10 Years after Northern Rock – is the UK more or less likely to see another bank run?

Stephen Clarke

(Northern Rock image – Lee Jordan – Flickr, reproduced from wikimedia commons under CCA licence)

Ten years ago this month, queues of people started to form early in the morning outside Northern Rock branches across the UK, to withdraw their money out of fear that their bank would soon collapse.  As the day wore on panic spread, and the run continued until when the government stepped in to guarantee all Northern Rock deposits. It was the UK’s first retail bank run since the 19th century and one of the first symptoms of the global financial crisis.  This anniversary is an appropriate time to reflect on those events, but also to look forward and assess how things have moved on in the last decade, and whether something similar could ever happen again.

I won’t go over the underlying liquidity issues at Northern Rock here, although it is worth noting that the events did not follow the traditional model of a retail bank run leading to a shortage of liquidity.  Rather, public knowledge of the bank’s existing liquidity shortage (initially arising from wholesale funding issues) acted as the trigger for a retail run.

Nor will I consider the probability of a bank run at any specific bank today.  Or how changes to capital regulation may have reduced the likelihood that insolvency threatens a bank.  Rather I’ll consider some of the general changes over the last ten years that might influence either the likelihood of a bank run starting, or the way in which a bank run takes place, under four broad headings: new media, new banking platforms, deposit insurance, and changes to liquidity regulation.

1. New Media: a breeding ground for panic?

An integral element of the classic bank run scenario is a sense of panic among depositors based on their expectation that *other* depositors have withdrawn or will soon withdraw their cash.  For this panic to take hold, some new information (not necessarily correct) must make its way to depositors.

In the case of Northern Rock, there was a perception  (disputed by journalists) that many of the depositors queueing outside branches on the Friday morning were there because they saw the previous night’s TV news reports that the Bank of England was providing a liquidity support facility to Northern Rock.

Now imagine if a rumour of Bank of England support for a bank started circulating today.  In 2007, Twitter was a year old and hashtag tracking had not yet started.  Facebook had 50 million users, which sounds like a lot.  But today, Facebook has 2 billion users, and in the time it takes you to read this sentence, 50,000 tweets will have been sent.  There is an opportunity for that rumour to “go viral” much more quickly in today’s world of 24/7 social media and non-mainstream news sources, so it is certainly conceivable that bad news would spread more quickly today than it did in 2007.  This could impact both the likelihood of a bank run and the severity of a run if it does occur.  But social media, as well as spreading messages that encourage depositors to run, could be used to stem outflows by disseminating accurate information on the true state of the bank, in a measured and controlled way.  The retail bank, and possibly the authorities, might consider whether social media can be used to help to counteract the bank run.  This assumes, of course, that the bank’s financial position is fundamentally sound.  The authorities might even analyse the use of social media to help predict bank runs before they happen.

Overall effect:   Rumours can be spread through new channels, but panics can also be countered through those channels.

2. New Banking Platforms: virtual bank runs?

The traditional image many people have of a bank run is of a crowd of depositors rushing to a physical counter to turn their deposits into cash while they still can – like the classic scene in It’s a Wonderful Life.  Indeed, the television pictures of the Northern Rock run looked a lot like that famous scene (albeit with much more orderly British queueing).

There is another way to take money out though, which is to move it from one bank account to another.  Because no physical cash need change hands, moving money between banks need not take place at a branch, or during office hours – it can be done over the phone or online, day or night.

While the large UK banks, including Northern Rock, had a significant online presence in 2007, compared to today’s world of ubiquitous smartphone and app banking, online banking was in its infancy.  A run can now take place via a swarm of smartphone swipes, with no physical barriers to slow it down.

And Faster Payments technology means that customers will be able to access their money instantly once they move it, rather than wait for nearly a week under the BACS system.  A UK retail bank could even experience a run without anyone leaving their homes.  On the other hand, a virtual bank run does not make for very newsworthy photographs, and so the strategy of joining a queue because the length of queue is growing is less likely to take hold.

A bank might choose to switch off online banking altogether to stop a run, but this is a nuclear option that would negatively impact customers, reflect poorly on the bank, and probably drive customers to alternative methods of withdrawals anyway.

Overall effect:  Hard to say – the increased availability of online and mobile banking might increase the speed at which a run takes place, but it is harder for panic to take hold in a virtual run.

3. Deposit Insurance: a prevention and a cure?

In 2007, the Financial Services Compensation Scheme fully covered the first £2,000 of each depositor’s deposits, but the system of co-insurance meant that only 90% of the next £33,000 was covered.  In the case of Northern Rock it was only the Chancellor’s announcement that the government would fully guarantee all deposits that finally slowed the run.  Since then, the FSCS has been expanded so that £85,000 of each retail depositor’s cash at each institution is fully insured, and FSCS works to build awareness of the scheme.  Of course its effectiveness is yet to be tested in a crisis situation: there has not been a bank run since FSCS was expanded.

