Aggregate Demand
Aggregate Demand
FREE RESOURCES
EXAM RESOURCES
PREMIUM
Aggregate demand guide: the components of AD and what shifts the AD curve
Aggregate demand is the total spending that buyers at home and abroad plan to make on one country's output.
Every year, households, firms and the government in Norla plan to spend money on goods and services, and buyers abroad plan to buy goods made in Norla. Norla is a middle-income country with an open economy, which means it trades a large share of its goods with other countries. Economists add up the planned spending on Norla's own output into one total.
Aggregate demand (AD): the total planned spending on goods and services produced in an economy, at a given price level, over a given period of time.
Put simply, AD adds up everything that buyers at home and abroad plan to buy from one country's producers in a year, into one total.
Norla's AD is $1,000bn: the spending planned on its output this year, at today's price level.
Exam Matters: In a Define question, give the meaning in one sentence: planned spending, on a country's output, at a given price level, over a period. A diagram or a list of the components answers a different question, so leave them out unless the question asks for them.
Aggregate demand adds up four components: spending by households, firms and the government, and the net trade balance.
Economists sort planned spending into four components: spending by households, by firms and by the government, and the net trade balance with the rest of the world.
Consumption (C): spending by households on goods and services.
Put simply, C is what families buy for their own use, from food and rent to haircuts.
Investment (I): spending by firms on capital goods, such as machinery, equipment and buildings.
Put simply, capital goods are goods used to produce other goods: a tailor's new sewing machine is investment, but the clothes it sells are not.
Government expenditure (G): spending by the government on goods and services.
Put simply, G is what the government buys, such as teachers' work, medicines and new roads.
Net trade balance (X − M): the value of exports minus the value of imports.
Put simply, exports (X) are foreign spending on Norla's output, and imports (M) are Norla's spending on other countries' output. X − M is the gap between the two.
Together they give AD = C + I + G + (X − M). Why are imports subtracted? C, I and G count all the spending of households, firms and the government, including spending on goods made abroad. A laptop that a household in Norla buys from abroad is counted in C. A machine a firm imports is counted in I, and medicine the government imports is counted in G. This spending buys another country's output, not Norla's. Subtracting M removes it, so what is left is spending on Norla's own output.
So $1,000bn is the total planned spending on Norla's output this year. The net trade balance is negative because Norla spends more on imports than foreign buyers spend on its exports, so it makes AD $40bn lower than C + I + G added together.
The components differ in size and in how much they change. C is the largest, at 64% of Norla's AD. It is also usually the steadiest, because households keep buying food and paying rent even in a bad year. I is much smaller than C, at 18% of AD, but it is usually the most volatile: it changes the most from one year to the next. A firm can delay buying a machine when its future sales look uncertain, so investment can fall sharply in one year and rise sharply in the next.
Exam Matters: When you calculate AD, work out the net trade balance first and write down its sign. A negative balance lowers AD, so keep its minus sign when you combine it with C + I + G.
The AD curve shows that real output demanded rises when the price level falls.
The AD curve plots the price level on the vertical axis against real output demanded on the horizontal axis. Real output is output measured at constant prices, so it rises only when more goods and services are bought, not when prices rise. At each price level, the curve shows aggregate demand: the real output that buyers plan to buy.
In Norla, the price level is measured as an index, with this year's level set at 100, and real output demanded is $1,000bn. If the price level falls to 94, real output demanded rises to $1,036bn. A lower price level goes with a higher real output demanded, so the curve slopes downwards.
Why does it slope downwards? In the market for one good, a lower price makes buyers switch to that good from its substitutes. The price level is different. When it falls, prices across the whole economy fall together. No good in Norla becomes cheaper than the others, so buyers have no reason to switch. The AD curve needs three reasons of its own:
Each effect starts with a change in the price level and ends with a change in a component of AD. A rise in the price level reverses each effect. The money households hold buys less, so consumption falls. Households and firms need more money for the same purchases, so they save less. Banks have less to lend, so interest rates rise, and borrowing to spend costs more, so consumption and investment fall. Norla's goods become more expensive than foreign goods, so X − M falls. In Norla, a rise in the price level from 100 to 106 lowers real output demanded from $1,000bn to $964bn.
Exam Matters: To explain one reason for the slope, start from the change in the price level, then name the component of AD that changes and say which way.
A change in the price level moves the economy along the AD curve; any other change shifts the whole curve.
The price level is on the vertical axis. So when the price level changes, the AD curve already shows the new real output demanded: the economy moves along the same curve to a new point. In Norla, a rise in the price level from 100 to 106 moves the economy up along AD, and real output demanded falls from $1,000bn to $964bn. A fall to 94 moves it down along AD, and real output demanded rises to $1,036bn.
Anything else that changes C, I, G or X − M is on neither axis. It changes planned spending at every price level, so the whole curve shifts. A shift to the right is an increase in AD, and a shift to the left is a decrease. So the test is one question: did the price level change, or did something else?
