In September 2007, following Ia number of news reports concerning the international liquidity difficulties generated by the US sub-prime problems, the UK experienced the first sustained run on a major retail bank for over 100 years.
Although the causes and consequences of the run on Northern Rock have been explored elsewhere, this paper provides the first systematic account drawn from the perspective of NR’s customers at that time.
Over the years NR had pursued an aggressive programme of saving account expansion, gaining market share primarily through its internet offerings.
This was a cost-effective strategy given that the branch structure of NR had a strong high street concentration in the North and North-East, but a weaker presence elsewhere – in total it had 72 branches with only four in London.
Furthermore, many branches had only a couple of counters and due to the requirements of the legislation associated with potential money laundering, large withdrawals could normally take up to 15 minutes to be completed.
During August and September 2007, NR began to encounter serious liquidity problems caused by its over-reliance on access to wholesale money markets to finance its business. When those markets froze in August 2007, NR’s management entered into secret negotiations with the regulator and the Bank of England to secure access to emergency funding to help meet its daily financial obligations.
The intention was to make a carefully worded announcement to the markets on Monday September 17. But a leak of this planned announcement to the BBC on September 13 provided it with an exclusive report and it broadcast the news across all media.
These then are the key facts which eventually led to the run on the limited branch network that started on the weekend of September 14 2007.
Through the conduct of an extensive three-phase research programme, combining qualitative and quantitative web-based data collection, we have focussed upon the factors that influenced consumer behaviour during the short time period from September 13 through to, effectively, the end of the run one week later.
In particular we have tested the hypothesis that, in the first instance, customers were simply seeking information from NR’s website rather than looking to withdraw funds. The failure to obtain that information caused customers to be concerned about their ability to withdraw their deposits, leading to queues at branches with the full-scale run on the bank that followed.
Thus the paper also establishes customers’ opinions as to the role the media played, particularly the BBC, whose business editor Robert Peston first revealed that NR had applied for emergency financial support from the BoE.
Furthermore, we identify custo-mers’ perceptions of the relative res-ponsibility for the situation they faced, whether they consider it to have been primarily due to the international background (US sub-prime), the various UK authorities (BoE, the Treasury, the Financial Services Authority), NR’s management or the media.
Finally, customers’ attitudes to events have been placed within the context of their understanding of their exposure to financial risk and their understanding of the Financial Services Compensation Scheme as it applied at the time of the run on NR.
The paper closes with NR’s customers’ expectations from any future compensation scheme, including the role of the various responsible authorities and the requirements for communicating information.
A short history of NR
The origins of the bank go back to the middle of the 19th century, with the foundation of the Northern Counties Building Society in 1850 and shortly thereafter the Rock Building Society (originally formed in 1865). Both were mutual societies, known in the US and elsewhere as savings and loan organisations, owned by their members and not by shareholders, whose activities were regulated by law.
NR itself was formed in 1965 from a merger of these two societies and it remained a mutual operation through to becoming demutualised in 1997.
During the period from the initial merger to its formation as a bank, the society had continued to expand by taking over a number of other local mutual societies, most notably the North of England Building Society in 1994.
The process of demutualisation and becoming a member of the UK Stock Exchange involved share offers at favourable rates to all members (both savers and mortgage holders), as well as the injection of funds from a number of City institutions and interested parties overseas.
As a result, the bank had an unusually large number of small shareholders, primarily based in the north of England.
The bank expanded aggressively over the following 10 years, with ex- tensive growth in its customer base beyond the confines of its relatively small branch network.
Within three years of joining the stock exchange, NR was promoted to the Financial Times Stock Exchange Index of the 100 largest publicly quoted companies in the UK. New customers were encouraged to communicate with the bank through a variety of means, using the post, telephone call centres and, in particular, the internet via online banking.
Its success was frequently remarked upon in the financial press and its shares were often tipped as a good buy. For example:
“Northern Rock’s strong interims last week provided an excellent start to the bank reporting season … record first- half net mortgage lending of £3.8bn, representing a market share of about 9.5% – just over twice its natural market share – forecast another good year for mortgage volumes. Buy at 689p” (Sunday Express, July 2003).
As a result of the company’s strong performance and favourable press comments, its share price climbed steadily over the years, almost doubling in value from July 2003 to a peak of £12.40 on February 6 2007, remaining at over £10 per share through to June 20 last year.
With such a strong share performance and the quality image that the bank had attracted in financial circles, NR’s management was able to be innovative in employing sources of funding other than that from savers and investors and became dependent upon interbank funding to cover shortfalls in operational cash flow.