Overall effect:  Greater deposit insurance makes retail runs less likely – the 2007 run slowed down and eventually stopped once full deposit protection was announced, and strong deposit protection is now in place as standard.

4. Higher liquidity requirements: red tape, or silver lining?

UK banks are required to hold more liquidity now than they did before the financial crisis.  Banks must, under Liquidity Coverage Ratio requirements, hold sufficient high quality liquid assets to cover outflows they might experience in a 30 day stress scenario.  And the PRA can require specific banks to hold additional amounts to guard against specific risks.  Partly because of these requirements, UK supervisors scrutinise banks’ liquidity very closely.  Other things being equal, higher amounts of business-as-usual liquidity and higher levels of supervisory activity should both reduce the chance of a run in the first place (by reassuring customers), and enable banks to better manage a run if it does occur.

Overall effect:  Increased liquidity requirements make runs less likely: banks have more chance of managing a run with their own resources, and the knowledge of this will reassure depositors.

Conclusion

While the trigger of a bank run will always be an unknown unknown, I believe a retail run is less likely in the UK now than it was in 2007.  Although changes in technology, both in the finance world and in the way people communicate generally, might create a small increase in risk, far more critical than these are the changes that have taken place to deposit insurance and to the regulatory framework.  One thing we can say with some confidence is that the announcement of full deposit insurance helped to stop outflows from Northern Rock.  The fact that strong deposit insurance is now in place for all UK banks as standard must give some comfort.  In addition, it is hard to dispute that higher liquidity requirements make banks better able to withstand a run.

Stephen Clarke works in the Bank’s Customer Banking Division.

If you want to get in touch, please email us at bankunderground@bankofengland.co.uk or leave a comment below.

Comments will only appear once approved by a moderator, and are only published where a full name is supplied.

Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.

2 Comments

Filed under Economic History, Financial Stability, Macroprudential Regulation, Microprudential Regulation

2 responses to “10 Years after Northern Rock – is the UK more or less likely to see another bank run?

  1. Interesting but too kind.
    – The capacity for ubiquitous and universal online banking to accelerate bank runs and make them more violent is not “hard to say”, it’s obvious and really quite scary.
    – The authorities’ soothing remarks were not reliable. eg. “We are satisfied that HBOS is a well-capitalised bank that continues to fund its business in a satisfactory way” (FSA statement, 17 September 2008)
    – Almost all of the Northern Rock depositors in those queues were well inside the deposit guarantee amount
    – The big change is the provision of massive, almost unlimited backstop liquidity support from central banks. For example, the ECB provided €233bn of four-year funding this year, in addition to existing facilities.

  2. Laurence Russell Sanders

    Firstly, it is very important to focus on the impact of technological change on financial stability. This paper is an encouraging sign that the Bank of England is focusing on a key business risk – I attach an even greater weight to technological risk than the author of this paper. My view reflects the rapid growth of internet and smart phone transactions, both in the UK and globally. Central banks should ensure that they have sufficient expertise in this important area – although not excessive resource, since this could impact adversely on the speed of decision making.

    As a Member of the Association of Corporate Treasurers who was working in the financial sector in 2007, I fully concur with Stephen Clark’s professional view that “a retail run is less likely in the UK now than it was in 2007 … reflecting changes that have taken place to deposit insurance and to the regulatory framework.” The one proviso is that the more rigorous regulatory regime should not give rise to complacency.

    An important element of financial oversight and control in any financial institution is a powerful financial risk committee – and for larger financial organisations, a committee in each jurisdiction and at group level. This should be a statutory prudential requirement. The group risk committee should ideally comprise executives and senior professionals with a high level of expertise in key areas including risk management, liquidity management, customer business (both retail and business finance), prudential regulation, and modern technology. Risk committees should be headed by the divisional / group treasurer or chief financial officer. Minutes of each monthly meeting should be made available to board members and to the Bank of England regulator. Key metrics of a professional risk management report are well known to members of the treasury management profession and to financial regulators. Each report should quantify key business risks, together with rigorous probability analysis.

    Fundamental reforms in prudential regulation have greatly reduced the risk of a repeat of the events of 2007. The UK must never return to: (1) a regime of light touch regulation (2) division of prudential responsibility between legally separate regulatory bodies (3) political pressure on regulators to accommodate an excessive expansion of financial sector lending.

    It is important to focus on global risks. Events that led to the collapse of Northern Rock had one key origin in the weaknesses in the US non-prime mortgage market that were clearly apparent in mid 2006. Global risks in the current era take many forms. Whatever the outcome of UK government negotiations with the EU27, the UK economy and financial system will become increasingly dependent upon the path of global events, especially in the more rapidly expanding developing economies.

    Laurence Russell Sanders

    Member, Association of Corporate Treasurers; Member, Society of Business Economists

    Independent Economic and Treasury Advisor

    22nd September 2017