An increase: Norla's government raises G from $220bn to $260bn. At a price level of 100, real output demanded rises from $1,000bn to $1,040bn. The government chose the extra $40bn, and its choice does not depend on the price level, so spending is $40bn higher at every other price level too. So the curve shifts right, from AD₁ to AD₂.
A decrease: Norla's central bank raises its interest rate. The rates on households' $600bn of debt follow it, so the rate on that debt rises from 5% to 8%. The yearly interest on that debt rises from $30bn to $48bn. Households have $18bn less to spend on goods and services. If consumption falls by that amount, real output demanded at the same price level falls from $1,000bn to $982bn. The higher interest bill does not depend on the price level, so AD is lower at every other price level too, and the curve shifts left, from AD₁ to AD₃.
Both the central bank's decision and the interest-rate effect change interest rates. Why is only one of them a shift? The interest-rate effect starts with a change in the price level: at a lower price level, more money is saved and lent, so interest rates fall. Because it starts with the price level, it is a movement along AD. The central bank decides at an unchanged price level, so its decision shifts AD.
Exam Matters: When you draw a shift, label both curves, AD₁ and AD₂, and name the cause beside the arrow. At the original price level, read real output demanded across from both curves to the axis.
Households spend mostly from the income they have left after tax, so disposable income is the base of consumption.
A worker's payslip shows two figures. Gross pay is what the employer pays, and income tax is then taken off it. Income tax is a direct tax: a tax taken straight from a person's income. So the worker receives less than the gross pay. Some households also receive money from the government, such as a pension. What a household can actually use is its income after both changes.
Disposable income: income after direct taxes have been taken off and state benefits have been added.
Put simply, disposable income is the money that really reaches the household. A retired person with no job still has disposable income, because the pension counts.
A household can do only two things with its disposable income. It can spend it on goods and services, which is consumption, or it can keep it unspent. The part it keeps is called saving.
Saving: the part of disposable income that is not spent on consumption.
Put simply, saving is the part of this month's income that you do not spend, whether it goes into a bank account or stays at home.
In Norla, households have $800bn of disposable income this year. They spend $640bn on consumption and save the other $160bn. Most other influences on consumption work by changing how much of this income households spend, not the income itself.
Suppose Norla's government cuts the rate of income tax, and gross pay stays the same. Each worker pays less tax, so disposable income rises. Households now have more to spend, so consumption rises. Households usually save part of the extra as well, so consumption rises by less than the tax cut.
Consumption is part of AD, and the price level has not changed, so the whole AD curve shifts to the right. A rise in the rate of income tax reverses each step: disposable income falls, consumption falls, and AD shifts to the left.
Exam Matters: When a question names a change in tax or benefits, write the step through disposable income before you reach consumption. Without it, the reason why households spend more or less is missing.
A higher interest rate leaves households with less to spend, so consumption falls.
An interest rate is the price of borrowing and the reward for saving. So a change in interest rates reaches households through two channels, and both point the same way.
Why does a rise matter to households that borrowed years ago? Many loans have a variable rate: the rate on the loan changes whenever interest rates in the economy change. So a rise in interest rates makes old borrowers pay more too, not only new borrowers.
If households pay the extra $18bn by spending less, consumption falls from $640bn to $622bn. With I, G and X − M unchanged, AD falls by the same $18bn, from $1,000bn to $982bn. The price level has not changed, so the AD curve shifts to the left. A cut in interest rates reverses both channels: borrowers pay less interest, and each dollar saved earns less, so households spend more of their disposable income. Consumption rises, and at an unchanged price level the AD curve shifts to the right.
How fast consumption falls depends on the kind of loans. Where most loans have variable rates, borrowers pay more at once, so spending falls quickly. Where most loans have a rate fixed for many years, only new borrowers pay more at first, so the effect is slower and smaller.
When households expect their income to fall, they spend less now, even before their income changes.
Consumer confidence is how households feel about their future income and jobs. A household that expects a pay rise, or feels its job is safe, is confident. A household that fears losing its job is not.
Confidence can change consumption even when income has not changed. Think of a family in Norla whose pay, taxes and interest rate are the same as last year. News arrives that firms in the area are cutting jobs. The family's disposable income has not changed, but it worries about next year. So it saves more of its income now and spends less.
Which spending falls? Mostly spending that can be postponed: a new car, new furniture, a holiday. A family must still buy food and pay for electricity every month, but a car can wait a year. Each of these purchases costs a lot. So when many families delay them at the same time, even a small change in expectations can lower consumption by a large amount.
Confidence is not a component of AD. It changes AD through consumption. When confidence falls, C falls, and at an unchanged price level the AD curve shifts to the left. When confidence rises, households buy the things they had delayed, so C rises and AD shifts to the right.