While wholesale funding to NR grew markedly, there was no correspondingly rapid growth in its retail funding. Between 1997 and 2006, retail deposits grew from £9.9bn to £22.6bn, whereas there was a six-fold growth in assets over the same period.
Retail deposits and funds as a proportion of liabilities and equity fell from 62.7% to 22.4% over the same period, which marked the bank out as a significant outlier when compared with some of its key competitors.
In the first half of 2007, NR continued to expand its business at a rapid rate. In that period, its loans to customers underwent a net increase of £10.7bn. It thus became particularly vulnerable to the international squeeze in liquidity caused by the US sub-prime crisis, a risk which finally crystallised on August 9 2007.
Events of September 13 to 19 2007
NR’s share price decline, combined with the liquidity squeeze affecting all banks across Europe and the US, caused its management to approach the BoE with a request for a special loan to cover a seemingly temporary shortfall in liquidity. This fact was first conveyed to the public at 8.30pm onThursday September 13 when BBC business editor Robert Peston broke the news that NR was in the process of requesting emergency financial support from the BoE.
In fact, the governing body of the BoE, known as The Court, was meeting to approve the request at the time when the BBC was broadcasting the story on its 24-hour news channel.
The story was then put out on all channels, news programmes and other platforms. In particular, the BBC led that evening’s 10 o’clock news bulletin with its exclusive. It was the lead item on the BBC’s website and radio bulletins. Peston also featured the item on his blog, Peston’s Picks.
Indeed, that Thursday evening, his was the only voice on the story. No spokesperson for NR, no government minister nor any other official was prepared to go on the record or appear in a broadcast. The following morning of Friday September 14, the Treasury and BoE confirmed Peston’s story and the government then launched a media campaign in an attempt to counter the fallout.
Nevertheless, many NR customers who had been used to its online banking service sought to find out more from its website, including some who wished to access their accounts to check that everything was OK.
Unfortunately, the website was unable to meet demand and the news media went on to convey that fact without any explanation from NR’s management, compounding the belief that something could be wrong with clients’ accounts.
This led to many customers att-empting to gain information and access to their accounts through other means, including visiting their nearest branch. Queues quickly formed at many branches, a fact also reported by the media from the Saturday through to the following week.
Headlines such as ‘Police called to break-up Northern Rock panic queues as customers withdraw millions’ and press comment that “the bank also had to close down its website, which could not cope with the flood of savers trying to withdraw their money” contributed to the widespread belief that the bank was insolvent.
Meanwhile, officials from the BoE and NR as well as representatives from the FSA remained off-camera until after the run had been staunched. Although Adam Applegarth, then chief executive officer of NR, is reported to have said on Saturday September 15 that “if I was a depositor – and I am – I would be reassured if the BoE was behind me”.
Subsequently, Applegarth reported to the House of Commons Treasury Select Committee that “one of the things we had intended to do over that weekend was to widen the bandwidth on the internet account, so you would not have had so much frustration from our internet customers”.
These were the key events leading to the run on the bank and the collapse in NR’s share price from a peak of over £12 in February 2007 to just 132p by October 1 2007.
The bank run
The research evidence confirms customers’ awareness of the above details and even though the events surrounding the run on NR occurred eight months prior to the study being undertaken, only 5% of respondents could not recall how they found out about the problems. Just over half of respondents (54%) found out via broadcasts on BBC news and 13% found out via the radio.
A further 10% found out by reading a newspaper. No respondents said they found out from NR management or from any official source (government, BoE or FSA). Word of mouth (friends/ relatives) was a relatively insignificant source at 10%.
The general impression portrayed by the media was that on finding out about the problems facing NR, the vast majority of savers panicked and attempted to withdraw all or some of their savings.
While there is clear evidence that large numbers of savers did seek to withdraw some savings, in fact the initial reaction of the majority when they first heard of the problems was different from the image portrayed by the BBC and others.
Half of all savers said that, initially, they did nothing in particular. A fifth (21%) tried to access the NR website and just over one in 10 (12%) tried to get information from other websites.
In addition to the 21% who did try NR’s website, a further 10% would have done so but had heard it was down. Furthermore, only 20% of those respondents who did try the NR website had intended to withdraw funds, far less than the final proportion of around 50% of customers who subsequently went on to make a withdrawal.
All the others were seeking some form of reassurance, with 58% claiming that they wished to access the website to find out whether NR was stable and secure and a further 18% saying they were seeking general information.
