Big Bang 2.0: should ring-fencing of banks be phased out?

It is now almost 15 years since a number of major UK financial institutions failed, which put millions of retail bank accounts at risk. This was part of a global financial crisis that sparked a wave of reforms around the world. 

The UK’s answer was to set up a ring-fencing regime that separated retail deposit-taking from other activities and treated them as being too vital to fail. This regime came into force in January 2019. 

Last year, the Skeoch Review suggested these reforms had made the banking system safer and that retail deposit-taking entities were now easier to supervise and resolve. However, it also warned the benefits “will likely diminish with time”, particularly as the Bank of England (BoE) develops the ‘resolution regime’ – an alternative way to deal with failing banks in which shareholders and some creditors take the hit.

So, is it time to replace ring-fencing? And is resolution the right option?

Repo rate reduction

New evidence has come to light which demonstrates that ring-fencing has been effective in de-risking banks and lowering the price they pay for liquidity.

In a paper I wrote with Irem Erten of Warwick Business School and Ioana Neamţu of the BoE, we looked at data on bank repo rates before and after the ring-fencing regime was implemented. These are the rates banks demand of each other in exchange for cash. They reflect how risky banks perceive each other to be. 

We found that ring-fenced banks enjoyed a special reduction in these borrowing costs when the regime was brought in: peers judged ring-fenced banks to have become less risky. This lowers their financing costs. 

We found that a banking group containing a ring-fenced subsidiary could borrow on average 0.86 basis points more cheaply. This is a roughly 2.5% reduction in borrowing costs. At the ring-fenced subsidiary level, the reduction in borrowing costs was approximately three times bigger. Further, in times of stress such as those experienced during the Covid-19 pandemic, this cost of borrowing reduction was enhanced.

Would being outside the ring-fence suggest to others that the BoE would be washing its hands of that part of the bank? My co-authors and I looked for evidence of this. But there was no equivalent rise in borrowing costs, and so no measurable damage, to the subsidiary outside the ring-fence.

Resolution regime a better option?

There are some disadvantages to ring-fencing, however. Ring-fencing was costly to implement and there is evidence that some deposit-taking institutions cap their growth in the UK to stay outside the regime. This stifles competitive expansion and could inhibit innovation. 

However, the untested resolution regime is not necessarily a better way of protecting retail savings, as big banks are intertwined with other financial market players. If you allow one domino to fall by pulling the trigger on resolution, it is impossible to know who else might be hit as the debts and write-downs cascade through the system. 

Nobody wants another incident like that with Lehman Brothers in 2008, but relying on resolution is something of an act of faith. This act of faith has taken a further knock from the collapse of Silicon Valley Bank (SVB) in the US and Credit Suisse in Switzerland. 

The resolution regime is designed to give the regulator the powers to bail-in debt and organise a banking rescue quickly enough to prevent collapse. But technology has long since made it possible for institutional depositors to withdraw money incredibly quickly. 

In the SVB case, depositors attempted to withdraw $42bn in one day. It was not possible for the US authorities to save SVB fast enough in the face of such a torrent of withdrawals. 

The second aspect of the resolution regime is the use of a priority regime for losses to be apportioned. Regulators plan to force equity to bear losses, and then some debt holders.

Many thought this standard ranking would apply to Credit Suisse, but when push came to shove, the losses were not apportioned as investors expected. The authorities determined that Credit Suisse equity holders should receive some protection while a class of debt holders was wiped out, reversing the standard ordering. The court cases have already begun. 

Both of these events are a warning that relying on resolution has not worked in other jurisdictions. We need to be clear why we think it would work in the UK.

A third path

In my view, there is another way to reduce the regulatory burden that ring-fencing has created while protecting deposits. 

The UK should overhaul its deposit insurance scheme and keep real-time information as to deposits protected. If a bank fails, depositors and small businesses should be automatically credited with their money in another regulated bank account. 

We now have the technology to do this by keeping a digital ledger of assets using blockchain technology. This would happen naturally if the UK introduced a central bank digital currency. Implemented properly, the first a depositor might know of a bank failing would be when they are told that their digital account has been credited with their savings. 

A financial regulator will never be able to match the speed with which depositors can run. But blockchain technology can. And with deposits protected for retail depositors and small businesses, much of the economy would feel secure. And so we would not need ring-fencing.

The UK government is keen to show that the country can be a leader in the financial sector. Harnessing the technologies of blockchain and digital currencies is one way this can be done.

Whatever system is put in place, any government will have to protect voters’ deposits. Memories of the global financial crisis may be fading, but there will always be a lot of political pressure to save voters’ savings and salaries from any banking collapse.

John Thanassoulis is professor of financial economics at Warwick Business School and course director for the Global Central Banking and Financial Regulation online course.