Welfare payments raise disposable income, and low-income households spend most of any extra they receive.
Governments pay money to some households: pensions to older people, benefits to people out of work, and cash payments to families on low incomes. These are welfare payments.
Transfer payment: a payment from the government for which no good or service is given in return, such as a pension or unemployment benefit.
Put simply, a transfer moves money from the government to a household. When it is paid, nothing is made and nothing is bought.
Welfare payments are transfer payments. They are not part of government expenditure (G), because G counts only what the government buys, such as teachers' work or a new road. A transfer buys nothing. It reaches AD only when the household spends it, and then it is counted in consumption.
A welfare payment is a state benefit, so it raises disposable income. How much of it becomes consumption depends on who receives it. A family on a low income often has needs it cannot yet pay for, and little saving. So it is likely to spend most of an extra dollar. A household on a high income already buys what it needs, so it is likely to save much of an extra dollar.
So $1bn more in welfare payments to low-income families is likely to raise consumption by more than a $1bn tax cut for high earners. A cut in welfare payments reverses this: those families lose disposable income and buy less, so consumption falls by most of the cut. At an unchanged price level, a rise in welfare payments shifts the AD curve to the right, and a cut shifts it to the left.
The effect also depends on how the payment is paid for. If the government raises another tax to pay for it, the households who pay that tax have less disposable income, and their consumption falls. This cancels part of the rise in consumption that the payment brings.
When the things households own rise in value, households tend to spend more, even with no change in income.
Income is what a household earns each year. Wealth is what it owns at one moment: a home, land, bank savings, or shares, which are small parts of the ownership of companies. The value of this wealth changes when house and share prices change, even if nobody's income changes.
Wealth effect: a change in consumption caused by a change in the value of the assets households own.
Put simply, when your family's home is worth more, you feel richer and spend more, even though your pay is the same.
Suppose house prices in Norla rise, while incomes stay the same. Consumption then rises through two routes. At an unchanged price level, the AD curve shifts to the right.
A fall in asset prices reverses both routes. Owners feel poorer, so they save more of their income. Banks lend less against homes that are worth less, so owners can borrow less. Consumption falls, and at an unchanged price level AD shifts to the left.
Do not confuse this with the wealth effect behind the slope of the AD curve. That effect starts with a change in the price level, so it moves the economy along the AD curve. The wealth effect in this part starts with a change in house or share prices. The price level stays the same, so the whole curve shifts.
Whether banks will lend at all can change consumption, even when the interest rate stays the same.
Many households borrow to buy a car, furniture or a home. Two things decide whether they can. The first is the price of the loan: the interest rate. The second is the availability of credit: whether a bank will lend to them at all. A bank decides this by setting conditions, the rules a borrower must meet to get a loan.
Common conditions are:
None of these conditions is a price. For example, a bank can ask for a deposit twice as large as before, while its interest rate stays the same. Some households who could borrow before now cannot, so they buy fewer cars and less furniture, and consumption falls. At an unchanged price level, the AD curve shifts to the left.
This is why availability and price can move apart. Interest rates can be low while banks still refuse many loans, for example when banks fear that borrowers will not repay. Then consumption can fall even though borrowing is cheap.
Households borrow mostly for large purchases that can be postponed, such as cars and furniture, so these are the purchases that fall most when banks lend less. When banks ease their conditions, more households can borrow, consumption rises, and at an unchanged price level AD shifts to the right.
The savings ratio shows how much of each dollar of disposable income households save rather than spend.
Disposable income has only two uses: households spend it on consumption, or they save it. So consumption plus saving always equals disposable income. In Norla, $640bn + $160bn = $800bn.
This has an important result. When disposable income does not change, every influence on consumption is also an influence on saving, pointing the other way. A change in disposable income itself is different: when it rises, households usually spend more and save more. If falling confidence leads households to spend $10bn less of an unchanged disposable income, they save $10bn more. If rising house prices lead them to spend more, they save less. These are not two separate effects. They are one decision about how to split disposable income.
Savings ratio: saving as a percentage of disposable income.
Put simply, the savings ratio tells you how many dollars out of every $100 of disposable income households save. They spend the rest of each $100.
A question may give the savings ratio instead of saving. Then start from the ratio: 20% of $800bn is $160bn of saving. Consumption is the part not saved, $640bn, and AD follows in the same way.
So with the same disposable income, a higher savings ratio means lower consumption, and a lower ratio means higher consumption.
Exam Matters: In a Calculate question, write the formula first, then put the figures into it, one stage per line, and give the units: a percentage for the ratio and $bn for consumption.
A rise in the savings ratio lowers consumption and AD this year.
The savings ratio changes when households decide to save a different share of their disposable income. Its causes are the influences that change how households split an unchanged disposable income between spending and saving.
When any of these moves the opposite way, households save a smaller share, and the ratio falls.