Indeed, even for those respondents who used online banking to manage their savings accounts, the proportion intending to withdraw funds at that time was only 24%.
These results support the hypothesis that the lack of any official reassurances to obtain information about the stability of NR was a critical factor leading to what became a run on the bank.
Ultimately, despite the run, just half of all respondents to the main survey who had savings with NR on September 13 had withdrawn some or all of their money, either at the time of the run or subsequently.
Nevertheless, those respondents with higher value savings were more inclined to have withdrawn some, so that 74% of those who had more than £35,000 in savings had withdrawn all or part of their savings at the time of being interviewed.
For those who have withdrawn any savings, the majority (55%) did so through their online banking facility, with 27% having visited a branch. The remainder did so by post (11%) and by telephone (7%).
We have, therefore, drawn the following initial observations in respect of the bank run:
a) Initially, the majority of customers were more concerned to obtain information rather than necessarily withdraw any money.
b) The media coverage, combined with a lack of any official communications, increased concern among NR customers.
c) The non-availability of NR’s web-site for almost four days after the first news report was a major factor in the initial loss of confidence for many investors.
d) The guarantee provided by the government when publicised did assuage the fears of many of those who had less money saved. But despite that assurance many customers still removed their money, even among those whose savings were now covered by the guarantee.
Who did the customers consider was primarily responsible?
Although many observers consider the US sub-prime crisis to have been a principal cause of NR’s difficulties, that possibility was not a major consideration for customers.
Instead, their concerns were primarily domestic and were almost entirely related to the bank’s manage- ment, various authorities or the media, especially the BBC.
As one saver (who was also a shareholder) wrote in an online focus group, “and then good old Robert Peston – shouting from the rooftop that NR was up the creek without a paddle”.
Another said, “you’re right about Robert Peston. I still think that the media is to blame for a lot of it”.
Another saver (who did not have shares) said, “I felt totally confident my money was safe. To be honest I thought it was only going to be a minor problem and was being blown out of proportion by the press”.
Peston was, indeed, conscious of the manner in which a number of individual savers and others had criticised him for being in some way responsible for the run.
When both the chairman and the chief executive of NR claimed before the Treasury Select Committee that they believed the leak to him had exacerbated the woes of the bank, he wrote a spirited defence of his journalism in his blog of October 16 2007.
But the media was not alone in receiving criticism from ordinary customers. Comments from NR’s share- holder saver focus group included “the government could have acted more effectively earlier”, “I was annoyed at the lack of management response” and “the ineffectiveness of the regulators. It seems like we have masses of regulations and regulators but they miss the big stuff”.
Of course, the primary responsible regulator was the FSA, which ultimately admitted a number of errors in the report of its own internal review, published on March 26 this year.
Among other observations made in that report, it admitted that no senior regulator had met with NR between January 2005 and August 2007.
Finally, of course, there is the matter of the management of NR, of whom it was once said that “the directors of NR are the most hated people in Britain”. We have seen above that both the chairman and the CEO did comment that the BBC and the press had exacerbated the bank’s difficulties.
Ultimately, however, Applegarth may be judged as being somewhat complacent with his initial remarks on September 14 2007 that the crisis is “a short-term liquidity problem” and that it was a “global issue” affecting all lenders.
However true that comment may be, it was never effectively understood by those customers directly affected.
By the time the research was conducted, during the intervening period following the run there had been considerable press comment on the various issues and much consideration as to what should happen to NR.
In the end, the fact that the government guaranteed £50bn of taxpayers’ money led to the bank being taken over by the government. Not only was such intervention considered politically taboo in the current economic age, it also raised critical questions in Europe about state aid.
Nevertheless, the nationalisation of NR was undertaken after lengthy consideration of other alternatives such as the sale of the bank to other interested parties, one being a consortium including Sir Richard Branson.
Seemingly, none of these alternative solutions were guaranteed to repay the government within a reasonable time and, therefore, the bank was nationalised on March 31 2008, just prior to the start of our research programme.
During the period from the time of the original run through to the date of the nationalisation, there had been a number of reports identifying one or other of the above institutions as having a significant responsibility for the problems. They included:
a) Mervyn King, governor of the BoE, who was blamed by Jon Wood, chairman of the SMR hedge fund, a leading shareholder of the bank.
b) Gordon Brown, the Prime Minister was, by implication, blamed by King when he claimed to have requested special legislation from the government which could have averted the crisis while Brown was still chancellor (i.e. prior to July last year).
c) Similarly, King suggested that Alistair Darling, chancellor of the Exchequer, had missed an opportunity for Lloyd’s Bank to take over NR.
