The Northern Rock Bank Run: Britain's First Bank Run in 150 Years (2007)

A Queue That Changed Everything

On the morning of 14 September 2007, something happened on British high streets that no living person had ever seen. Outside branches of Northern Rock — a respectable, FTSE 100 mortgage lender based in Newcastle — ordinary savers formed queues stretching around the block. Some had arrived before dawn. They carried newspapers, thermoses, folding chairs. Television cameras captured the scenes and broadcast them around the world. It was, financially speaking, a moment from another century.

It was. Britain had not witnessed a bank run since Overend, Gurney and Company collapsed in 1866 — the panic that had prompted Walter Bagehot to write Lombard Street, his foundational text on central banking. One hundred and forty-one years later, the same spectacle was playing out in digital-age Britain, live on the BBC, with an audience of millions watching on the internet as well as on television.

The run itself lasted days. By the end of it, roughly £2 billion had been withdrawn. But the run was almost the least important part of the story. What Northern Rock revealed — about wholesale banking, about deposit insurance, about the relationship between liquidity and solvency — reverberated through every subsequent crisis decision of the next twelve months.

The Machine That Printed Mortgages

Understanding the collapse requires understanding the business model. Northern Rock had been a building society, a mutual institution owned by its members, until it demutualised in 1997. Demutualisation gave it access to capital markets, and it used that access aggressively.

Traditional mortgage lenders raise money by attracting retail deposits — ordinary savers who put money in current accounts and savings accounts. The lender then lends that money out as mortgages at a higher rate, pocketing the spread. The model is slow to grow but inherently stable: retail deposits are sticky, and depositors generally don't all ask for their money back at once.

Northern Rock chose a different path. Rather than waiting to accumulate retail deposits, it funded its mortgage book primarily through wholesale money markets and securitisation. The bank would originate mortgages, bundle them into securities under a programme called Granite, and sell those securities to investors. In the meantime, it borrowed short-term funds from other banks and institutional investors to keep new lending flowing. This "originate and distribute" model allowed Northern Rock to grow at extraordinary speed. Between 1998 and 2007, its mortgage book expanded from around £15 billion to over £100 billion. By 2007, it had become the fifth-largest mortgage lender in the United Kingdom.

The risks were visible to anyone who looked. By 2007, approximately 40% of Northern Rock's funding came from securitisation and a further significant portion from other wholesale sources. Its retail deposit base was thin relative to the size of its loan book. As long as money markets remained open and securitisation markets remained liquid, the engine ran. If either market seized up, Northern Rock would have no way to roll over its short-term borrowings or fund new lending — and it would face collapse not because its mortgages were going bad, but because it simply could not fund itself.

August 2007: The Freeze

The trigger arrived from three thousand miles away. Through late 2006 and into 2007, the US subprime mortgage market had been deteriorating. House prices had peaked. Delinquency rates were rising. Investors who held subprime mortgage-backed securities were taking losses. On 9 August 2007, the French bank BNP Paribas suspended withdrawals from three investment funds, saying it could no longer value their subprime holdings because "the complete evaporation of liquidity in certain market segments of the US securitisation market" had made it impossible. That phrase — the complete evaporation of liquidity — became the defining description of what happened next.

Interbank lending dried up almost overnight. Banks did not know which counterparties held subprime exposure, so they stopped lending to each other. The London Interbank Offered Rate — LIBOR — spiked. The commercial paper market, which institutions used to fund themselves over periods of days and weeks, effectively closed for any issuer with any connection to structured products. Northern Rock had extensive such connections, through Granite and through its wholesale borrowing.

Over the following weeks, Northern Rock's management worked frantically to find alternatives. They approached potential acquirers. They held discussions with other banks. Nothing materialised. By early September 2007, the board concluded that the bank could not survive without emergency central bank support (Shin, 2009).

On 13 September 2007, Northern Rock formally approached the Bank of England for emergency liquidity assistance — invoking, for the first time in living memory, the role of lender of last resort that had been enshrined in British financial doctrine since Bagehot wrote about it in 1873.

The Leak, the Queue, and £1 Billion in a Day

The Bank of England's support, when agreed, was supposed to remain confidential until an announcement could be made in an orderly fashion. That plan lasted approximately hours.

