Commentary: Northern Rock: Solutions and Problems

Introduction

September brought the United Kingdom its first mass market bank run. Although there had been bank runs in the nineteenth century, very few people then had bank accounts so the runs were not in any sense comparable with what happened to Northern Rock, which showed a failure of mass retail banking.

The crisis had its roots in sub-prime lending in the United States. Sub-prime mortgages, like other mortgages, were converted into traded securities with complex structures. European banks bought these securities in the belief that they were diversifying their portfolios but without any real understanding of the risks that they were running. The crisis broke on 9 August, when Paribas reported that it was impossible for it to value some of the mortgage-related securities which it owned. Since many banks were believed to have invested in similar securities, banks felt that they did not know enough about the solvency of other banks to which they might lend. As a consequence lending on the inter-bank market stopped.

This meant that banks with wholesale deposits falling due were unable to renew them in the wholesale markets. They therefore looked to central banks to provide the liquidity which was needed to replace that which was no longer available in wholesale markets. When it became public that Northern Rock was receiving help from the Bank of England, depositors sensibly concluded that their savings were at greater risk there than elsewhere, leading to the run on Northern Rock which ended only when the Chancellor of the Exchequer announced a government guarantee for their deposits.

Regulatory issues

The Governor and the Bank of England has drawn attention to the sorts of changes which would prevent individual banks being affected by runs. Raising the extent to which bank deposits are guaranteed would help. So too could liquidity insurance. In the run on Northern Rock people withdrew their deposits to place them with other banks. They did not, for the most part, ask for cash. Had Northern Rock held liquidity insurance with other banks, then the insurance policies might have required the banks to make longterm loans to Northern Rock, providing them with the funds that they needed to cope with the loss of deposits. Liquidity insurance would, in effect, recycle deposits from the other banks to Northern Rock. In a sense these insurance contracts would oblige the other banks to do what they did not do during the crisis, to lend money wholesale to Northern Rock.

How well would this work in practice? For the system to operate presumably all banks would be obliged to insure each other. It is not very likely that banks could insure with bodies outside the banking system Ð because non-banks do not typically have the liquidity needed to provide the resources required in a banking run. If banks pay each other insurance premia then, unless some banks can be identified as more prone to liquidity problems than others, the revenue earned by each bank from providing insurance to other banks might be expected more or less to match what it pays for its own insurance. This has the result that liquidity insurance becomes little more than a mutual aid arrangement Ð an agreement between banks to lend to each other on occasions when the inter-bank market dries up. Such an arrangement would be more likely to work in practice if no public announcement were made when liquidity facilities were needed.

Nevertheless recent experience suggests that such an arrangement will be set up only if it is compulsory and it is a shortcoming of the Financial Services Authority that they had not thought of this in advance. On the other hand there is an issue as to how such insurance would operate if there were doubts about the solvency of a bank which wanted to draw on its liquidity insurance/mutual aid. The risk of losses on the loans that banks might face on loans made under a mutual aid scheme could be insured with bodies outside the banking system. Bodies which sold such insurance polices would in essence be agreeing to provide extra capital to banks in difficulties.

There is some suggestion that small banks are more in need of liquidity insurance than are large banks. If this is the case then the cost of such insurance will presumably be offset against the fleetness of foot and otherwise lower costs that small banks may benefit from. But since bank runs are very rare it is not clear on what sort of basis it is possible to say that small banks are more prone to such problems than are large banks.

Finally we note that the problems at Northern Rock were triggered because of the public announcement that the Bank of England was providing aid. It is suggested that the Market Abuse Directive prevented the Bank from providing aid discreetly because Northern Rock was a publicly traded company. Separately the Stock Exchange Rules would required Northern Rock to make public that they had received help.

It was this publicity which provoked the bank run. Given current rules there may be little that could have been done about this, although we note that similar problems at IKB (Deutsche Industriebank) were handled more discreetly despite the fact that it is quoted on the German stock exchange. In any case there is a real question to be asked why the Bank of England was not aware of the risk that the rules pose for lender of last resort activities and had not developed a framework which allowed it to issue Public Interest Immunity Certificates to allow people to depart from the rules when they would put lender of last resort activities at risk. They need to address this shortcoming of the current institutional framework.