The price level is unchanged in both cases, so the AD curve shifts: left when the ratio rises, right when it falls.
A rise in the ratio can also have a second effect, which takes longer. Households keep much of their saving in banks. With more saving, banks have more funds to lend, and to find borrowers for them they may lower the interest rates on their loans. Cheaper loans reduce the interest a firm pays to buy a machine or a building, so more of these purchases become worth making, and investment may rise in the following years.
This later rise is not certain. If firms expect weak sales, partly because households are spending less, they may not borrow even at a lower rate. So a higher savings ratio lowers C this year, and it may or may not raise I later. A fall in the ratio reverses this second effect: banks have fewer funds to lend, so interest rates on loans may rise, and investment may fall in the following years. This is not certain either.
Exam Matters: In an Examine question on the savings ratio, follow the chain through consumption first. Then weigh the later effect through bank lending and investment, and say what it depends on.
Firms spend on capital goods, such as machines and buildings, both to replace worn-out ones and to add new ones.
Every year, some of the machines, vehicles and buildings that firms own wear out, and some become out of date. All the capital goods that firms own make up the capital stock. Part of what firms spend on capital goods replaces worn-out ones, and the rest adds to the capital stock. Two measures keep these apart.
Gross investment: the total spending on capital goods in a period.
Put simply, gross investment counts every capital good that firms buy this year, whether it replaces an old one or adds a new one.
Depreciation: the fall in the value of the capital stock over a period as machines and buildings wear out or become out of date.
Put simply, depreciation is the part of the capital stock used up during the year. A delivery van that can no longer run must be replaced before the firm owns any more vans than it did.
Net investment: gross investment minus depreciation.
Put simply, net investment is the change in the capital stock over the year: positive when it grows, negative when it shrinks.
Why keep both figures? They answer different questions. Aggregate demand counts gross investment, all $180bn, because every capital good bought is spending on output. A van that replaces an old one is made and sold, just like an extra van. Net investment changes the capital stock, and so it changes capacity, the most the economy can produce. Only the $60bn of new capital lets Norla produce more next year. In the weak year, all $110bn still counts in AD, but Norla starts the next year with less capital than before.
Exam Matters: In an Explain question on gross and net investment, define both measures, then use the data: subtract depreciation, and say what the sign of the result means for the capital stock.
Firms buy new capital goods when they expect sales to grow beyond what their machines can produce.
A machine runs for many years, so a firm buys one for the sales it expects in the future, not for this month's sales. When Norla's economy grows, households and firms buy more, and many firms expect their sales to keep rising.
Whether a firm needs new machines for those extra sales depends on its spare capacity. A firm has spare capacity when its machines and buildings could produce more than they produce now. A firm with spare capacity can meet more orders with the machines it already has, so it does not need to invest. A firm running near the limit of its machines cannot meet more orders that way. If it expects sales to keep growing, it must buy new capital goods.
So investment depends on the rate of economic growth, which is how fast output is rising, not only on the level of output. Suppose output in Norla has grown quickly for some years, and firms have bought machines to keep up with their rising sales. Now growth slows. Output is still rising, but more slowly than the pace those machines were bought for. Many firms now have spare capacity, so they order fewer new machines. This is why investment can fall even in a year when output is still rising. If growth speeds up instead, sales rise faster than before. More firms reach the limit of their machines sooner, so investment rises. Investment is a component of AD, so at an unchanged price level a fall in investment shifts AD left, and a rise shifts it right.
Exam Matters: In an Analyse question on growth and investment, say whether firms have spare capacity before you say they will invest: the chain runs from expected sales, to capacity, to new capital goods.
A firm buys a capital good only when its expected rate of return is higher than the interest rate.
Kovo, a furniture maker in Norla, is thinking about buying a new machine that costs $1 million. Kovo expects it to add $80,000 a year to its profit. That yearly amount, as a percentage of the cost, is the machine's expected rate of return: $80,000 is 8% of $1 million. It is "expected" because nobody knows future sales for certain.
Suppose Kovo borrows the $1 million from its bank. It must then pay interest on the loan every year. So the machine is worth buying only if it adds more to profit each year than the interest costs. This is the same as asking whether the expected rate of return is higher than the interest rate. This comparison is the return test: a project passes it when its expected rate of return is higher than the interest rate.
What if Kovo pays with its own money? The test still holds. Kovo could put its $1 million in a bank deposit instead and earn interest. At 10%, that is $100,000 a year, more than the machine adds. The interest Kovo gives up is the opportunity cost of buying the machine: the benefit of the next best choice, which Kovo gives up. So the same comparison decides.
Firms in Norla make this comparison for each project they could invest in. When the interest rate rises, projects whose expected rate of return lies between the old and the new rate no longer pass, so investment falls. Kovo's machine is one of them: 8% passed at 7% but fails at 10%. Investment is a component of AD, and the price level has not changed, so AD shifts left. A cut in the interest rate reverses each step: if the rate falls from 10% back to 7%, Kovo's machine passes again, and across Norla more projects pass, investment rises and AD shifts right.
