Before you start
What you'll be able to answer
- How did the Bank of England buy government bonds, and who did it buy them from?
- What kind of money did the Bank pay with, and why could banks not lend that money out?
- How was the Bank's buying of gilts (government bonds) different from buying them straight from the government?
Where this sits
Quantitative Easing · this module is lit
- 1998The Bank of England Act gives interest rate decisions to a committee at the Bank; the Chancellor keeps the target
- Mar 2009Bank Rate is cut to 0.5 per cent, and the Bank starts buying government bonds with newly created money
- Aug 2013The Bank says it will not consider raising interest rates at least until unemployment falls to 7 per cent
- Mar 2020Bank Rate is cut to a new low
- Jan 2022Banks' reserves at the Bank reach their highest level
- Sep 2022The Bank buys government bonds again, to stop forced selling by pension funds
- Oct 2022Consumer price inflation reaches 11.1 per cent
- Nov 2022The Bank starts selling its bonds to investors, among the first central banks to do so
- Aug 2023Bank Rate reaches its peak after rises at consecutive meetings
- Apr 2026The Governor writes an open letter to the Chancellor after inflation overshoots the target
The Bank began buying government bonds with money it had created
In 2009 the Bank of England began buying government bonds, known as gilts. It paid for them with money it had just created. Paying with new money can sound like printing banknotes and handing them to the government to spend. Three details made the real process different: who did the buying, who did the selling, and what kind of money was paid.
When the Bank of England bought gilts from 2009, who did it buy them from?
Investors that already held the gilts, such as pension funds and insurers.
A separate company did the buying
A company owned by the Bank of England, the Asset Purchase Facility, bought the gilts. Early in 2009 the facility was buying small amounts of company debt. The Chancellor, the minister in charge of the government's finances, then gave permission for gilt buying. The Bank's Monetary Policy Committee, the group that sets interest rates, could now use the facility to buy gilts with newly created money.
The gilts were listed in the facility's accounts, not directly in the Bank of England's own accounts. The Treasury, the government's finance department, agreed to cover any losses the facility made. So any losses would fall on the government's finances, and the Bank would not bear them.
The gilts were bought second-hand from investors
The government borrows by selling gilts to investors. Each gilt is a promise to pay interest every year and to repay the loan on a set date. Once sold, gilts are traded second-hand between investors, in the secondary market.
The facility bought gilts only in the secondary market. It aimed to buy mainly from investors outside the banks, such as pension funds and insurance companies. A pension fund that sold gilts was likely to use the money to buy other assets, such as company shares and bonds. When pension funds bought shares and bonds, the prices of those shares and bonds would rise. Higher prices for company bonds mean lower interest rates on them. So companies could borrow more cheaply. Cheaper borrowing should lead companies to borrow and spend more.
A pension fund sells gilts to the facility. What does the pension fund receive for the gilts?
The pension fund's own bank adds the money to the pension fund's account.
The Bank paid with reserves instead of banknotes
Paying in banknotes would have meant handling huge quantities of notes, so the payment was electronic. Reserves are money that commercial banks hold in accounts at the Bank of England. Banks use reserves to make payments to each other.
The facility paid for the gilts. It borrowed the money from the Bank of England. The Bank created that money as new reserves. The reserves went to the pension fund's bank, because a pension fund has no account at the Bank of England. The pension fund's bank then added the same amount to the pension fund's account, as a new deposit (a bank balance). So each purchase created new reserves for a bank and a new deposit for the seller.
Reserves rose steeply while cash in circulation did not
The chart shows the reserves banks held at the Bank of England, as a monthly average. In February 2009, just before the purchases began, reserves averaged about £39 billion. By December 2013 they averaged about £299 billion.
Notes and coin in circulation are the cash people use in shops. Cash rose only from about £52 billion in February 2009 to about £69 billion in December 2013.
Source: Bank of England Database, sterling reserve balance liabilities, monthly average (LPMBL22). Open Government Licence v3.0.
Check yourself
Work from the two reserve figures above, for February 2009 and December 2013. About how many times their February 2009 level were banks' reserves by December 2013?
7.7
Nearly eight times: divide the December 2013 figure by the February 2009 one.
A bank now holds an extra £100 million of reserves from the purchases. What can it do with them?
Reserves can only be paid to others with accounts at the Bank of England.
Banks could not lend the reserves out
Households and firms do not have accounts at the Bank of England. So reserves can only pass from one bank to another. When a bank makes a loan, it adds the loan amount to the borrower's account as a new deposit. A bank decides how much to lend by asking how much it can lend profitably and safely. A bank's reserves do not limit how much it lends.
Even in 2009 the Monetary Policy Committee did not expect the extra reserves to bring much extra lending. Banks were under pressure to reduce their total loans and other assets. So they were unlikely to lend much more, even with extra reserves. The new money that reached the rest of the economy was the new deposits of the sellers, such as pension funds. The sellers used these deposits to buy other assets, such as shares and bonds.
Check yourself
When the facility bought gilts from a pension fund, what grew as a result?
Banks' spare reserves did not push interest rates down
Normally a central bank sets the interest rate it pays on reserves. It then creates as many reserves as banks ask for at that rate. In quantitative easing (QE), the gilt buying described above, the Committee chose an amount first: how many gilts to buy. The reserves to pay for them followed.
Bank Rate is the interest the Bank of England pays on reserves left with it overnight. Banks ended up with far more reserves than they needed. If they lent the spare reserves to each other, many banks would want to lend and few would need to borrow. So the interest rate banks charge each other would fall. But a bank could earn Bank Rate just by leaving reserves at the Bank of England. So the rate between banks stayed close to Bank Rate, however many reserves existed.
Check yourself
Three of these were part of how quantitative easing worked in Britain. Which one was not?
Suppose a central bank bought gilts directly from the government when they were first sold. It paid with money it had created. What would that be the same as?
The government would be spending money the central bank had created for it.
Buying second-hand was not monetary financing, on the usual definition
Monetary financing means paying for government spending with newly created money, instead of borrowing from investors. The Bank bought gilts second-hand, from investors who had already lent to the government. The new money went to those investors. So on the usual definition, the purchases were not monetary financing.
Some disagree. In 2021 some witnesses told a House of Lords committee that the Bank's 2020 gilt buying had much the same result as financing the government directly. Each gilt pays a set amount of interest, so a higher price means a smaller return. The Bank's buying pushed gilt prices up, so the government could pay less interest on new gilts. The Governor, the head of the Bank, said it had stayed within its powers. He said the amount bought was set by what would best keep inflation at the Bank's target. The amount the government borrowed did not decide it.
Check yourself
The key questions
How did the Bank of England buy government bonds, and who did it buy them from?
The Bank bought gilts through a company it owns, the Asset Purchase Facility. The facility bought them second-hand from investors such as pension funds and insurers. It never bought gilts from the government when they were first sold.
What kind of money did the Bank pay with, and why could banks not lend that money out?
With new reserves, created electronically and held by banks in their accounts at the Bank, not with banknotes. Households and firms have no accounts at the Bank, so reserves can only move between banks.
How was the Bank's buying of gilts (government bonds) different from buying them straight from the government?
No, by the usual definition of monetary financing. The gilts were bought from investors after the government had sold them. Critics said the 2020 purchases had much the same effect, because the buying lowered what the government paid to borrow.
The numbers
Check yourself
Who sold gilts to the Bank of England's Asset Purchase Facility?
Check yourself
What did the Bank of England create to pay for the gilts?
Check yourself
By the usual definition of monetary financing, was buying gilts second-hand the same as paying for the government?