Monetary policy — interest rates and money supply — Economics

Before this

This leads to

Conflicts between objectives
How interest rate changes work · coming soon
Quantitative easing · coming soon

Nine people, one number

Eight times a year, nine people meet at the Bank of England in London. They study inflation, wages, spending and forecasts. Then each of them votes on one number: Bank Rate, the interest rate at the centre of the UK's money system.

Here is the puzzle. The Bank of England lends almost nothing to households or ordinary firms. Yet its vote changes what millions of families pay on their homes, and what firms pay to borrow. And sometimes the nine people make borrowing more expensive on purpose, knowing that it will slow the economy down.

How can one interest rate reach a whole economy? And why would anyone want to make borrowing dearer? The fiscal policy lecture ended by naming a second tool that works through a different channel: the price of borrowing. This lecture opens that channel up.

What monetary policy is

Monetary policy is the use of interest rates, and of the amount of money in the economy, to influence aggregate demand and so inflation. In most countries it is run by a central bank. In the United Kingdom it is the Bank of England's Monetary Policy Committee (MPC): the Governor, three Deputy Governors, the Chief Economist and four outside experts.

The government still sets the goal. Since December 2003 the target has been CPI inflation of 2%, as the measuring inflation lecture showed. If inflation moves more than 1 percentage point away from 2%, the Governor must write an open letter to the Chancellor explaining why, and what the MPC will do about it.

But the government does not choose the interest rate. The Chancellor handed that power to the Bank on 6 May 1997, and the Bank of England Act 1998 made it law. Why give it away? The fiscal policy lecture described the political business cycle: governments are tempted to boost the economy before elections. Cheap borrowing before a vote is the same temptation. And when people trust that inflation will stay near 2%, their expectations stay anchored, as the causes of inflation lecture showed. A central bank that is believed has to do less.

How one rate reaches every rate

Commercial banks keep accounts at the Bank of England, just as you keep an account at a commercial bank. The money in those accounts is called reserves. Bank Rate is the interest rate the Bank of England pays on reserves.

That small fact gives Bank Rate its power. A commercial bank can always earn Bank Rate, with no risk, by leaving money at the Bank of England. So it will not lend to another bank, a firm or a household for less than that. And competition between banks stops them charging very much more than they need to. So when Bank Rate rises, the rates on new mortgages, business loans and savings accounts rise with it. When it falls, they fall.

Where does the money supply fit in? The money supply is all the money people and firms can spend: notes, coins and, above all, bank deposits. Most of it is created by commercial banks. When a bank makes a loan, it adds the amount to the borrower's account, and that deposit is new money. So the money supply grows faster when people borrow more. Today the Bank of England does not fix the amount of money. It sets the price of borrowing, and the amount of money responds. Some central banks once tried to control the quantity of money instead. The real case below is one of them.

From Bank Rate to spending: the transmission mechanism

The deflation lecture said that cheaper borrowing encourages households and firms to spend, and promised that this lecture would explain how. The path from Bank Rate to prices is called the transmission mechanism. It has four main channels, shown below for a cut in Bank Rate.

The transmission mechanism. A change in Bank Rate moves market interest rates. These work through four channels at the same time, and all four change aggregate demand. Output responds first, and inflation later.

Borrowing and saving. Cheaper loans make it easier to buy a house, a car or a machine, so C and I rise. Households with variable-rate mortgages have more to spend each month. Saving earns less, so some people spend more.
House and share prices. When a bank account pays less, a house or a share that earns an income becomes more attractive, so its price tends to rise. Owners feel richer and spend more: the wealth effect from the causes of inflation lecture.
The exchange rate. Lower interest rates make the country's savings less attractive to foreign investors, so the currency tends to fall. Exports become cheaper abroad and imports dearer, so $X - M$ tends to rise. Other news often moves exchange rates more, so this channel is the least reliable.
Expectations and confidence. A cut signals that the central bank will support demand, which can encourage spending and investment. But if people read it as a sign of bad times, it can make them more careful.

A rise in Bank Rate works through the same four channels, in the opposite direction. Borrowing becomes dearer, saving pays more, asset prices tend to fall, and the currency tends to rise. Aggregate demand grows more slowly. Firms then find it harder to raise prices, workers ask for smaller pay rises, and inflation falls.

That is it. That is the whole idea. Bank Rate is the price of borrowing. Changing the price of borrowing changes how much people want to spend now rather than later. And spending decides how hard the economy is pushing against its capacity.