That evening, the BBC's business editor Robert Peston broadcast the news that Northern Rock had sought emergency support. The report reached millions of people before any official statement was ready. In the age of rolling news and early internet, containment was impossible.

Savers face a brutal game-theoretic logic in such situations. If the bank survives, those who queue and withdraw lose nothing except time. If the bank fails, those who queued first and got their money out are the winners; those who waited and trusted are the losers. When the downside risk of not queuing is potentially catastrophic and the cost of queuing is merely inconvenience, the rational choice is to queue. The Financial Services Compensation Scheme — the UK's deposit guarantee — covered only the first £2,000 in full and 90% of the next £33,000, leaving anyone with more than £35,000 at risk of real loss (Shin, 2009). For savers with substantial deposits, the incentive to withdraw was overwhelming.

On 14 September 2007, approximately £1 billion was withdrawn from Northern Rock in a single day. It was Britain's first bank run in 141 years.

Mervyn King and the Moral Hazard Problem

The Bank of England's initial response was shaped by a principled but politically explosive argument. Governor Mervyn King was a committed believer in what economists call "moral hazard" — the idea that rescuing institutions from the consequences of their own risk-taking teaches them that risk-taking carries no penalty, thereby encouraging future recklessness.

King's position, which he articulated to the Treasury Select Committee in September 2007, was that a generous and unconditional bailout of Northern Rock's depositors would send exactly the wrong signal to other banks. "The provision of large liquidity facilities penalises those who bear the cost of insuring against liquidity risk and benefits those who took risks but were not insured," he told MPs. It was a textbook statement of the moral hazard concern, and it was not wrong as economic theory.

What it underestimated was the speed and ferocity with which an uncontained run could spread. By the time King's position became publicly known, queues were already forming. The political pressure on Chancellor Alistair Darling was intense. Darling later described the days after the BBC leak as among the most difficult of his political career. "I was being asked to make decisions that would affect millions of people, in real time, with incomplete information," he wrote in his memoir (Darling, 2011).

On 17 September 2007 — three days after the run began — the government announced that it would guarantee all deposits at Northern Rock. Not just the first £35,000, but all of them. The queues stopped. The run was over. But the damage to confidence was not.

The Numbers Behind the Crisis

The scale of Northern Rock's structural fragility is best understood through data. By the time the run occurred, the bank's funding profile was almost uniquely vulnerable among British institutions.

A conventional bank of comparable size would have relied on retail deposits for 50–70% of its funding. Northern Rock's retail deposit ratio was the lowest in the British mortgage sector (Shin, 2009). When wholesale markets closed, the bank had almost no stable funding base to sustain it even for weeks.

The government's guarantee announced on 17 September stopped the run but did not resolve the underlying problem. Northern Rock still needed to refinance its wholesale borrowings, and those markets remained effectively closed. Private sale negotiations — with Lloyds TSB and later with a consortium led by Virgin Money — failed to produce an acceptable deal. A Virgin Money bid valued the bank at approximately £1.3 billion, a fraction of its book value.

On 22 February 2008, Chancellor Darling announced that Northern Rock would be taken into temporary public ownership. It was the first nationalisation of a British bank since the nationalisation of the Bank of England itself in 1946. At its peak, the taxpayer's exposure — through loans, guarantees, and capital injections — exceeded £100 billion. Eventually most of that was recovered as the bank's mortgage book was wound down, but the experience was a shock to a political class that had assumed the era of bank nationalisations was long past.

The Canary in the Coal Mine

Northern Rock did not cause the 2008 financial crisis. But it prefigured it almost perfectly, and it provided the clearest possible early warning that the global financial system was built on assumptions that could not survive a shock to wholesale funding markets.

The bank's "originate and distribute" model — originating mortgages and packaging them into securities for sale rather than holding them on the balance sheet — was not unique to Northern Rock. It was the dominant model in US mortgage finance and was spreading through European banking. The assumption that securitisation markets would always be liquid, that short-term wholesale funding would always be available to roll over, that no major bank would face a funding crisis because someone would always lend — these assumptions were shared by Citigroup, UBS, Merrill Lynch, and dozens of other institutions that would face their own crises in the following fourteen months.

