdiction does, there will be a rush to do this and as soon as that happens, you will see the financial systems and the monetary sovereignty of small, unstable countries just get demolished.”
The comments came in the broader context of a speech that focused on how to make banks safer in the aftermath of the bout of financial instability that gripped markets in March.
As regulators grapple with how to reform the system, some radical propositions are making a comeback. Among them is the idea of moving the financial system towards a narrow banking model, where the deposit-taking and payment activities of banks are separated from their financial intermediation roles. Under this system, there would be no need for depositors ever to ‘run’ on a bank out of fear of losing their money.
The structure, which is very similar to a full-reserve system, would prevent deposit-taking banks from lending to the private sector. But critics worry this would constrain credit creation and move risk elsewhere in the system, not least onto government balance sheets.
“CBDCs are a form of narrow banks,” Cecchetti told the conference. “I am not a fan of narrow banking. I don’t think it fixes the problem… we don’t have narrow banks because they are expensive.”
The economist said he preferred the idea of moving to the so-called “pawnbroker for all seasons” regime championed by former Bank of England governor Mervyn King. This would see banks pledge collateral to the central bank for every dollar’s worth of deposit liabilities. The quality of the collateral would then be assessed on a regular basis, with the central bank demanding a discount — known in the industry as a haircut — representing the level of capital it believed was at risk. The differential would have to be funded through bank equity or debt.
If the system went to such a regime it would, in Cecchetti’s mind, completely substitute the need for deposit insurance. “You don’t need it at all,” he said, adding that the haircuts the central banks set would drive credit allocation in your system.
There is an argument, he added, that the Eurosystem is already doing this, and pricing the risk “via some machine sitting in the bowels of the Banque de France.” For the system to work, however, it would have to be applied to the whole financial system. “Nobody has ever really tried to do this. It might be worth it.”
Banks must be allowed to fail
For now, what is clear is that the current regulatory framework is wanting: neither depositors — even those who knew their accounts were too big to be covered by Federal deposit insurance — nor the stock market were alert to the risks at Silicon Valley Bank and Signature as the unrealized losses on their securities holdings stacked up.
And yet, these issues were well known to banking analysts, Cecchetti said. “They were computing these numbers and publishing them in the junk that you get in your email when you’re on their mailing lists,” he said. “Some of this spam is worth opening it turns out. The problem is which stuff?”
Fed officials too were warned. A quarterly briefing to the Federal Reserve Board in February, cited in a harshly-worded review of the failings at SVB by the Fed’s top banking supervisor Michael Barr, noted such unrealized losses exceeded 50 percent of the capital of 722 banks by the third quarter of 2022, while another 31 banks had negative tangible equity.
“They weren’t getting shut down and the question is why not,” said Cecchetti, highlighting that equity prices of failing banks were inexplicably high until two days before they failed. The reason, he said, could be because accounting rules may have vastly inflated regulatory capital.
His advice to supervisors from now on? To look for risk in the parts of a bank’s balance sheet that are making the largest amount of money.
“I don’t believe in geniuses,” Cecchetti said. “What I do believe in is people hiding risk, so I need to figure out where people are hiding it, which is to study the places making a lot of money.”
In that context, Cecchetti poured cold water on institutional practice that sees banks treating deposits as a hedge against large losses on securities — a formula known as “deposit beta”.
“This deposit beta business, ask yourself whether or not if, in the worst circumstance, the hedge really works,” he said. “You want a hedge that works in the bad state of the world? This one doesn’t… that seems to me to be wrong-way risk, not a hedge. “
Ultimately, he said, the system needs to shift to a regime where bank assets are continuously valued, at least on a quarterly basis, against market prices — although some in the audience pointed out that could be difficult when it came to more complex and less liquid assets.
One Response