Exam Matters: In an Explain question on interest rates and investment, use the return test: work out the expected rate of return, then set it against the interest rate before and after the change.
The return on a capital good is a forecast, so when firms feel less confident, the return they expect falls.
Kovo expects its $1 million machine to add $80,000 a year. This is a forecast, not a fact: it depends on how much furniture Kovo expects to sell over the years the machine will run. With an expected return of 8% and an interest rate of 7%, the machine passes the return test. Business confidence is how optimistic firms feel about their future sales and profits. When confidence is high, firms expect strong sales. When it falls, they expect weaker ones.
So confidence works inside the return test. It changes the expected return, which the firm compares with the interest rate. Suppose Kovo's managers now fear a slowdown and expect the machine to add only $40,000 a year. That is 4% of $1 million. At an interest rate of 7%, a year's interest is $70,000, more than the machine is now expected to add, so Kovo does not buy it.
The interest rate, the machine and Kovo's bank are all unchanged. Only the forecast changed. If many firms lose confidence at the same time, many projects fail the test, so investment across Norla falls. At an unchanged price level, AD shifts left. A rise in confidence reverses each step. If Kovo's managers again expect $80,000 a year, the expected return is back at 8%, above 7%, so Kovo buys the machine. Across Norla, more projects pass the test, investment rises and AD shifts right. Forecasts can change quickly, which is one reason investment is usually the most volatile component of AD.
A project can pass the return test and still not go ahead, because the firm cannot borrow enough.
The availability of credit is whether a bank will lend at all, and on what conditions; its price is the interest rate, and a bank may refuse a loan at any interest rate. Availability can matter a great deal to firms, for three reasons:
Take Kovo's $1 million machine, which is expected to add $80,000 a year: a return of 8% against an interest rate of 7%, so it passes the return test. Now Kovo's bank will lend only half the cost, $500,000. If Kovo cannot find the other $500,000 itself, it does not buy the machine, even though the project still passes.
When banks across Norla lend less to firms, investment falls at an unchanged interest rate, and AD shifts left at an unchanged price level. If banks lend more, more of the projects that pass the test can be paid for, so investment rises and AD shifts right. This also explains why a cut in the interest rate may not raise investment. A cheaper loan does not help a firm that cannot get a loan at all.
A firm compares the interest rate with the return it keeps after tax, not with the return before tax.
A company pays part of its profit to the government as tax. A tax on company profits and corporation tax are two names for this one tax.
Corporation tax: a tax on the profits of companies.
Put simply, corporation tax is the government's share of each company's profit. The higher the rate, the less of each extra dollar of profit the company keeps.
This changes the return test. Kovo can only use the profit left after tax, whether to repay a loan or to pay its owners. So the return Kovo compares with the interest rate is its kept return: the return it keeps after tax, not the return before tax.
Before tax, the machine passed the test, because 8% is above 7%. After tax it fails, because 6% is below 7%, even though nothing else changed. A higher rate of corporation tax lowers the kept return on every project of every company that pays the tax, so fewer projects pass and investment falls. A lower rate raises the kept return on every project of a company that pays the tax, so more projects pass.
Tax relief and subsidies lower what a firm pays for a capital good, so the same yearly return is a higher percentage of the cost.
To raise investment, a government can make capital goods cheaper for firms in two ways. Tax relief lets a firm subtract what it spends on a capital good from its taxable profit, the profit that corporation tax is charged on. A subsidy is a payment from the government that covers part of the price of a capital good.
Start from Kovo's position after corporation tax of 25%. Kovo keeps $60,000 a year from its $1 million machine, a kept return of 6%, which is below the 7% interest rate. So without help, Kovo does not buy the machine.
Neither policy changes the $60,000 that Kovo keeps each year. Both lower what Kovo pays for the machine, so the same $60,000 is a larger percentage of the cost. Across Norla, projects whose kept return was just below the interest rate now pass, so investment rises and AD shifts right. If the government ends the subsidy or the tax relief, each step reverses: the machine costs Kovo $1 million again, $60,000 is 6%, below 7%, and Kovo does not buy it. Across Norla, investment falls and AD shifts left.
Both policies cost the government money: tax relief lowers its tax revenue, and a subsidy is spending. Some firms would have invested anyway, so part of the money pays for investment that would have happened without it.
Exam Matters: In a Draw question on investment, decide first whether the change acts on a component of AD at an unchanged price level. If it does, shift the whole AD curve instead of moving along it.
A cut in corporation tax raises the return that a company paying the tax keeps on its projects, but it changes only some decisions.
Unlike tax relief and a subsidy, a cut in corporation tax does not change what a capital good costs. It raises what every company that pays the tax keeps from every project, each year.