The collapse of Long-Term Capital Management in 1998 had demonstrated that liquidity could evaporate suddenly in stressed markets. The Northern Rock run made the same point about the entire model of wholesale-funded banking. Neither lesson was learned quickly enough.

What distinguished Northern Rock from later failures was the specific visibility of the retail run. When Lehman Brothers failed in September 2008, there were no queues outside offices — the creditors were institutions, not individuals with savings accounts. The Northern Rock run was viscerally comprehensible to ordinary people in a way that the subsequent collapse of the interbank market was not. That visibility made it politically impossible to ignore, which is partly why the government response — guarantee, then nationalisation — moved faster than it might otherwise have done.

What Changed: The Regulatory Aftermath

The Northern Rock crisis exposed three specific failures of regulation that were subsequently addressed — imperfectly, but substantively.

First, deposit insurance. The £35,000 limit, with its 90% coinsurance provision for balances between £2,000 and £35,000, was designed as a backstop against occasional bank failures. It was not designed to prevent runs. A guarantee structure that leaves savers partially exposed creates a strong incentive to withdraw at the first sign of trouble. The Financial Services Compensation Scheme limit was raised to £50,000 in October 2008 and subsequently to £85,000, reflecting European Union requirements. Full protection was extended for temporary high balances — proceeds from property sales, insurance payouts — to address the specific concern about depositors facing large but transient balances.

Second, the resolution framework. Before 2009, there was no legal mechanism in the UK to take a failing bank into public ownership quickly without triggering a formal insolvency process. The Banking Act 2009 created a Special Resolution Regime, giving regulators powers to transfer a failing institution's business to a private-sector purchaser, to a bridge bank, or to public ownership — without the chaos of a full insolvency. It also introduced a bank insolvency procedure specifically designed to pay out insured depositors quickly.

Third, and most consequentially in global terms, liquidity regulation. Basel II, the international capital framework in force at the time of Northern Rock's collapse, said relatively little about liquidity. Banks were required to hold capital against credit risk; they were not required to hold liquid assets sufficient to survive a prolonged closure of funding markets. Basel III, agreed in 2010 and phased in over the following decade, introduced the Liquidity Coverage Ratio — a requirement for banks to hold enough high-quality liquid assets to cover net cash outflows over a 30-day stress scenario — and the Net Stable Funding Ratio, which required banks to fund long-term assets with long-term liabilities. Had these requirements been in place in 2007, Northern Rock as it existed could not have operated.

The Ghost in the Machine

The deeper lesson of Northern Rock is about the gap between apparent solvency and actual stability. At the moment the run began, Northern Rock was not technically insolvent. Its mortgages were performing. Its capital ratios met regulatory requirements. By the tests that existed, it was a sound institution.

What it was not, was liquidity-resilient. It had traded structural safety for speed of growth, and had done so in a regulatory environment that rewarded capital adequacy while essentially ignoring funding structure. The institution that Bagehot had described in 1873 — the lender of last resort that would lend freely against good collateral at penalty rates — stepped in exactly as the doctrine prescribed. And the institution it was lending to, by Bagehot's own test, was solvent but illiquid: exactly the case his doctrine was designed to handle.

But the world had changed since 1873. Northern Rock was not a small provincial bank whose depositors could be reassured quietly by a Bank of England official. It was a national lender with millions of retail customers and a twenty-four-hour news ecosystem that could transmit panic faster than any official statement. The lender of last resort doctrine assumed a world in which information moved slowly. The twenty-first century had made that assumption obsolete.

Just as Barings Bank's collapse in 1995 exposed the inadequacy of controls over derivative exposures, Northern Rock exposed the inadequacy of liquidity regulation in an era of wholesale-funded banking. Both crises prompted reforms. Both reforms came too late to prevent the next, larger catastrophe.

Alistair Darling, writing after the event, captured the essential quality of what happened: "Northern Rock was the warning that the financial system chose not to hear" (Darling, 2011).

The queues dispersed. The branches reopened. The cameras moved on. Eleven months later, Lehman Brothers failed, and the warning became the catastrophe it had been trying to announce.