Why the Bank must look two years ahead

The transmission mechanism is slow. In the Bank of England's own model, a change in Bank Rate has its fullest effect on output after about one year. Its fullest effect on inflation comes after about two years (Bank of England, 1999). Building a factory takes time to plan, and shop prices change only once costs and demand have moved.

Mortgages show the delay most clearly. In 2022, about 86% of UK mortgages were on fixed interest rates (ONS, January 2023), usually fixed for two or five years. A rise in Bank Rate cannot touch a fixed payment until the deal ends. The ONS counted more than 1.4 million fixed deals ending during 2023. In July 2023 the Bank of England expected nearly 1 million households to be paying at least £500 a month more by the end of 2026. Those rates had started rising in December 2021. Use the model below to see how long the effect takes.

Real-world model — who feels a rate rise, and when?

An invented country has many identical mortgages of $200 000, repaid over 25 years at 2%. The central bank raises rates. Choose the size of the rise, the share of mortgages on fixed rates, and how long the fixed deals last. Watch how the share of households paying more grows month by month.

Rise in interest rates: 3 percentage points

Share of mortgages on fixed rates: 85%

Fixed deals last:

An invented model: identical mortgages of $200 000 over 25 years at 2%, and a rise of 3 percentage points. Variable-rate mortgages pay more at once. Fixed deals end evenly over their length, then pay the new rate. One mortgage's payment rises from about $848 to about $1 169 a month, about $321 more.

Months after the rise	Paying more (85% fixed for 2 years)	Paying more (85% fixed for 5 years)
0	15%	15%
12	57.5%	32%
24	100%	49%
60	100%	100%

With five-year deals, a year after the rise the average extra payment across all mortgages is only about $103 a month, a third of the full $321.

What should you have seen? With mostly fixed five-year deals, only about a third of households pay more after a year. So the MPC must set Bank Rate for the economy it expects in one or two years, not the one it sees today, and forecasts are often wrong. The model leaves a lot out. Real loans differ, and lenders often raise their rates before Bank Rate moves.

Stop and think

In August 2023 the MPC raised Bank Rate to 5.25%, even though inflation had already been falling for several months. Why raise rates when inflation is already falling?

Show the answer

Because of the lags. Inflation was still far above 2%, and the MPC was deciding about inflation one to two years later, not about the month it was in. Much of the earlier rises had not yet reached households, because so many were on fixed deals. The question was whether enough pressure was in the pipeline to bring inflation all the way back to 2%, and stop it there. Falling inflation is not the same as inflation at target.

A real case: the United States, 1979–82

Worked Example 1 — Policy, prediction, outcome, verdict

The situation. By 1979 the United States had lived with high inflation for most of a decade. Consumer prices rose by about 13% in 1979. Earlier attempts to bring inflation down had been abandoned. In 1973 the Federal Reserve, the US central bank, tightened policy, then eased again when unemployment rose, before inflation was under control (Federal Reserve History).

The policy. Paul Volcker became chair of the Federal Reserve in August 1979. On 6 October 1979 he announced a new approach. The Fed would stop steering its interest rate in small steps. Instead it would limit the growth of bank reserves, and so of money, and let interest rates go wherever that took them (Federal Reserve History; FRASER). The key interest rate reached a record of about 20% in late 1980, and came close to it again in 1981.

The prediction. Many doubted the Fed would hold on. A Treasury official told the Wall Street Journal in 1979 that the Fed would be forced to cut rates once unemployment reached 7% (Federal Reserve History). Economic theory predicted that the policy would work only if people believed it would continue.

What happened. Consumer price inflation peaked at about 14.8% in March 1980, then fell. On the Fed's measure it was 3.7% in 1983 (Federal Reserve History). But the cost was severe. The economy went through two recessions, and unemployment reached 10.8% in late 1982, the highest since the Great Depression of the 1930s.

The verdict. The policy achieved its aim. Inflation came down and stayed down, and the long period of steadier growth that followed is widely credited to it. The cost was millions of lost jobs and many failed firms. Some economists argue that a slower squeeze could have cost less. Others reply that the gentle approach had already been tried in the 1970s and had failed, because nobody believed it. A fair verdict: monetary policy can defeat high inflation, but when expectations are not anchored, the price is a deep recession.

Numbers to work with

Worked Example 2 — Is policy really tight? (UK, 2022 and 2024)

The effects of inflation lecture showed that savers and borrowers care about the real interest rate: $1 + real rate = \frac{1 + nominal rate}{1 + inflation rate}$ . The same idea tells you how tight monetary policy is.

In October 2022, CPI inflation reached 11.1% (ONS). Bank Rate had already risen seven times, to 2.25%. The real rate was $1.0225 \div 1.111 \approx 0.920$ , so about $- 8.0%$ . Borrowing was still very cheap in real terms. Rates had risen, but policy was still loose.