So a cut changes a decision only where the kept return was below the interest rate, but close enough for the cut to lift it above. Take another project that adds 9% a year before tax. At a tax of 25% it keeps 6.75%, below 7%. At 20% it keeps 7.2%, above 7%, so it now goes ahead.
The cut also reaches firms whose decisions do not change. A company that was already investing keeps more profit, but it would have invested anyway, so the government gives up revenue for no extra investment. A company that makes a loss has no profit to tax, so it pays no corporation tax. A lower rate therefore gives it nothing.
This is the key difference from a subsidy, a payment towards the price of a capital good, which goes only to the goods the government chooses to pay for. A tax cut goes to every company that pays the tax. So the extra investment it brings depends on how many projects were close to the interest rate, and how many firms make a taxable profit. Where projects do go ahead, investment rises and AD shifts right at an unchanged price level.
Real Example: In September 2019, India cut its basic rate of corporation tax from 30% to 22% for companies that gave up most other tax breaks. New manufacturing companies could pay 15%, if they started production by March 2023. The general cut was open to existing companies, including those already investing. The lowest rate was only for new manufacturing companies, so only new investment could gain from it.
Exam Matters: In an Examine question on a tax cut, end with a judgement that says what the result depends on, such as how many firms make a loss or how many projects were close to the interest rate.
Fiscal policy is a government's choice of how much to spend and how much to tax, and its spending sets G directly.
Each year, Norla's government decides in its budget how much it will spend and how much tax it will collect. This decision is one influence on government expenditure (G).
Fiscal policy: the use of government spending and taxation to influence the economy.
Put simply, fiscal policy is the government using its own budget to change what happens in the economy, for example spending more on roads when it wants more output and jobs.
G is the one component of AD that the government sets directly. Households decide C, firms decide I, and buyers abroad decide exports. But when the government decides to buy more roads, medicines or teachers' work, G rises by the amount it decides.
Suppose Norla's government raises G from $220bn to $260bn, and nothing else changes. AD rises by the same $40bn, from $1,000bn to $1,040bn. The price level has not changed, so AD shifts to the right: at every price level, real output demanded is $40bn higher.
A cut works the same way in reverse: if the government cut G by $40bn instead, AD would fall by $40bn, to $960bn, and the curve would shift to the left.
Taxation, the other half of fiscal policy, reaches AD by a different route. A tax cut buys nothing, so it does not change G. Instead it raises households' disposable income, their income after taxes and benefits. Households spend part of the extra income and save the rest. Only the part they spend adds to C.
Why does this matter? Because of this difference, the same $40bn adds more to AD as a rise in G, spent on goods and services made in Norla, than as a tax cut.
Exam Matters: In a Draw question on AD, label the axes "Price level" and "Real output", draw the new curve beside the old one, and add an arrow to show which way AD moved.
When output and jobs rise or fall, the government needs to buy different things and can afford different amounts.
The level of economic activity is how much an economy produces and how many people are in work. It changes two things for the government: what it needs to buy, and what it can afford.
Suppose Norla enters a downturn, a period when output falls and many workers lose their jobs. A downturn like this is also called a recession. More people now need training courses and employment offices to help them find work. If the government buys $35bn more of these services, G rises from $220bn to $255bn. This adds $35bn to AD at every price level, which cancels part of the fall in C and I.
Unemployment benefit also rises. But a benefit is a transfer payment, money paid with nothing bought in return, so it is not part of G. It reaches AD as C only when households spend it.
Tax revenue falls too, because people and firms earn less. So to pay for the extra services, the government may have to borrow, raise tax rates or spend less on something else.
Now take the reverse case: a boom. Output grows fast and unemployment falls. Fewer people need training places, so the government buys $10bn less of them, and G falls from $220bn to $210bn. This lowers AD by $10bn at every price level, which holds back part of the rise in C and I. But higher incomes and profits raise tax revenue, which makes other spending, such as new schools, easier to afford.
So the level of economic activity moves G through two changes at once. In a downturn, what the government needs to buy rises while what it can afford falls. In a boom, what it needs to buy falls while what it can afford rises. The effect on G depends on which of the two changes is larger.
Governments buy goods and services that markets provide too little of, and these purchases are part of G.
Some goods are provided too little, or not at all, when only private buyers and sellers decide. This is a market failure: the market does not produce the quantity best for society. Correcting market failures is another influence on G.
There are two common reasons. The first is public goods, such as defence and flood protection. A sea wall is one kind of flood protection. Once it is built, it protects every house behind it, and no one can be stopped from benefiting. So people can benefit without paying, and many will choose not to pay. A seller cannot charge enough of the people who benefit to cover its costs, so the market provides too little or none. The second is external benefits. Vaccination and basic research help people who did not pay for them. Buyers count only their own benefit, so they buy less than is best for society.