That, then, is the detailed background to the main survey concerning the respondents’ opinions of the various parties responsible for the situation in respect of NR.
When asked to assess each of the responsible parties on a scale of 0 to 10, respondents did not, on balance, consider any of them to have served them well. Not one achieved even an average rating score.
The BoE did best with a score of just 4.1 out of 10, whereas NR’s management obtained the lowest score of 2.7. The FSA was only slightly ahead at 3.
The media and the government both achieved the relatively respectable score of 3.8.
Although ratings for each of the responsible parties were relatively uniform across each of the various subgroups in the sample, there were some interesting differences to be observed – most notably:
– The media achieved a much higher rating (4.2) among those who had removed all their savings but only a rating of 3.5 from those who had kept their savings with NR.
This fact is clearly associated with the manner in which those with higher savings were more likely to have withdrawn their funds and, accordingly, appreciated the warnings received from the media.
– Those who had maintained their savings within NR gave a higher rating to all parties other than the media (see table top).
– Savers who were also shareholders gave lower ratings to each of the parties than non-shareholders did. We attribute this to the fact that, presently, they have lost the value of their shares – although, at the time of writing, shareholders are taking action against the government to recover some of their value.
– In fact, the media has its lowest score (2.6) among those savers who are also holders of NR mortgages.
However, these relatively small differences in the uniformly poor ratings shown in the above tables conceal the fact that, overwhelmingly, respondents blamed NR’s management above any of the other parties.
Almost twice as many respondents (65%) blamed the management for subsequent events, compared with just 35% blaming all the other institutions put together.
Nevertheless, there were some important differences across the key behavioural subgroups, most signifi- cantly as follows:
– 10% of all savers who did not withdraw their money blamed the media, whereas only 4% of those who withdrew all their savings did the same (see table bottom).
– 16% of those respondents who also hold a NR mortgage blamed the BoE (see table on facing page).
The others blamed in the above tables include some references to the background represented by the sub-prime problems, as well as a variety of other causes – including 2% of respondents who, overall, felt that no-one was to blame.
Our conclusions from the above information endorse the previous view that communication from all parties has been poor – not only at the time of the run but subsequently.
Furthermore, the manner in which the BoE came out so poorly among NR mortgage holders may have implications for it in respect of the wider audience of all mortgage holders, a con- stituency that represents over 35% of the UK’s adult population.
What are customers’ perceptions of financial risk and the implications for the regulatory authorities?
The qualitative work provided considerable background information in respect of NR’s customers’ attitudes to financial risk-taking, the most fundamental difference being between those savers who were also shareholders and those who weren’t.
The former could be classified as mild to medium risk-takers with a few extremes since for the most part they understood that some risk is inevitable but are cautious as to how much they should accept.
For example: “I’m good with my finances. Having worked for a bank for five years I have a good understanding of what good and bad attitudes look like. I’m willing to take a punt on an investment if I think I will make a good return. Sometimes it works, other times it does not.“
Another said,”I’m a little bit cautious but prepared to take a risk with some of my investment provided I understand what the risk is”.
Non-shareholders are more risk-averse, seemingly less confident in their ability to manage their finances, thus, “My attitude to risk is fairly cautious, having had a couple of really bad experiences with stocks and shares based products“.
Another said, “I also agree about advisers. Have tried twice – the first time we were oversold and it was over-risky. Second time we were more cautious and decided not to proceed as we didn’t feel we could trust someone else, especially after the previous occasion“.
The main survey therefore examined the issue of respondents’ attitudes to risk and to other aspects of savings and investments to understand the degree to which this may have influenced their behaviour.
The results were, however, not as significant in respect of their subsequent behaviour as were the facts of the ownership of products and the amounts of money saved, as previously reported.
For example, a key issue that emerged during the run on NR focussed on clients’ awareness of the level of protection offered to ordinary retail savings accounts.
Many of our respondents with retail deposits were simply unaware that they would not receive their deposits in full beyond £2,000, with only 90% of funds being reimbursed between £2,000 and £35,000.
For example: “The FSA makes sure that banks, building societies, insurance companies and other financial institutions hold enough funds to pay out to customers in normal circumstances.”
Some 73% believed this to be the case and 27% ‘definitely’ knew they did so. A further 45% thought that the FSA must do this.
Indeed, a further 13% believed that ‘some organisation’ was responsible for ensuring that their savings were fully covered.
These high confidence results remained true across all sub-groups – even those depositors with higher sums at risk.