In May 2024, CPI inflation was back at 2.0% (ONS) and Bank Rate was 5.25%. The real rate was $1.0525 \div 1.020 \approx 1.032$ , so about $+ 3.2%$ . Now policy was clearly tight.

The lesson: judge policy by the real rate. A rising rate can still be a loose policy if inflation is higher still.

When Bank Rate cannot fall: quantitative easing

The deflation lecture showed that interest rates cannot go far below zero. It promised that this lecture would show what a central bank can still do. The answer is quantitative easing (QE).

In QE, the central bank creates new reserves and uses them to buy bonds, mostly government bonds (in the UK, called gilts). It buys them in the market, from investors, not from the government. More demand for bonds raises their prices. When a bond's price rises, the interest it pays, as a share of the price, falls. So QE lowers long-term interest rates, the ones that matter for long loans and big investments, even when Bank Rate is already near zero.

Bank of England researchers estimated that the first £200 billion of QE, completed by February 2010, may have lowered gilt interest rates by about 1 percentage point (Joyce and others, 2011). By the end of 2021 the Bank's purchases totalled £895 billion (House of Lords Library, 2021). Before you read on, test your picture of QE.

Predict — who receives the new money in QE?

The Bank of England creates new money and buys £50 billion of gilts. Choose the answer you think is right before you read any explanation.

Where the new money goes in QE, and why.

Most of it goes to the investors who sold the gilts, such as pension funds and insurance companies. The Bank of England designed its purchases so that the gilts came mainly from such non-bank investors (Bank of England, 2012). They end up with fewer gilts and more money in their bank accounts, and their banks hold more reserves. The government receives none of it directly, and households receive none of it as cash. The investors then look for other things to buy, such as company bonds and shares, which pushes those prices up and their interest rates down.

Now the question the fiscal policy lecture left open. During 2020 the Bank bought £450 billion more bonds. In the financial year 2020–21 the government borrowed about £311 billion (ONS). Did QE "pay for" the deficit? In 2021 a House of Lords committee warned that a widespread perception that QE was financing government borrowing could damage the Bank's ability to control inflation. The Bank replied that the evidence did not support that claim. The committee's chair later said it had not concluded that the Bank was financing the deficit. What is not in doubt is that QE made government borrowing cheaper. Whether that was its purpose or a side effect is the debated part.

What monetary policy cannot do

It cannot fix the cause of cost-push inflation. Higher interest rates cannot make gas cheaper. They can only reduce demand elsewhere, so that a price shock does not spread into a wage-price spiral, and that costs jobs.

It works with long and uncertain lags. By the time the full effect arrives, the economy may have changed.

It hits some people much harder than others. A rate rise hurts borrowers on variable rates at once, and helps savers. A rate cut does the reverse. Low rates and QE also raise house and share prices, which helps people who already own them. The Lords committee concluded that QE had widened wealth inequality. The Bank replied that QE had supported all borrowers at a time of extreme stress. Supporters add that protecting jobs helped people with little wealth. Both effects are real, and judging the balance between them is a matter of debate.

Trap 1 — "Higher interest rates bring prices down"

A rate rise aims to lower inflation, the speed at which prices rise. When UK inflation fell from 11.1% to 2.0%, prices did not fall. They kept rising, but more slowly. The price level stays higher. Only deflation would bring it down, and central banks try hard to avoid that.

Trap 2 — "The government sets interest rates"

In the UK, the government sets the 2% target, and the MPC chooses Bank Rate to meet it. Fiscal policy belongs to the government, and monetary policy to the central bank.

Trap 3 — "A rate change works at once"

Answers often say "rates rise, so spending falls, so inflation falls" as if it happens next month. The fullest effect on inflation takes about two years. Always mention the lag, and what it means for the central bank: it must act on forecasts.

Try it yourself

Work each one on paper first. Open a hint only when you are stuck.

Find the real interest rate, to one decimal place, in three invented cases: (a) nominal rate 4%, inflation 6% (b) nominal rate 4%, inflation 1.5% (c) nominal rate 10%, inflation 13%. In which case is monetary policy loosest?

Hint 1
Use $1 + real rate = \frac{1 + nominal rate}{1 + inflation rate}$ .

Hint 2
(a) is $1.04 \div 1.06$ . Subtract 1 and turn it into a percentage.