So the government can buy these goods and services itself. It pays for an army, builds sea walls, buys vaccines and pays scientists. Each is a purchase of goods or services, so it is part of G. A government can also correct a market failure without buying anything, for example by passing a law or by taxing a good. These actions do not change G, because G counts only what the government buys.
The reason for a purchase does not change how it is counted. A vaccine bought to correct a market failure adds exactly as much to G as one bought for any other reason.
The reason matters in a different way: it makes this spending hard to cut. If the government stops paying for flood protection, a private firm is unlikely to provide it, because it cannot charge everyone who benefits, and the failure returns. So when a government must spend less, it may find it easier to cut other parts of G first.
What a government chooses to spend on depends on its political priorities, and this can change both what G is spent on and its total.
Governments have different aims. One promises better schools, another a stronger army, another lower taxes. These political priorities also influence G. Two governments facing the same economy and the same tax revenue can still spend in different ways.
Priorities change the composition of G: how its total is divided between different uses. Suppose a new government in Norla spends $15bn more on teachers and $15bn less on building roads, both bought from Norla's own workers and firms. G stays at $220bn and nothing else changes, so AD this year is unchanged.
The change still matters for the future. More teachers can raise the skills of the students who will be workers in later years. Fewer new roads can make it slower and more costly to move goods. So a change in composition can leave AD unchanged this year and still change what the economy can produce in the future.
Priorities can also change the total of G. A government that wants a smaller state may spend less and tax less, so G falls and AD shifts left. One that wants more public services may spend more and tax more, so G rises and AD shifts right. Priorities also shape how a government uses fiscal policy, how much extra it spends in a downturn, and which market failures it pays to correct.
Real Example: Indonesia's president, Prabowo Subianto, promised free meals for schoolchildren in his election campaign. His government began the programme in January 2025, with a budget of 71 trillion rupiah (about $4.3 billion) for 2025. The government pays for meals that kitchens run by catering services prepare, so this spending is part of G. It shows a political priority deciding what G is spent on.
When real income at home rises, households buy more imports, but exports do not change.
When people in Norla earn more, they spend more, both on goods made in Norla and on goods made abroad. What matters is real income: income after allowing for inflation, so it measures what people can actually buy.
Suppose Norla's real income rises by $100bn. Households spend $80bn more in total, and $25bn of this extra spending is on imports.
Why do exports not change? Exports are bought by households and firms in other countries, and their spending depends on their own incomes. Those incomes have not changed. So a rise in real income at home moves M, not X.
Does AD fall, then? No. The $25bn spent on imports buys another country's output, so M removes only that part. The other $80bn − $25bn = $55bn is spent on Norla's own output, so AD rises by $55bn.
A fall in real income reverses each step. If households spend $80bn less and imports fall by $25bn, to $275bn, X − M rises from −$40bn to −$15bn. Exports still do not change, and AD falls by $55bn, because C falls by $80bn while M falls by only $25bn.
Exam Matters: In a Calculate question on the net trade balance, work out X − M before and after, each with its sign. Then say whether the balance rose or fell, and by how much.
When a country's currency loses value, its exports become cheaper abroad and its imports more expensive at home.
Norla's exporters are paid in Norla's currency, but many foreign buyers pay in US dollars. So each sale abroad changes one currency into another.
Exchange rate: the price of one currency in terms of another currency.
Put simply, it is how much of one currency one unit of another buys. Norla's rate is quoted in units of its currency per US dollar.
Depreciation (of a currency): a fall in the value of a currency against another currency.
Put simply, the currency buys less foreign money, so foreign goods cost more units of it. This is not the depreciation of capital, the fall in the value of machines and buildings as they wear out.
Suppose Norla's rate moves from 4 to 5 units per dollar. Each unit is now worth less, because a dollar costs more of them: a depreciation.
Foreign buyers now pay fewer dollars for each of Norla's goods, so they are likely to buy a larger quantity of them: X tends to rise. Buyers in Norla now pay more units of their currency for each import, so the quantity of imports they demand is likely to fall: M tends to fall. Both changes tend to raise X − M. How much depends on how strongly buyers respond to the new prices.
A rise in a currency's value is an appreciation, and it reverses each step. Suppose the rate moves from 4 to 3 units per dollar instead. Each unit is now worth more, because a dollar costs fewer of them. An export priced at 1,200 units now costs a foreign buyer $400 instead of $300, and imports cost buyers in Norla fewer units of their currency. X tends to fall, M tends to rise, and X − M tends to fall.
Real Example: On 6 March 2024, Egypt's central bank let the market set the value of the Egyptian pound. That day the pound fell from about 30.9 to about 50.6 pounds per US dollar. This was a large depreciation: goods priced in pounds became much cheaper for buyers paying in dollars, and imports priced in dollars cost Egyptians far more pounds.