Meanwhile, the FSA conducted its own research on an annual basis for a number of years covering a sample representing all UK consumers.
Focussing primarily on the aspect that measures the FSA’s prudential functions of ensuring that financial institutions have qualified staff and enough money to operate, it is not surprising that the regulator’s research shows a higher level of confidence than that shown by our research.
For example, only 2% of our respondents considered that all financial institutions in the UK are adequately staffed and funded compared with 12% for the FSA’s research undertaken this year.
More surprising is the fact that the FSA’s 2008 survey indicates that many more consumers now feel that most firms reach these standards – up by almost 40% on the corresponding 2007 figure. Even for our own respondents, our survey figure is 20% higher than the 2007 value for the total population.
Nevertheless, our sample of NR depositors (all of whom are now aware of the FSA) shows rather more scepticism concerning the regulator’s overall effectiveness – with a net confidence score of -12% compared with a net confidence score of +52% for the FSA’s survey based upon the 33% of all re- spondents who were aware of the FSA as a regulator.
It is also of interest to note that for the respondents to our own survey, there is little significant difference in the levels of confidence between those NR customers who withdrew money and those who did not.
Subsequent to the events of last September, the role and effectiveness of the FSA as a regulator has been closely scrutinised by the authorities and the level of protection and compensation afforded to depositors has also been reviewed, with the result that in July 2008 new levels of compensation were proposed.
Such changes in compensation are, however, unlikely to satisfy all the concerns of shareholders, a significant number of whom had expected to be able to grow their investments in NR.
The crisis had come as a surprise to them – examples from the qualitative work include “I am more cautious and cynical about bankers and the government”, “My attitudes have not changed a lot really – but I will try to steer clear of anything involving the government” and finally, “I wouldn’t say that my attitude has changed much, I still have a savings account with NR. Even though NR was a major risk for an investment, I still went ahead with it and would do so again“.
Overall, the evidence clearly de-monstrates the enormously negative impact that a run on a retail bank has on the confidence of its depositors and others in the regulation of the financial system.
Even eight months after the event those confidence levels do not appear to have recovered.
Furthermore, NR depositors are evidently now more aware of the social costs of potential failure, whereas others still tend to believe that there exist special companies and government safety nets protecting such firms.
In our opinion, the above information also calls into question the interpretation of the FSA’s research. Prima facie it shows a relatively high level of satisfaction with the FSA and other regulatory bodies, whereas a different conclusion may be that the public continues to show a high level of ignorance as to the facts of regulation and compensation.
We draw just three primary conclusions from NR’s difficulties as evidenced by the research undertaken, which are as follows:
1. The growth of internet banking and the failure to effectively manage NR’s website was a primary factor in the bank run.
Thus we confirm and even extend observations made by John Sviokla and Kevin McGilloway of Harvard Business School on January 14 2008 in their piece How technology amplified the mortgage crisis.
Compared with previous financial crises when firms have failed, the events of 2007 are characterised by the speed with which consumers demanded information and expected access to it via the internet.
The reliance of so many consumers on the internet for information and for access to their accounts appears to have been an oversight.
We can only speculate on whether during NR’s rapid transition from a provincial mutual to a thrusting and innovative player in the global financial services marketplace, its management had overlooked the changing character of its underlying customer base.
2. With the development of 24/7 media, NR may be seen as only one of many examples where the events in one country can quickly have consequences for consumers elsewhere.
It is evident from the research that communication from every one of the responsible parties is considered to be poor.
The absence of any official communications once the leak had occurred clearly demonstrates how those responsible for managing major corporations and their regulation can be caught on the back foot.
The failure of the tripartite authority to co-ordinate communications with consumers at speed, once the leak had occurred, is also apparent.
This conclusion is supported by the findings of the Treasury Select Committee:
“In view of the role that fears of a leak of a support operation had played in the decision on Tuesday September 11 that a covert operation was not possible, the tripartite authority was unwise initially to accede to Northern Rock’s request for the announcement of the support operation to be delayed until Monday September 17.
“In the light of subsequent events, it seems evident that the tripartite authority and NR ought to have strained every sinew to finalise the support operation and announce it within hours rather than days of the decision to proceed with the operation.”
3. The failure of a mid to large-sized financial services institution and the subsequent need for government intervention on a major scale has had a hugely negative impact on the confidence of those depositors who were customers at the time the bank failed in regulation.
This lack of confidence contrasts markedly with financial services consumers more generally.
It is, perhaps, significant that in these modern times of customer-led developments, the research conducted by the authors has been the only case of consultation and review with customers since the run on the back itself.