The Economics Book

320 BANK RUNS

In September, 2007, the first

serious British bank run

since 1866 took place. Northern

Rock, Britain’s eighth-largest

bank, was a fast-growing

mortgage lender. To expand

its business, it had become

over-reliant on “wholesale”

funding—funding provided

by other institutions—rather

than personal deposits. When

wholesale financial markets

froze on August 9, 2007, a

gradual, unseen wholesale

A modern bank run

run began. At 8:30 p.m. on

Thursday, September 13, BBC

Television News reported that

the UK central bank, the Bank

of England, would announce

emergency liquidity support

the next day.

It emerged later that Mervyn

King, the Governor of the Bank

of England, had opposed a

rescue offer by Lloyds, another

British bank. King had

suggested that central bank

support might reassure

depositors. However, this

reassurance did not happen,

and a run on personal deposits

began over the internet that

evening. Under Britain’s deposit

insurance program, deposits

above $3,300 (£2,000) were not

fully insured, and the next day,

long lines formed outside

Northern Rock branches. The

run ended the following

Monday evening after the

government announced a

guarantee for all deposits.

Diamond and Dybvig showed that
this property also makes the bank
vulnerable to a run. A run occurs
when, on Tuesday, patient people
become pessimistic about what
they will receive from the bank on
Wednesday, and so withdraw their
deposits on Tuesday. Their actions
mean that the bank must sell
investments at a loss; it will not
have the resources to pay all of its
patient and impatient customers,
and those later in the line will not
receive anything. Knowing this,

customers become eager to be at

the front of the line.

Pessimism can arise out of

concerns about investments, other

people’s withdrawals, or the bank’s

survival. Crucially, this allows for

the possibility of a self-fulfilling

bank run even if the bank is sound.

For instance, suppose that on

Tuesday I believe that other people

are going to withdraw their

deposits—I then decide to do so as

well because I fear that the bank

may fail. Then suppose that many

other people think in the same way

that I have. This itself can cause

a run on the bank, even if the bank

would otherwise be able to meet its

obligations today and tomorrow.

This is an example of what

economists call “multiple

equilibrims”—more than one

outcome. Here there are two

outcomes: a “good” one in which the

bank survives and a “bad” one in

which it is sunk by a run. Where we

end up may depend on the people’s

beliefs and expectations rather than

the true health of the bank.

Preventing bank runs

Diamond and Dybvig showed how

governments could alleviate the

problem of bank runs. Their model

was partly a defense of the US’s

system of federal deposit insurance,

under which the state guarantees

the value of all bank deposits up

to a specified amount. Introduced

in 1933, this system reduced bank

failures. In March, 1933, President

Franklin D. Roosevelt also declared

A panicking crowd is held back

by police outside a German bank

in 1914. The declaration of war had

caused pessimism among savers,

leading to a number of bank runs.