Norla's exports depend on the incomes of the countries that buy them, not on incomes in Norla.
Norla's exports are spending by households and firms in other countries, so incomes abroad affect them. The state of the global economy means whether the economies of a country's trading partners are growing or shrinking. When Norla's trading partners grow, their real incomes rise and they buy more, including more goods from Norla. When they enter a recession, a period when their output and incomes fall, they buy less.
Suppose Norla's two largest buyers enter a recession, and Norla's exports fall from $260bn to $240bn. Nothing has changed for households in Norla, so imports stay at $300bn.
This is the mirror of a change in real income at home. A change in real income at home changes M, because it changes what Norla's own households spend, but X stays the same. A change in real income abroad changes X, because it changes what foreign buyers spend, but M stays the same. The lost exports were spending on Norla's own output, so AD falls by the same $20bn.
Growth abroad reverses each step. Suppose that, instead of a recession, the real incomes of the two largest buyers rise, and Norla's exports rise from $260bn to $280bn. Imports stay at $300bn, so X − M rises from −$40bn to −$20bn, and AD rises by $20bn.
How much a recession abroad matters depends on where a country's exports go. A country that sells mostly to one or two partners loses many sales when they slow. One that sells to many is hit less, because some may still be growing.
Real Example: In 2009, during a worldwide recession, the volume of world trade fell by about 12%, its sharpest fall since the end of the Second World War. The main cause was a large fall in spending around the world. It shows the state of the global economy at work: many countries' exports fell mainly because their buyers abroad were spending less.
Protection at home cuts a country's imports, but protection abroad can cut its exports too.
Many governments protect their own producers from foreign competition. The degree of protectionism is how much of this a country does, and how much its trading partners do.
Protectionism: the use of measures such as tariffs and quotas to restrict imports in order to protect domestic producers.
Put simply, protectionism makes foreign goods harder or more expensive to buy, so that buyers buy goods made at home instead. A tariff is a tax on imports. A quota is a limit on the quantity of a good that may be imported.
Suppose Norla's government puts a tariff of 25% on an import priced at $100. Buyers in Norla now pay $100 + $25 = $125. At the higher price, some buyers switch to goods made in Norla, so M falls. A quota cuts M directly: once the limit is reached, no more of the good may be imported. Either way, M falls, and because M is subtracted, X − M rises at first.
Why only at first? Norla's trading partners may respond by protecting their own producers. If they put tariffs on Norla's goods, foreign buyers pay more for Norla's exports, so the quantity of them demanded is likely to fall, and X falls. Whether X − M rises or falls in the end depends on which falls more: it rises if M falls by more than X, and it falls if X falls by more than M.
Less protection reverses each step. If Norla removes its tariff, the import costs $100 again instead of $125, so buyers switch back to it: M tends to rise, and X − M tends to fall. If its partners remove their tariffs, Norla's exports cost foreign buyers less, so the quantity of them demanded is likely to rise: X tends to rise, and X − M tends to rise.
Real Example: In 2018 the United States put a 25% tariff on steel imports, including steel from Canada, Mexico and the European Union. Canada, Mexico and the European Union responded with tariffs on US goods. Canada's covered about $12.8 billion of US exports, including steel and food products. So the US made its steel imports more expensive, but some of its own exports then faced tariffs abroad.
Exam Matters: In an Examine question on protection, follow the import side and the export side separately. Then judge which change is likely to be larger in the case you are given.
Buyers choose between exporters by more than price, so quality and service can change exports when prices stay the same.
Imagine two firms that sell the same machine abroad at the same price. One delivers in two weeks and sends an engineer when a machine breaks. The other delivers in two months and sends nobody. Most buyers are likely to choose the first, although the prices and the exchange rate are the same.
These are non-price factors: everything except price that makes buyers choose one good over another. The main ones are quality, design, reliability, speed of delivery and after-sales service.
Non-price factors change X and M at unchanged prices. If Norla's firms make their goods more reliable, foreign buyers order more of them, and buyers in Norla choose them over imports, so X rises and M falls. If foreign firms improve their designs faster than Norla's firms do, buyers abroad and in Norla choose the goods of the foreign firms instead: Norla's X falls and its M rises. In both cases, prices and the exchange rate have not changed.
Non-price factors usually change slowly. A currency can lose value in a day, but a reputation for quality or reliability usually takes years to build. So non-price factors are more likely to explain lasting differences in countries' exports than changes from one year to the next.
So the two work in different ways. The exchange rate changes the price that buyers pay. Non-price factors change what buyers receive for that price, such as quality and service.
Exam Matters: When a question gives a change in exports with no change in prices or the exchange rate, look for a non-price factor such as quality or delivery, and name it.
Aggregate demand is the total spending that buyers at home and abroad plan to make on one country's output.
Aggregate demand adds up four components: spending by households, firms and the government, and the net trade balance.