Full answer
(a) $1.04 \div 1.06 \approx 0.981$ , so about $- 1.9%$ . (b) $1.04 \div 1.015 \approx 1.025$ , so about $+ 2.5%$ . (c) $1.10 \div 1.13 \approx 0.973$ , so about $- 2.7%$ . Policy is loosest in (c), even though its nominal rate is the highest. A 10% rate sounds tight, but with 13% inflation, borrowers repay less buying power than they borrowed.
An invented household has a $150 000 mortgage. Its interest rate rises from 3% to 6%. Looking only at the interest, how much more does the household pay (a) in a year (b) in a month? Then explain why the effect on the whole economy is smaller in the first year than this suggests.

Hint 1
Interest for a year is the loan times the rate.

Hint 2
For the second part, think about fixed deals.

Full answer
(a) $150000 \times 0.03 = $4500$ before and $150000 \times 0.06 = $9000$ after, so $4 500 more a year. (b) $4500 \div 12 = $375$ more a month. In the whole economy, only households on variable rates, and those whose fixed deals end, pay more in the first year. If most mortgages are fixed for several years, most households feel nothing at first, so spending falls only gradually.
An invented central bank cuts its rate by 1 percentage point during a recession. Trace the effect through each of the four channels of the transmission mechanism to aggregate demand. Then say which component of AD each channel mainly affects.

Hint 1
The four channels are borrowing and saving, asset prices, the exchange rate, and expectations.

Hint 2
AD is $C + I + G + (X - M)$ . Monetary policy does not change G directly.

Full answer
Borrowing and saving: loans become cheaper and saving pays less, so households spend more (C) and firms invest more (I). Asset prices: house and share prices tend to rise, so owners feel richer and spend more (C). The exchange rate: lower rates make the currency less attractive, so it tends to fall, making exports cheaper and imports dearer ( $X - M$ rises). Expectations: firms and households expect support for demand, which can encourage investment and spending (I and C). The effect on inflation comes later, after output has responded. In a deep recession some channels may be weak, for example if firms are too worried to invest at any interest rate.
Two invented countries both have inflation of 8%. In Country A, inflation comes from strong spending: unemployment is very low and firms have no spare capacity. Most mortgages there are on variable rates. In Country B, inflation comes from a jump in the price of imported energy: unemployment is rising. Most mortgages there are fixed for five years. Both central banks are thinking of raising rates by 2 percentage points. Compare how quickly, and how appropriately, the rise would work in each.

Hint 1
Ask two questions for each country: is the cause demand-pull or cost-push, and how fast will the rise reach households?

Hint 2
Think about what a rate rise can and cannot change in Country B, and what it does to jobs there.

Full answer
Country A: the cause is demand-pull, which a rate rise is designed to treat. With variable-rate mortgages, households pay more almost at once, so spending slows quickly. The rise is appropriate and fast. Country B: the cause is cost-push. A rate rise cannot make imported energy cheaper. With five-year fixed deals, most households will not feel it for years. Meanwhile unemployment is already rising, and the rise would push it higher. The main case for a rise in B is to stop the energy shock spreading into higher pay and prices, if expectations start to drift up. So B's central bank faces a harder choice. It might raise rates less, or wait, if expectations stay anchored. The same inflation rate can need different policies, because the cause and the speed of transmission differ.

The bigger idea

Monetary policy is often presented as a technical tool, set by experts with a single target. It is quick to decide and quick to reverse, which makes it more flexible than fiscal policy. But it is not neutral. Every change in Bank Rate moves money between borrowers and savers, and between people who own houses and people who do not. That is one reason it was handed to an independent committee with a clear target, which must explain itself in public.

The fiscal policy lecture ended by saying that the two tools are strongest when they push in the same direction. In 2020 they did: low rates and QE kept borrowing cheap while the government supported incomes. When they pull against each other, each cancels part of the other. Both work on demand. The next lecture turns to a third tool, supply-side policy, which aims to raise the economy's capacity itself.

Key concepts in this lesson

Monetary policy uses interest rates, and sometimes the amount of money, to influence aggregate demand and inflation. In the UK the independent Monetary Policy Committee runs it.
Bank Rate is the rate the Bank of England pays on reserves, and other interest rates follow it because banks will not lend for less than they can earn there.
The transmission mechanism works through borrowing and saving, asset prices, the exchange rate and expectations, and affects output after about a year and inflation after about two.
Policy should be judged by the real interest rate, so a rising nominal rate can still be loose policy when inflation is high.
Quantitative easing creates reserves to buy bonds from investors, lowering long-term interest rates when Bank Rate is near zero.
Monetary policy cannot remove the cause of cost-push inflation, works with long lags, and affects borrowers, savers and asset owners very differently.